--- title: "Ch 25 — Systems as an act of recognition: observer, variety, control" description: "A system is not found but recognized: coherence, pattern and purpose are all acts of mind, and because variety is counted by an observer, control begins with a…" url: "https://wiki.cybernetic-futures.org/pure-vsm/part-05-method-and-epistemology/ch25-systems-as-an-act-of-recognition" modified: "2026-11-12T15:26:26.128Z" language: "en" site: "Cybernetic Futures" --- # Ch 25 — Systems as an act of recognition: observer, variety, control [Chapter content](https://wiki.cybernetic-futures.org/pure-vsm/part-05-method-and-epistemology/ch25-systems-as-an-act-of-recognition/ch25-systems-as-an-act-of-recognition-content) · Overview diagram # Ch 25 — Systems as an act of recognition: observer, variety, control **Part V — Method and epistemology.** The most detailed account Beer gives of what a system is — and of how much of control theory follows from the observer rather than from the world. ## The question What is a system — and who decides where its boundary lies and how much variety it has? ## The reading To speak of a system is to speak of the coherence of a number of entities called its parts (DC 1966: 241); a system is not something given in nature but something defined by intelligence (242). A collection is recognized in three steps — coherence, pattern and purpose — and is not dignified by the unitary notion of a system until some unifying purpose is devised for it; and all three are acts of mental recognition rather than characteristics of physical things (242–43). It follows that it is not possible to prove that a system exists, only that treating a certain collection of things as a system is helpful (243), and that this act of recognition is precisely the formulation of a conceptual model of the system (246). Beer repeats the point in 1980: a system is not something presented to the observer but something recognized by him (TBYT 2009, "Preface to Autopoiesis": 68). The boundary is part of the argument. Systems are never completely closed; the relatively isolated system is a set of entities separable from the rest of the world except for two chosen sets of relationships, its inputs and outputs (DC 1966: 273). Which relations count as inputs and outputs depends on how the system is defined, and that depends on who the observer is — the dilemma of modern physics appearing once again (274); to discuss control the open system has to be closed by an artificial convention, and the closure is artificial (288). What management deals with is a particular kind: exceedingly complex to the point where full definition becomes impossible, extremely probabilistic to the point where causal relations are no more than tendencies, and internally self-regulating to some extent (275). Variety inherits the same condition. Complexity needs a clear concept and its scientific name is variety (DC 1966: 247), but measuring it depends on how the system is defined and on who defines it (247); every conceptual step that enriches the system increases the information about it and proliferates its variety (252), which is why what makes control difficult is exactly that proliferation (253). The laws follow. *Decision and Control* gives Ashby's law in Ashby's own words — only variety can destroy variety (279) — and adds that requisite variety need not exist as a raw count of items, provided the purposive assemblage is capable of proliferating it (280). The image later used for the Three–Four homeostat appears here already: the self-regulatory control implicit in a viable system absorbs proliferating variety like a sponge (284, 293). And the black box is not a problem but a resource: assumed able to take on any internal arrangement of input–output connectivity, it can proliferate maximal variety, and that variety is the raw material of a control mechanism competent for real situations (293, 296). Against this, the plain man and especially the classical scientist copes with complexity by killing off variety until almost none is left (296) — which is why Beer needs the deliberate variety generator, a device organized to produce it for a purpose (297). ## Beer locators DC 1966: 241–47, 252–53, 273–75, 279–80, 284, 288, 293–97; TBYT 2009, "Preface to Autopoiesis": 68. ## Status Draft, *Pure VSM* v0.8 (2026-09-23). Beer's texts govern where the draft and they differ. Checked: the passages were read in the *Decision and Control* text layer; the italics of the e-book were used where the print could not be consulted. ## Open points - A possible addition to §25.1, reported but not verified here: *Decision and Control* chapter 14 says the observer projects his own notion of purpose onto the system (DC 1966: 357, also 361–62). - Two italics cannot be settled without a printed copy: "Cybernetics and Management, I" (288), where the e-book italicises the comma and the pronoun as well as the title, and "95" set in italic (298). ## Related [Part V — Method and epistemology](https://wiki.cybernetic-futures.org/pure-vsm/part-05-method-and-epistemology), [Ch 24 — Operational research, models and homomorphism](https://wiki.cybernetic-futures.org/pure-vsm/part-05-method-and-epistemology/ch24-operational-research-models-and-homomorphism), [Ch 26 — Knowledge of knowing](https://wiki.cybernetic-futures.org/pure-vsm/part-05-method-and-epistemology/ch26-knowledge-of-knowing), [Observer and recognition](https://wiki.cybernetic-futures.org/pure-vsm/concepts/observer-and-recognition), [Variety and regulation](https://wiki.cybernetic-futures.org/pure-vsm/concepts/variety-and-regulation), [Algedonic signal](https://wiki.cybernetic-futures.org/pure-vsm/concepts/algedonic-signal), [Decision and Control](https://wiki.cybernetic-futures.org/pure-vsm/entities/decision-and-control), [Viable System Model](https://wiki.cybernetic-futures.org/pure-vsm/entities/viable-system-model).