co-criticality

If we truly believe that we treat the systems we manage as if they really are systems, we have to be extremely careful about the notion of priority. Developing countries, for example, need to build houses, roads, schools, and hospitals. What are their priorities? Do they build houses, and find that there is nowhere for the children to go to school; or do they build schools that are unattended because no-one lives anywhere near? Do they build hospitals that no-one can reach as soon as it rains; or do they build roads that stop suddenly in the middle of nowhere? Any reader who thinks that these questions are fatuous cannot have been to a developing country. Setting priorities is a very dangerous game indeed. The process partitions the total system, and gives power on a relative scale to its parts. It is notorious in the history of World War II that Lord Beaverbrook's single-minded success as Minister for Aircraft Production, when planes were desperately needed, nearly lost Britain the tank war in North Africa....

Priorities can be set: very often people become disoriented if they are not. The point is that there is a price to pay; and it always has to do with a failure to understand the system, and a subsequent loss of systemic synergy. Synergy has to do with physiology; that is to say with the body that happens to consist of all the parts. One can imagine the concept mechanically too. Think of a piece of machinery in which a dozen subsystems are all working away to do something useful. On top of each is a revolving cogwheel, not connected to anything. Then suddenly a lever is thrown. The twelve floating cogwheels come together. The dozen subsystems continue to play their separate roles; but now the entire machine rises up on caterpillar tracks, and moves off. We could have set priorities between the subsystems, considering each on its own merits, and in so doing we could have failed to understand the first thing about the nature of the system as a whole. But this is not going to concern those whose psychological paradigm demands that priorities necessarily exist, because they think that they understand the purposes of the system. We come full circle: if the purposes are known, then the priorities are clear. It is all very convenient — and all very damaging.

Pray consider this issue in an entirely human context. We are all members of a board which is due to reconsider its fundamental raison d'être. What is this 'thing' of which we are board members; what are we really trying to do; and how? Now this scenario is very familiar. Boards of companies and of national services are implicitly faced with this problem all the time. They usually pretend that they are not, by carrying forward the minutes of one meeting into the agenda of the next, and simply slipping in — unobtrusively — any new item which has raised its head. And if this device fails to work, then they have that ultimately incompetent item 'Any Other Business' in which to catch the residue. But sometimes a meeting really will convene with the explicit purpose of reviewing purpose. Or a whole conference may be set up to consider, ab initio, how to handle an entirely new issue in human affairs.

Pray consider, then, how the convenors of this meeting set about their task. They identify key issues (— and yet if we actually knew what are the key issues, maybe there would be no problems: all of this hinges upon the acknowledgement of purpose). Next, they identify alternative strategies within the key issues (— and yet if we actually knew all the strategies available, the answers would probably be obvious: all of this hinges upon the acknowledgement of systemic boundaries). Finally, they put the agenda in a particular order (— and yet if this order were actually known, a computer could track through to the proper conclusions: all of this hinges upon the attribution of priorities).

Again, it is surely clear that this systematic structuring of the meeting is unsystemic. We are meeting to try to understand the nature of the system with whose management we are charged. We do not understand it; but all of these preliminaries are based upon the implicit contention that we do. The situation is like that of a man on his hands and knees in the middle of the night, grovelling in the ground under a lamp-post. A policeman asks him what he thinks he is doing. 'I am looking for a lost ring.' 'Where did you lose it?' 'Over there', says the man, pointing ten yards away into the dark bushes. 'Then why are you hunting for it here?' 'Because this is where the light is.'

The board has a number of lighted lamp-posts. They are labelled 'cost effectiveness', 'locations', 'staff availability', 'liquidity', and so on. They illuminate very familiar patches of terrain. But the answers the board seeks are in the dark bushes. There is no light over there; there is no guide into that unknown. Thus even when the board recognizes that it does not understand its priorities, it has no hesitation in setting up specialist committees to consider 'obvious areas of concern'. It will set up the priorities when it receives the committees' reports. That is an advance: it shows some care for the priority trap. However, all the committees are sitting around lamp-posts. Well, fair enough — they have to be able to read the minutes of the last meeting, surely....

Let us now demolish the whole of this apparatus, on the grounds that it is systemically incompetent. Then we have to ask what that leaves. It leaves a collection of individuals each of whom (we may certainly hope, because of his or her membership of the board) understands something about the issues. Of course Mr A is well-known to be a world expert on x. Of course Dr B is well-known to ride a hobby-horse about y. They have respectively bored us to death already on their respective topics. Ms C, on the other hand is wise and learned, and does not seem to represent a vested interest: etcetera.

Pray consider now a large room. There are (say) ten tables in this room, surrounded by easy chairs and supportive refreshment. Beside each table is a vertical pole; on each table is a stack of large white cards. The protocol of the meeting is that anyone who reckons that he or she has a point to make, one that is relevant to the policy that we are met to create, should proceed to one of these tables, write the name of a topic on a white card, and fix it to the top of the pole. Obviously Mr A and Dr B will install themselves without delay. It is to be hoped that Ms C will do so as well — although the topic that she announces may well occasion surprise, because no-one else had thought of that.

The protocol continues thus. Members of the board may sit down at any table, and join in the discussion. Meanwhile, they may observe the apathy or frenzy of other groups, and watch the emergence of new tables, set up under new banners. It should be remembered that these people know each other, and can 'read the signs'. Anyone is entitled to move, without apology. Anyone is entitled to initiate a new table.

It should be obvious already that boredom cannot set into this situation. Much experience of meetings has convinced me that boredom is the dominating factor in decision-taking. Any conclusion, however obnoxious, is better than sitting through one further minute of this absurd drivel. And yet we sacrifice to the graces: it is simply not permitted that one should scream 'enough! I'm leaving'. Under the new protocol, this is possible without offence; one simply wanders away.

As soon as the group sitting round a table recognizes that nothing will come of its deliberations, it may either disperse, or change the topic inscribed on its masthead. Others will notice those changes, taking note of topics, and taking note of those involved. If, on the other hand, those at the table consider that they have agreed on something worth saying, they write it down, and each person present signs the note. There are no rules whatever about the format of any such declaration. It could be a sentence asserting a belief, a declaration about a boundary, a draft paragraph towards a policy: it could be anything considered worth saying. Outstandingly, it could formulate a crucial question. A signal is then attached to the masthead, while the group continues, or not, as it wishes.

Staff assistants, observing the signals, collect the notes, and deliver them to the Chairperson. The Chairperson (as always) has a most important role. This is, however, no longer to bully and cajole the board into observing the rules set by the agenda — for there are none. It is to sort the notes coming in; it is to be an intelligent computer for the meeting (and to keep quiet).

Gradually, the issues that the board considers worth considering will emerge. They may not be at all in line with the 'lamp-posted' areas of established light. The Chair must of course recognize that likelihood from the start. The Chair's job is instead to recognize what the areas of concern actually are. Thus the Chairperson cautiously sorts the incoming notes into foci of concern, and tries to create an appropriate taxonomy. The staff assistants are at hand to undertake typing and duplicating.

By the first agreed break, let us say luncheon, all members of the board are issued with a first review of their deliberations. Anyone is entitled at this point to add his or her name to any declaration that has been made, via the secretariat. All right: he or she missed the whole discussion (which is a relief), but had thought something like that all along. People are also entitled to start a 'denial' list, or to propose amendments, and these will also gain signatures or not as the process continues.

For, indeed yes, the process is intended to be iterative. It may go on for two or three days, if so arranged, and given that the issues are weighty. The role of the Chair is crucial, and remains so: some issues will be declared closed, as having no general interest, while others will be amalgamated, as entailing similar outcomes. The Chair was always important, but its role now is wholly subservient to the meeting's needs: that may call for a different chairperson. These duties call for considerable mental dexterity, since the first attempts to derive a suitable taxonomy for the system may have to be radically changed as the meeting iteratively shifts its foci of concern. Moreover, there is no scope, under this protocol, for writing the minutes of the meeting before it has taken place...

Quite apart from the flexibility and sense of ease that an enquiry conducted in this way exhibits, the protocol is explicitly designed to avoid the traps that we have been considering. The purposes of this board are gradually emerging as an amalgam of the purposes of each member, considered as a participant observer in the process. This amalgam is not the same thing as a consensus, which may be generated by the traditional meeting. The search for consensus is made within artificial limits, artificial codifications of topic, and artificial orders of priority. That is why the consensus tends to state a lowest common denominator of purposes which often ends up to everyone's chagrin as a 'motherhood' statement. But the amalgam on purpose generated by the new protocol may be expected to include elements that are perceived as important by subsets of board members, while earning no more than amiable acceptance by others. People are inclined to say of such a purpose: 'well, if all those good people think that this aspect of the matter ought to be made explicit and pursued, I daresay they are right. It doesn't really turn me on'. In the orthodox meeting this is an unlikely outcome, because the issue is an agenda item to which all present are forced to address themselves; thus those who are indifferent, finding it embarrassing to say just that, become hostile. In short, the orthodox meeting tends to polarize the board — and this is not only unnecessary, but damaging to a constructive outcome.

Secondly, the new protocol enables board members to probe the boundaries of the system. In the orthodox meeting, someone says: 'why should we not consider what would happen if...?' Someone else says: 'I think that's outside our terms of reference', and appeals to the Chair. The Chair is prudent; nothing happens. Under the new protocol, a small group of members is enabled to develop the what if question to the point where it can be considered as a serious prospect. After all: if something is regarded as important to our purposes, as we observe those to be emerging, how can it be that their consideration takes us outside the boundaries of the system? Surely, those boundaries have been wrongly defined. Even if the definition is merely a matter of internal convention, it will be difficult to change. I have seen the situation in which the definition was determined by the legal interpretation of an Act of Parliament. The agreement remained at this level; no-one on this board (a nationalized industry) would even consider the possibility of moving to amend the Act. It is not surprising. People do accept 'the system' as if it were given in nature, or handed to them on an institutional plate.

What truly is given in nature is a collection of principles that systems cannot violate if they are to maintain their identity. These principles are the content of the science of cybernetics, which may conveniently be defined as the science of effective organization. This book is about the cybernetics of a particular kind of system, that I call the viable system. We shall discover what that means as we proceed. But this ENTRY into the whole discussion (which is the loosening-up exercise) is pointing to an unassailable fact. We cannot successfully handle any system that we are disposed to manage unless we obtain an insight into its nature. That nature, as we have seen, has to do with purposes, with boundaries, and with other matters which are not customarily investigated within our institutions — because their mores effectively forbid it.

The story about 'tails' took us back thirty years. The story about 'meetings' is quite recent. I devised the new protocol in 1970, and have had four opportunities to use it since — although only once in its full setting. (The other occasions did not offer sufficient preparation time, and were reduced in the 'technology' of the meeting.) But in the examples that I have quoted, spanning a quarter of a century in time, and spanning the range of essentially mechanical to essentially human affairs, the questioning of the nature of the involved system itself was paramount.

Ought we not to question that more often?

some diagrams here... yeah. excited

and then use THAT to explain criticality