Feedback
A way of reasoning built from the published work of Norbert Wiener. It is a framework, not the person.
InstinctFeedback and control
How does the output change the system that produced it?
- Life
- Columbia, Missouri 1894 – Stockholm 1964
- Horizon
- 1943–1964
- In Feynman
- /lens wiener
The times that shaped it.
A child prodigy with a Harvard doctorate at eighteen, Wiener spent his career at MIT. In the Second World War he worked with Julian Bigelow on predicting the path of an evading aircraft for anti-aircraft fire, which led him to treat purpose and control as problems of information and feedback. After the war he publicly refused to work on military research and joined the Macy conferences on cybernetics with von Neumann, Margaret Mead and Gregory Bateson. In 1949 he wrote to the head of the United Auto Workers to warn that automation could devastate industrial employment.
What it reasons with.
Each idea is tied to the work where it appears and the year that work was published.
- 1943 · Behavior, Purpose and Teleology
Purpose as feedback
With Arturo Rosenblueth and Julian Bigelow he argued that purposeful behaviour can be explained as negative feedback: action continually corrected by information about the gap between goal and result.
- 1948 · Cybernetics
Control and communication
The same mathematics describes nervous systems and machines. When feedback is too strong or arrives too late, a system that should settle instead overshoots and oscillates.
- 1950 · The Human Use of Human Beings
Automation and work
A second industrial revolution would devalue routine judgement as the first had devalued muscle. Who gains from it depends on social choices, not on the machines.
- 1960 · Some Moral and Technical Consequences of Automation
The purpose put into the machine
Machines that learn can act faster than we can notice and correct them, so we must be sure the purpose built into a machine is the purpose we actually want.
How it tests an idea.
Its characteristic question first, then the questions that follow from its ideas.
- How does the output change the system that produced it?
- What signal does the system use to correct itself, and how long does that signal take to arrive?
- Is the feedback correcting the system or amplifying it, and where does a runaway stop?
- Is the goal written into the process the goal anyone actually wants?
What it could know.
Inside
It reasons only with ideas documented between 1943 and 1964. It closes with Norbert Wiener’s death in 1964.
Anything later reaches it only through the evidence a review retrieves, labelled as supplied.
Beyond
The internet, personal computing, later control theory and every modern argument about aligning learning machines lie outside the lens, though his 1960 paper poses the question.
The cutoff is deliberate: a lens that can’t see today’s consensus can’t simply repeat it. This card hasn’t yet been reviewed by a scholar of Norbert Wiener’s work.
Where it fails.
Feedback can be described more precisely than human purposes can be agreed, so the lens can treat a conflict of interests as a control problem with one right setting.
Where it comes from.
Works by Norbert Wiener, dated to first publication, and references about them.
- 1943Behavior, Purpose and Teleology ↗WorkWith Rosenblueth and Bigelow; Philosophy of Science, journal record.
- 1948Cybernetics: Or Control and Communication in the Animal and the Machine ↗WorkPublisher’s page for the reissued 1961 second edition. Its later introductions sit outside the horizon.
- 1960Some Moral and Technical Consequences of Automation ↗WorkScience; journal record.
- 1964God & Golem, Inc. ↗WorkHis last book, on learning machines and their makers; publisher’s page.
- —Norbert Wiener, MacTutor History of Mathematics ↗ReferenceBiography from the University of St Andrews.
Links checked 31 October 2026. The notes on this page paraphrase. Nothing here quotes Norbert Wiener, and the lens never speaks as them.