Communication and Control Are Intertwined

Control is exercised through communication, and a controller can control no more than its channel can carry. Regulation, read in cybernetic terms, is a communication problem (Ashby — An Introduction to Cybernetics).

Promoted 2026-11-09 — partially confirmed by Ashby — An Introduction to Cybernetics (originally captured as the human's hunch 2026-06-01; evaluated and left unpromoted 2026-06-02). Ashby grounds the control→communication half: cybernetics is "the study of systems that are open to energy but closed to information and control," regulation is interpreted "in terms of communication," and "R's capacity as a regulator cannot exceed R's capacity as a channel of communication." Not grounded: the converse ("to communicate is to influence"), and the conversational half (pauses and discourse markers as control signals). See Open questions.

The idea

Communication and control are not separate phenomena but two faces of the same thing: to communicate is to influence, and to control is to send signals that steer a system's behavior. The act of conveying information already carries the machinery of regulation — feedback, correction, and steering toward a goal.

This is the founding intuition of cybernetics — Wiener literally subtitled his work control and communication in the animal and the machine — but the hunch here also reaches into human conversation: the small devices people use to talk (pauses, "uh"/"um", discourse markers, backchannels) may be control signals that regulate the joint activity of a dialogue, not mere noise. If so, the same coupling shows up at both the engineering scale (feedback systems) and the interpersonal scale (managing a conversation in real time).

What would ground this

• A primary source on cybernetics tying communication and control together as one subject (Wiener's Cybernetics or The Human Use of Human Beings).
• Psycholinguistic work showing conversational devices function as control/coordination signals rather than disfluencies (Fox Tree's research).
• Either would confirm; a source arguing the two are usefully distinct (e.g. information theory treating transmission as separable from steering) would challenge and is worth flagging.

Grounding (2026-11-09)

Ashby builds cybernetics on Wiener's definition, "the science of control and communication," and then shows why the two words belong in one phrase. The Law of Requisite Variety ("only variety in R can force down the variety due to D; variety can destroy variety") is proved on a table of moves, owing "nothing to experiment." When he returns to regulation, the diagram's arrows "represent actual channels of communication," and the law becomes a bound on control stated in communication terms, in exact correspondence with Shannon's Theorem 10: disturbance is noise, the regulator is the correction channel. So the first half of the hunch holds as a theorem: steering is bounded by signalling. What the hunch adds beyond Ashby, that all communication is already steering, is a stronger claim he does not make.

In this wiki. The same bound explains Verification Requires Domain Knowledge (a reviewer regulates only the variety they can themselves produce) and gives Method Reflection T3 a measure: the human controls the agent through a channel, and no finer than that channel allows.

Open questions

• The converse: is every act of communication already an act of control, as the hunch says, or only those that close a loop? Ashby is silent; Wiener's own text (Norbert Wiener, #to-read) may not be.
• The conversational half: do pauses, "um," discourse markers and backchannels act as regulators of a dialogue in Ashby's sense? Still needs Jean E. Fox Tree.
• Ashby's law is about variety, not meaning. Does the coupling survive once content matters, or is it a statement about capacity only?

Grounding history

2026-06-02 — evaluated, NOT promoted. A Wiener source entered the wiki (The Human Use of Human Beings — Wiener (excerpt)), but it addresses information and originality (Clichés Carry Less Information) — not the communication⇄control coupling this seed claims. Right author and book, wrong passage (and only secondary commentary at that). The seed stays unverified: the control/feedback half still needs the relevant passages of Wiener's primary works, or Jean E. Fox Tree's research on conversational devices.

Leads (still open)

• Norbert Wiener — cybernetics' founder; the definition Ashby adopts. His primary works remain unread; they are where the converse half would be tested.
• Jean E. Fox Tree #to-read — psycholinguist on conversational devices (filled pauses, discourse markers); the grounding for communication-as-control at the human/dialogue scale.

Connections

• Ashby — An Introduction to Cybernetics — the grounding source: regulation bounded by channel capacity.
• Verification Requires Domain Knowledge — the same bound applied to review: variety can only be destroyed by variety.
• Human-in-the-Loop by Exception — the regulator acts only on the variety the table does not already block (Ashby §11/9).
• Agent Autonomy Spectrum — see also (weak resonance, not grounding): agent "steering," voice input, and OpenClaw "heartbeats" are control exercised through communication / feedback loops — a possible modern instance of this coupling. But The AI Daily Brief — Why Agents Still Need Humans does not cite Wiener or Fox Tree, so it does not ground this hunch. Flagged for the human to judge whether it's more than coincidence.
• The Human Sandwich — see also: a human steers an agent by framing and judging — communication as the control surface.

Sources

• Ashby — An Introduction to Cybernetics — chapters 1 and 11.

Future cybernetics, feedback, and conversation-analysis pages should also link back here.