17.1 The Massey Lectures (Designing Freedom)
“Individual freedom has been lost, momentarily at least, in Chile” (DF 1974, Lecture 3). Beer said this on the radio in autumn 1973, in the third of six lectures broadcast as the thirteenth series of the Massey Lectures and printed as Designing Freedom (DF 1974; cited by lecture, since the transcript has no print pagination, see sources.md). The lectures are one of two places where he treats freedom explicitly. The other is chapter 6 of The Heart of Enterprise. The lectures state the problem for a general audience; the later book gives the argument its most precise form, so they are taken here in that order.
The lectures avoid the notation of the viable system model, but the argument is the one developed in Brain of the Firm (see chapter 16 for the Chilean background). It proceeds in six steps, one per lecture.
Lecture 1 — institutions as dynamic systems. Beer begins by changing how an institution is to be seen. Institutions “are not just things, entities we recognize and label. They are instead dynamic and surviving systems” (DF 1974, Lecture 1). To make this concrete he uses a physical model, a tennis trainer: a ball hangs from a piece of elastic strung between two poles, and two men sitting on top of the poles pull on the elastic. When the ball is disturbed it bobs about and eventually comes to rest. “The time it takes to stop is called the relaxation time of the system” (ibid.). Relaxation time, the time a system needs to return to stability after a perturbation, is the key concept of the lecture.
Then Beer complicates the model. He adds some forty poles in a circle, each with a person and a piece of elastic of different strength, and notes a counter-intuitive result: “the harder all the men try conscientiously to manipulate the system so that it settles, the more unstable it is likely to become” (ibid.). He also lets the world intervene, in the form of a cat that swipes at the ball. The general statement follows: “All our major societary institutions are high-variety systems; all of them need to have a finite relaxation time; but all of them are subject to constant perturbation” (ibid.).
How do institutions cope? They reduce variety, Beer answers, and all of them do it in one form or another. He summarizes the vocabulary he has built: “A dynamic system is in constant flux; and the higher its variety, the greater the flux. Its stability depends upon its net state reaching equilibrium following a perturbation. The time this process takes is the relaxation time. The mode of organization adopted for the system is its variety controller” (ibid.). Hierarchy is one way of reducing variety, for example a boss who constrains the freedom of subordinates. Beer names the cost and offers no easy escape: “It is in this way that freedom starts to be subordinated to efficiency; but the only alternative—which we must face—is total anarchy” (ibid.).
The lecture ends in a hypothesis, stated in its accompanying notes: “The relaxation time of the institutional system is now on the average longer than the average interval between perturbations, with the result that the institutional system is permanently unstable” (ibid., notes). This is the core diagnosis of the series. In terms of the ball: if the cat strikes again before the ball has come to rest, the ball never comes to rest. The system is permanently in transit, and the variety-reducing measures it adopts to cope are applied more and more heavily.
Lecture 2 — the disregarded tools. The second lecture explains why this happens and what could be done. “Our institutions are failing because they are disobeying laws of effective organization which their administrators do not know about” (DF 1974, Lecture 2). Beer means the laws of variety set out in the earlier chapters of this book. He repeats the point of the first lecture in sharper form: “A social institution is not an entity, but a dynamic system” (ibid.). The tools that could help are available but misused. The computer in particular is treated as a data processor; yet it “is really their only hope” (ibid.). The distinction between processing data and producing information returns in Platform for Change (see chapter 18.5).
Lecture 3 — the liberty machine. The third lecture applies the argument to government. Beer recalls his earlier claim that government “is a kind of machine meant to operate the country in the interests of individual freedom”, but that it does not work well. He proposes a change of conception: “So, I declared, let us redesign this ‘liberty machine’ to be, not an entity characterized by more or less constraint, but a dynamic viable system that has liberty as its output” (DF 1974, Lecture 3). On this conception freedom results from how the system is organized, and is no longer a quantity that the state grants or withholds.
Beer immediately strips the state of its aura: “There is no need to be overawed by the pomp and circumstance of the state, once we have found the scientific way in to the problems of effective organization, and understood the basic laws of variety” (ibid.). He goes so far as to say that the state and a department store face the same problems, because “the pattern for handling variety, which last time I called the model, is the same” (ibid.). This is the invariance claim that underlies the viable system model generally: organizations of very different kinds map onto the same model.
Lecture 4 — science and the individual. The fourth lecture turns from institutions to the individual and asks how much of the individual’s variety actually reaches the system. Beer describes a chain of attenuators applied by culture and media, and singles out one of them: “The editorial decision is the biggest variety attenuator that our culture knows” (DF 1974, Lecture 4). His verdict on the present state of affairs is blunt: “I reckon our existing liberty to be largely illusory: we are fooling ourselves” (ibid.). Formal rights may be intact; what Beer points to is that the variety of what people can see, say and choose is heavily filtered before it matters.
Lecture 5 — bureaucracy. The fifth lecture puts the central question of the series in one sentence: “How do we sustain individual liberty and societary cohesion at the same time?” (DF 1974, Lecture 5). The Heart of Enterprise will answer it formally (17.2). The obstacle, in Beer’s diagnosis, is bureaucracy. Inside institutions there is a nucleus “that exists—narcissus-like—in self-regard” (ibid.), and institutions produce their social benefits “simply as by-products of their major bureaucratic undertaking, which is to produce themselves” (ibid.). This is the popular form of the argument about pathological autopoiesis (see chapter 15.4): a subsystem that becomes concerned with producing itself rather than serving the whole. The remedy Beer draws is structural: “If institutions really are to be changed, then their fundamental organization really must be altered; and a major component of that alteration will be to dismantle the bureaucracy” (ibid.).
Lecture 6 — the free man in the cybernetic world. The last lecture gathers the conclusions. The first addresses the fear that efficient organization must mean loss of freedom. Beer argues that behavior which looks like individual choice is often “an output of a dynamic system having a particular organization” (DF 1974, Lecture 6), and concludes: “And so my first conclusion to these lectures is: efficiency does not entail tyranny—if we can get the system right” (ibid.). He adds why this matters: “To do so is a top priority, because some version of efficiency is required to save our dinosaur society” (ibid.). The same lecture contains his argument about adaptive planning, taken up in chapter 19.4.
The series closes by drawing the line from diagnosis to task: “Thus is freedom lost; not by accident, but as the output of a system designed to curb liberty. My message is that we must redesign that system, to produce freedom as an output” (ibid.). The phrase “freedom as an output” is the thesis of the whole series. It puts freedom in the same category as any other output of a system: something that results from structure and can therefore be designed for.
17.2 The argument in HOE: freedom as a computable function
Chapter 6 of The Heart of Enterprise is entitled “Freedom”. Where the lectures used images, it gives the formal version in the vocabulary of the viable system model. Beer locates freedom in the geometry of the model. The model has two dimensions: a horizontal one, in which the management units, their operations and their environments interact, and a vertical one, in which the metasystem relates to System One. Beer states that in the VSM “the dear question of freedom is locked into its orthogonality” — that is, in the relation between horizontal (operations) and vertical (metasystem) variety, an interaction “which DEFINES FREEDOM within the viable system” (HOE 1979: 145–146). Freedom, then, is a property of how the two dimensions interact, and belongs inside the model from the start. From this Beer draws a conclusion that runs against a common assumption. The loss of freedom is usually blamed on the top. Beer argues that it often arises in the operations themselves, through the interaction of the operational elements with each other: “The loss of autonomy, the erosion of liberty, is already implicit in the total system […] — but the fact goes unrecognized, and is blamed on autocratic corporate management” (HOE 1979: 152). He also moves the question from feeling to structure. The reality of freedom in an institution “lies, not in the subjective impressions of the people involved, but in what the institution DOES to them by way of variety constraint” (HOE 1979: 156). This matches the fourth lecture’s claim that liberty can be illusory. Whether people feel free and whether they are free are different questions, and the second is answered by examining variety. Beer then develops the argument in seven steps (HOE 1979: 157–158). In summary:
- The management unit has less variety than its operation, and the operation less than its environment; the unit must therefore design amplifiers and attenuators to balance these horizontal variety equations.
- The other elements of System One intervene in these horizontal equations, because the operations interact. This “is a mark of the total cohesion, but it tends to vitiate elemental autonomy.”
- If the metasystem intervenes as well, in the interests of fairness, it is likely, on the whole, further to reduce the variety available to the management unit. “Therefore The metasystem should make minimal use of variety attenuators in its dealings with management units in a downward direction.” Beer comments: “This is the cybernetic argument for autonomy (or freedom), as distinct from the ethical, political, or psychological arguments. Unlike these others, it has no emotive content. It is basically mathematical.”
- The minimum of intervention is in principle zero.
- But without any intervention the elements would act inconsistently, or even contradictorily. “Therefore The metasystem must make some intervention, and should make only that degree of intervention that is required to maintain cohesiveness in a viable system.”
- Cohesiveness depends on purpose: systems of concentrated purpose are closely knit, systems of general purpose more loosely coherent.
- Systemic purpose, however, is “a subjective phenomenon, rather than a property of the system independent of its instigators, participants, and observers.” The argument squeezes from two sides. Steps 3 and 4 push the metasystem’s intervention toward zero, because every intervention costs the operational unit variety. Step 5 stops it from reaching zero, because a viable system must hold together. The permissible amount of intervention is therefore the minimum required for cohesion. Steps 6 and 7 say what fixes that minimum: the purpose of the system, as those involved perceive it. The ball of the first lecture shows the same shape in miniature: the harder the men pulled to make it settle, the less stable it became. The conclusion follows: “THEREFORE Freedom is in principle a computable function of systemic purpose as perceived” (HOE 1979: 158). Beer anticipates the reaction: “That is the explosive conclusion. It is explosive precisely because it sounds heartless, whereas the dear question of freedom is full of heart” (ibid.). He does not retract it. Calling the argument “basically mathematical” in step 3 means that it holds whatever one’s feelings about it. When he summarizes the model later in the book he restates the result in the same terms: “The important principle was deduced that freedom for the operational unit is a computable function of the purposes of the system as perceived” (HOE 1979: 200). Two features of the conclusion need spelling out. “Computable” does not mean that anyone has computed it. It means that the degree of autonomy is not arbitrary: given a purpose, the minimum intervention needed for cohesion is in principle determinate. And the purpose is “as perceived”. Because purpose is subjective in Beer’s sense, the calculation cannot be fixed from outside. People who perceive a system’s purpose differently will arrive at different permissible degrees of intervention. On my reading, freedom is therefore bound up with the question of who defines the purpose, which is the business of System Five. This link is an interpretation, not Beer’s statement. Beer extends the argument beyond the enterprise: this explication “can be extended to cover individual freedom within a society (since we are dealing with the identical cybernetics exhibited at a different level of recursion)” (HOE 1979: 160). With that sentence the two halves of the chapter meet. The “liberty machine” of the third Massey Lecture and the management unit of The Heart of Enterprise are the same problem at different levels of recursion. The fifth lecture’s question, how to sustain individual liberty and societary cohesion at the same time, receives here its formal answer: hold metasystemic intervention to the minimum that cohesion requires, and measure cohesion against a purpose that the participants themselves perceive. The same logic governs Beer’s rules for planning, where a uniform planning rubric imposed from above is ruled out because it would impair the autonomy of System One (see chapter 19.3).
References
- DF — Beer, Stafford (1974). Designing Freedom. London: John Wiley. Cited by lecture from a transcript edition.
- HOE — Beer, Stafford (1979). The Heart of Enterprise. Chichester: John Wiley & Sons.