10.1 Why System Four is needed
The external environment of a viable system, Beer writes in HOE, “includes all the environments of the elements of System One, but much more. In particular, it includes the problematic environment that the future represents: the playground of opportunity. Then it becomes a necessary condition of viability that there should be a system within the metasystem that can deal with this. System Three is debarred, because it is dedicated to the internal environment, and System Four is born” (HOE 1979: 143). Up to this point the metasystem (Systems Three, Two and the command channel) stabilizes the internal environment (see chapters 8 and 9). Everything it knows about the outside world reaches it through System One, where the environment appears as a disturbance to be absorbed. For the future, that is not enough. The reasoning follows the pattern of the other subsystems. Beer identifies a function the viable system cannot do without, and assigns a subsystem to it because none of the existing ones can take it on. System Three cannot, because it is “dedicated” to the inside-and-now. DTS adds a second task: “SYSTEM FOUR is not only concerned to manage the outside-and-then, but to provide self-awareness for the System-in-focus” (DTS 1985: 115). The second task becomes central in 10.2. Beer distinguishes two environments of System Four: the accepted environment (“the total envelope”) and the problematic environment, “special to System Four, and marked with the interrogation sign” (HOE 1979: 228). DTS draws the same distinction as two loops, Alpha and Beta. The Alpha loop watches what is going on in the environment of the outside-and-then; the Beta loop “is anchored in THE UNKNOWN FUTURE. Hence the ‘?’” (DTS 1985: 119). They face different problems: “Monitoring what is actually happening in the big wide world, and correctly assessing the trends, is different from being alert to NOVELTY. Folk too often assume that the future will be an extrapolation of the past — and then someone invents the transistor” (DTS 1985: 120). System Four also sends. It is “involved in containing environmental variety […]; it is also concerned in generating that variety” (HOE 1979: 237). Beer calls this “the issue of initiative” (ibid.). Forecasting is only part of the task: “most people appear to believe that ‘The Future’ is something that will happen to us, whereas it is certainly open to humankind to take charge of these events” (HOE 1979: 237). The problematic environment, Beer writes, was defined as special to the viable system in question: “how will it invent the future?” (ibid.).
10.2 System Four must contain a model of the whole
In HOE Beer recapitulates the Conant–Ashby theorem (chapter 2.9): “that any would-be regulator must contain a model of that which needs to be regulated, and that if this model has a given level of variety, it will handle only that equivalent variety in the outside world” (HOE 1979: 235). Then he applies it: “although the theorem is completely general […], it has a very special application in System Four. This is because System Four houses the viable system’s whole apparatus for adaptation. It follows that System Four must contain a model of the viable system of which it is the System Four. In current managerial practice, System Four is hardly recognized: still less does it contain this model” (HOE 1979: 235). He guards at once against a misreading. By a model he does not mean a corporate model in the usual sense of “a set of simultaneous equations fed into a computer”. “Instead, we are discussing a cybernetic model of the necessary and sufficient conditions of any viable system, as illuminating the managerial structure and behaviour of that organization” (HOE 1979: 235). The model System Four needs is the VSM itself, applied to its own enterprise. In practice this model “is held in the heads of the senior people: it is not made explicit, and cannot be interrogated or tested” (HOE 1979: 237). Beer draws a concrete consequence. The enterprise does generate creative ideas, “usually at junior levels, and these conflict with the paradigm held at the senior levels. They are usually squashed” (HOE 1979: 237). An implicit model cannot be checked against new ideas; it can only accept or reject them. The operations room of 10.4 is where Beer proposes to display the model. DTS draws the recursive consequence. System Four’s model of the whole includes System Four itself: “What we have to do is to constitute our Figure 30 as a model of its own (that is, System Four) self” (DTS 1985: 115). So System Four also contains a model of itself, and this model contains a model of itself, “indefinitely”. “As hinted earlier, it is just this infinite regression of self-images that seems to hold the key to the characteristic self-awareness of viable systems” (DTS 1985: 116). Beer marks this as a step towards closure: the self-reference in System Four “is the operational basis for the final self-referencing, system-closing, System Five to which everything is now leading” (DTS 1985: 115; see chapter 11).
10.3 System Four in practice: disseminated, not integrated
“This is the weakest component of the viable system in existing management orthodoxy. System Four exists — it has to: but it is disseminated, and lacks integration” (HOE 1979: 143). Beer’s diagnosis of existing enterprises holds across the books. The activities that belong to System Four are real, and he lists them: “Market Research counts. Corporate Planning counts. Economic Forecasting counts. Management Development counts” (HOE 1979: 231). Ask how they are integrated, and “The truth is: THERE IS NO SUCH INTEGRATION” (ibid.). They “are disparate, are disseminated, are active all over the organization in the cause of special pleading” (HOE 1979: 231); “It is quite normal, in a large enterprise, for the elements of System Four to have virtually no knowledge of each other’s activity” (HOE 1979: 232). DTS makes the same point about corporate planning, which may proceed “as if the R & D and market research were simply inputs to a computer programme”, so that the links between them may amount to “a total disjunction” (DTS 1985: 115). The verdict in DTS: “VERY FEW ENTERPRISES have a well-functioning System Four” (DTS 1985: 117). It matters more now than it used to, Beer explains. Before the Second World War the boss was primarily concerned with the internal environment: “He could deal with the problems of System Four in the bath-tub: such as once a month” (HOE 1979: 231). The expansion of technology “has meant that System Four is in continuous action” (ibid.). Disseminated System Four activity was tolerable when the rate of change was low. It is not tolerable when change is continuous. Under stress, enterprises show a characteristic reaction. One of their early responses to a threat is to cut back System Four, “this being (supposedly) the organ of adaptation. On the face of it, this reaction is suicidal” (DTS 1985: 121). Beer adds a qualification that places responsibility on both sides: “if it is not, then System Four was either unprofessional or its Senior Management was not in good faith” (ibid.).
10.4 Focus and the operations room
If System Four is disseminated, the obvious remedy would seem to be to put it all in one place. Beer rejects this: “That solution would be to amalgamate all the elements of System Four in one integral department. I think it is mistaken, because it would involve a colossal proliferation of variety that no management could handle” (HOE 1979: 233). His answer is “Focus”: “all elements of System Four (as it were, optically) ‘project’ their activities, in full variety, onto the same screen.” He sees at once that the optical metaphor breaks down, because detailed pictures on one screen would blur each other, and concludes: “Then I propose that the ‘screen’ on which the elements all focus is none other than the model of the viable system whose System Four they comprise” (HOE 1979: 233). The model that System Four must contain (10.2) is also the instrument that integrates its parts. The operations room gives that screen a physical form: “An Operations Room, considered as the physical manifestation of our focus — in which in particular the kernel of the System Four model of itself is displayed — might take on any form” (HOE 1979: 243). The Operations Room outlined in BOTF, Beer notes, “was subsequently built for President Allende in Santiago, Chile” (HOE 1979: 242), and the experience of using it did not outlast 11 September 1973 (ibid.). Such a room must satisfy three conditions, because its users are “human beings constrained by their own neurophysiological limitations”: “it is absurd to overload them with data that they cannot ingest, absurd to present those data in ways that actually suppress their inherent patterns, and especially absurd to maintain data that are out of date” (HOE 1979: 243). Every operations room he had visited, he reports, showed at least one of these faults, which is why the concept had fallen into disrepute (ibid.). DTS keeps the idea: “At best, we need some kind of management centre (I have also called it an operations room)” (DTS 1985: 117). It also warns against the usual substitute: “do not simply constitute those red dotted lines as a bunch of committees. They will not have Requisite Variety” (DTS 1985: 117). In chapter 11 the operations room returns as the place where System Three and System Four meet. BOTF calls System Four “the development directorate of the firm” (BOTF 1981: 181); HOE likewise calls an integrated System Four a “Development Directorate” (HOE 1979: 236, 242). Beer knows the proposal meets resistance. A Development Director in his sense “is not a ‘Chief Boffin’, a tame intellectual. He is a full member of the senior management” (HOE 1979: 241). He also records a common failure in appointments: “the man who usually gets the job (under whatever title) is the one who — his colleagues think — has the least chance of accomplishing change” (HOE 1979: 242).
10.5 System Four is not senior to System Three
System Four deals with the future and holds the model of the whole, so it is tempting to place it above System Three. Beer rejects this: “Let us also remember that System Four is in no sense senior to System Three: they are accountable to each other for the huge variety that each disposes in carrying through its own activity” (HOE 1979: 252). He continues: “But they are not in business to attenuate each other’s inputs, nor to inhibit each other’s disposable variety” (ibid.). HOE also places System Four “in the corporate management (and not the merely advisory) domain” (HOE 1979: 241), which is why the Development Director must be “a full member of the senior management”. With Systems Three and Four side by side, each handling high variety in its own domain, Beer describes the metasystem as “schizoid — torn between present and future preoccupations” (HOE 1979: 252). The channels between them cannot carry requisite variety without locking them “together in a solid embrace” (ibid.). Chapter 11 takes up how that tension is resolved, through the Three–Four homeostat, the operations room turned into a meeting place for both, and System Five.
10.6 System Four in BOTF: “Environments of decision”
The earliest full treatment of System Four is Chapter Thirteen of BOTF, “Environments of decision — system four” (BOTF 1981: 181–199). It opens with the two inputs to decisions at the top. The autonomic condition and the filtered information on the central axis reach System Five “mediated by System Three but, as we have seen, they are switched by System Four” (BOTF 1981: 181). The second input is information about the outside world: “All indications of relevance here are collected by System Four as direct input from the outside world, and they too are switched into System Five” (ibid.). Figure 33 enlarges the box of System Four, “which is the development directorate of the firm” (ibid.); System Five appears as a sensorium and a motor cortex (BOTF 1981: 182). Both the sensory and the motor representations of the world in System Five “are generated by the activities of System Four”, and they must match the world outside: “Were it not so, the organism would be in a state of hallucination” (ibid.). With this loop between sensation and action, “At last the story of the central command axis is complete” (BOTF 1981: 183). Beer draws a consequence for planning: “a corporate plan must continuously abort, for unless that plan is continuously aborting, how shall sanity (the match between the world and the firm’s notion of it) be preserved?” (ibid.). The second part of the chapter states the model requirement that 10.2 follows through HOE and DTS. “Remember that there is always a System Four, even if it is not identified in quite the form specified here. […] We may then say that System Four must contain a model — some model — of the corporation” (BOTF 1981: 183). Beer then builds such a model in stages. Figure 34 shows the cash flow of a new subsidiary for one epoch; Figure 35 closes the reinvestment loop, with a feedback box F(s): “It is precisely the corporate planning function of the enterprise” (BOTF 1981: 184–185). Figure 36, derived from work by R. H. Anderton, adds a second regulator, the match between product attributes and market demand, and five control parameters: capability to improve products (A), to innovate products (B), to improve efficiency ©, product market inertia (X) and money market inertia (Y) (BOTF 1981: 187–188). Figure 38 maps an extension of the model onto the corporate structure, and Beer warns against a misreading: “The diagram at Figure 38 is not an alternative to, nor even an elucidation of, Figure 27. It is explicitly the diagram of a model of System Four. The reason why it looks like the total figure is that System Four must in some sense mirror or map the totality it serves” (BOTF 1981: 192). The development director controls the functions that acquire and evaluate information and implement the adaptive planning agreed with System Five (BOTF 1981: 192). Beer’s primary concern, though, is “the physiological role of System Four as ‘the biggest switch’”, at “the conflux of information between the central command axis and the outside world” (ibid.). The existing fragments of System Four have no means of communication with the board or the operations directorate beyond formal committee work, and Beer takes his remedy from the wartime “war room” (BOTF 1981: 193). “I propose a control centre for the corporation which is in continuous activity. This would be the physical embodiment of any System Four. All senior formal meetings would be held there […]. PAPER WOULD BE BANNED FROM THIS PLACE” (BOTF 1981: 194). Its four walls display an animated, quantified Figure 27; information recalled in graphic form; prognoses; and an analogue version of the Figure 38 model whose knobs set the five parameters (BOTF 1981: 194–197). This is the Operations Room that HOE reports was later built in Santiago (HOE 1979: 242; see 10.4). The chapter ends on the purpose of the whole arrangement. Most System Four work serves “stating what has happened in the past”; the System Four envisaged here builds a model to be used “as a tool for inventing the future (in Gabor’s phrase), rather than fearfully predicting what it may hold” (BOTF 1981: 198). “Most enterprises are directed with the driver’s eyes fixed on the rear-view mirror” (BOTF 1981: 199). A note to the second edition says that the chapter “has been left in its original form for strong historical reasons, which will become manifest in Part IV” (ibid.).
References
- HOE — Beer, Stafford (1979). The Heart of Enterprise. Chichester: John Wiley & Sons.
- BOTF — Beer, Stafford (1981). Brain of the Firm. 2nd ed. Chichester: John Wiley & Sons. (1st ed. London: Allen Lane, 1972.)
- DTS — Beer, Stafford (1985). Diagnosing the System for Organizations. Chichester: John Wiley & Sons.