American Society for Cybernetics
ASC 2001 Conference
May 27-29, Vancouver 
 

Wrigley, Philip A. (2001). A Systems Theory of Deliberative Meetings in Organisations.
Online Proceedings of the American Society for Cybernetics 2001 Conference, Vancouver, May 2001.
http://www.asc-cybernetics.org/2001/Wrigley.htm

Malik and Probst (1984: 105ff) also offer support for the approach adopted here. They identify self-organising social systems created by human action that are only to a limited extent organised and guided by conscious and planned intervention. They must adapt to circumstances unknown to anyone completely. This can be why events deviate from people's designs for them.

These systems are viable because:

1. they are an effective form of co-ordination of activity through modification of the behaviour of members;
    
   
2. social actions often produce unintended consequences that themselves produce further actions; and
    
   
3. they are organised according to rules of behaviour that incorporate the means of adapting to an environment. The rules dominate over conscious goals.

Interestingly, Malik and Probst use discussions and negotiations as an illustration of self-organising processes. Results will usually differ from the goal of any single discussant or negotiator. Their course can be foreseen only within narrow limits. 'The negotiations have the participants under their control, or ... there is mutual control' (p.111). Hence a manager acts in the light of situations guided by general behaviour rules. She does not attempt direct control.

Maturana's position

Maturana (1988: 64ff) has proposed characteristics for social systems consistent with their autonomy (in Maturana 1980a: xxiv-Philip A. Wrigley & 1980b: 12-14). His descriptions are useful for developing the theoretical frame of the next section.

A social system, according to Maturana:

1. is organised as a network of interactions and relations (1980b: 12);
    
   
2. is made up of these interactions (1980b: 12);
    
   
3. is homeostatic, maintaining invariant the relations that define it (1980a: xxvii (6), 1980b:14);
    
   
4. recursively selects properties in its members that confirm it (1980a: xxvii (9), 1980b: 13); and
    
   
5. this process of selection leads them 'to continuously regenerate the same network of interactions' (1980b: 14).

Systems within systems: contained yet free

An autonomous entity is able to express itself with respect to its medium at the same time as its parts assert their own independence. One independent entity can exist within another. A meeting can act through decisions or understandings, but they arise from the self-expression of individuals. An observer is puzzled by the simultaneous observation of two distinct phenomenological domains. One of them is the operation of a simple unity (the meeting) in a medium. The other is the domain of interactions of its components (individuals) as it operates as a composite entity.

Effecting change in small groups

Family therapy has developed prescriptions for practitioners based on autopoietic theory. Family relationships have parallels with the small group processes of a meeting. This suggests an opportunity exists to develop prescriptions for meetings from the theory as well. The therapist can be considered to have the role of the chairperson or facilitator of a meeting.

The therapist must act indirectly. He is not a member of the family: he cannot structurally couple with it as a simple unity as its members do. Rather, the therapist observes the family's network of conversations to identify its organising relations. Then he acts orthogonally. An orthogonal interaction occurs when an agent acts outside the domain of a social system, but within the domain of a member of it (Brocklesby & Mingers, 1999a).

2. A Systems Theory of Deliberative Meetings

The meeting as a social system

Bringing together the theoretical base in autopoietic theory outlined above and the characterisations given to gatherings, it is possible to detail a model of a meeting as an autonomous social system.

I defined a meeting, in part, as an engagement in episodic talk, referring both to the progression through different topics and to the regulation of talk on a particular topic, through turn taking and ordered phases of discourse. Maturana's description of a social system as a network of recurrent and changing conversations resonates with this definition. The brevity of a meeting is no barrier; Maturana (1980b: 12) considers fellow travellers on a bus may form a social system.

A meeting fits with Hejl's characterisation of a social system. Its members interact with respect to a domain of reality that reflects a shared history of structural change (which he calls modulation). They also form a group.

More generally, a meeting is a social system if it conserves its organisation by selecting a trajectory of structural change in its members that leads each to stabilise the cultural relations that define it. This happens as each of them behaves according to its norms. The stabilising factor in the constitution of the system is goodwill, the seeing of the other as a partner in a dimension of living. A member makes an ethical choice to belong. (Maturana, 1980a: xxvi-xxvii.)

Systemic elements of the meeting

An observer brings forth a meeting as an object by making a distinction in language. She cleaves it from an organisational background. If the observer distinguishes that organisation as a composite entity, the meeting is a component of it or a relation between components. When an observer decomposes a meeting into the individuals engaged and the relations between them, it is a composite entity, a system. People are the components of its structure.

Someone refers to 'the meeting' in speech or thought before it starts as if it were an external reference point. Rather, it stands for a set of actions. These are the linguistic behaviours she considers proper for someone taking part. Mendez et al. (1988: 155-6) refer to all such domains as comprising 'the basic implicit operational premises (values, accepted truths, etc.) which [constitute] our social identity at any moment'. These premises in turn arise from a given cultural background.

For a business meeting, the operational premises shared by its participants need to include, as a minimum:

1. joint attention to the matters of business put before them;
    
   
2. a mechanism for drawing upon and channelling the differences between their domains of explanations into the creation of a new domain; and
    
   
3. goodwill amongst them.

Emotional membership is vital to carry the group across the inevitable 'breakdowns' [2] as people find their domains differ. Their history of prior coupling influences the set of actions considered appropriate. This history may have formal artefacts like an agenda and a problem situation as well as cultural and ceremonial aspects.

When an observer credits behaviours as arising from a meeting, she is distinguishing the meeting as a simple unity by describing its properties in interaction with a medium. 'John presented the points we had agreed on to the minister.'

Languaging gives rise to ever-increasing criss-cross patterns of action and response. These interactions form a network. They trigger changes in the structure of participants contingent on the course of their engagement. Maturana (1988: 50) calls this type of structural coupling in language a conversation. The partners in the conversation orient themselves to each other and to a subject through co-ordinations of actions. These in turn are distinguished as objects and are manipulated in a consensual domain. Further co-ordinations arise predicated on an initial co-ordination in a recursive process of interpretation.

The medium of a meeting is its containing background. It has two aspects, niche and environment. The meeting is structurally coupled with its niche, which is the operation of its members as observers of it. Its environment is what an observer sees surrounding it. (Maturana, 1987: 340-1.)

The human capacity to recursively make distinctions and interact with them as though they were independent entities gives us the ability to step outside our experiences in a meeting at the same time as we are components of it. The members comprise its niche as they shuffle back and forward between acting in the meeting moment-to-moment as components and appraising its flow as observers - independent unities.

Participants adjust their actions in consequence of their observations. The very act of formulating a description of the meeting changes their structure and hence their subsequent relationship to it (Varela, 1979: 57). They perturb its network of structural coupling in line with the co-ordinations that the meeting object as a simple unity stands for in their distinctions. The meeting responds to these perturbations through changes in the flow of its conversation that are determined by its structure.

The environment comprises a physical environment and the organisation that gave the gathering its genesis. A meeting has limited interaction with its environment during its course. As classically defined in cybernetics, it is a closed system with no apparent inputs or outputs. It is perturbed by persons leaving or arriving, distractions, deliberate disruptions, the arrival of new information and so forth.

The cognitive domain of a meeting is the set of interactions it engages in with its medium in the course of maintaining its organisation. Interactions that fall outside a meeting's cognitive domain are either irrelevant - they do not exist for it - or are destructive of its organisation. For example, if a whiteboard is available, but no one wants to use it, the whiteboard does not exist for the meeting. If the fire alarm goes off and everyone flees, a destructive interaction has occurred.

As with all composite entities, we can distinguish parts of a meeting at the same time as we can give classifying properties to it as a whole. The organisation of a meeting is the relations between the individuals participating in it that give its essential defining character as a totality, a simple unity. This character exists as properties in interaction with its medium and defines it as belonging to a particular class. The gathering expresses its character in the nature of the outcomes it produces.

The defining internal relations comprise a subset of its social dynamics. Those relations are organised as a conversation. They are only those that must remain invariant for the meeting's character to be maintained. Organisation may disintegrate or change according to the interactions entered into. The conversation has an overlay of meeting organisation that regulates it.

The meeting and the conversation within it are distinct. While relations between people at a meeting form its organisation, its structure includes a conversation. The attendees at a meeting regulate its talk. They confine its conversation within bounds they specify, dividing it into episodes. They maintain a consensual domain of interactions across divides of topic or emotion. An observer distinguishes a meeting object separate from the conversation within it. This is seen in such familiar phrases as 'the feeling of the meeting was... ' and 'I hate meetings!'

Autonomy and its consequences

A meeting is autonomous if it exhibits internal self-generating processes that maintain its organisation and specify the domain of interactions it can enter into in its medium, resulting in regularities in those interactions that an observer can use to identify its organisation. A meeting that meets the requirements to be a social system is autonomous.

A meeting that can reflect upon its own processes can orient consensual participants to co-ordinate their actions in the maintenance of its organisation. It selects behaviours in the group. While the meeting's course of change is a drift, participants perturb the flow to keep it within limits set by organisation. The meeting object sets the character of the conversation. Those taking part venture to maintain it through their actions. The trajectory of their individual structural change is congruent with the structural change of the meeting. The meeting adapts to the changes in those present while they adapt in the course of their conversation to it. The autonomous meeting is organisationally closed (Varela, 1979: 56).

Structural coupling brings about an interlocked chain of interactions. Change brings further change in a recursive process. The participants' autopoiesis - their individuality - and their unique ontogenies make the course of change unpredictable. New distinctions and co-ordinations of distinctions between them maintain a creative flow. Once accepted, an explanation can give rise to a new explanation in the flow of the conversation. Nothing external to the meeting is needed; no feedback loop exists to an environment. The process of explanation can sustain itself. There are no inputs or outputs.

Domain of reality: the generation of knowledge

An autonomous meeting creates knowledge as a consequence of its organising processes. An autonomous entity is able to act recursively on its experiences, applying them to new situations and thereby creating knowledge (Ghose, 1980: 201). Participants specify a domain of legitimate actions in their praxis of living and working together (Maturana, 1988: 33). This new domain of reality encapsulates changes in them that adapt them to their circumstances. Structural changes in them extend or change their individual cognitive domains. By a mixture of curiosity and the suspension of prior assumptions, they can carve out a new set of distinctions that enliven the explanations emerging from their conversation. They can 'breathe life into alternatives that had no previous existence' (Efran et al., 1990: 197).

Difference is an impulse for creative interaction. Meeting processes need to support the expression of differences between the domains of the individuals taking part so that a new frame of reference can be woven. Men and women, as autopoietic beings, are conservative, keeping to their usual patterns unless their experiential and conceptual pot is stirred (Efran et al., 1990: 184, 187, 193, 197).

Acts at a meeting may have significance in a context outside it, but within it involve nothing substantive. For example, a proposal may be adopted without discussion. The group gives it legitimacy by the act of adopting it. As the decision supports effective action afterwards in an organisation, it is nonetheless knowledge.

A new domain of reality is like a frame, or metaphor, with which to re-frame familiar situations to gain new insights. A frame is a re-formulation of experience in language in the relational domain. It is not part of one individual's cognitive process. (Brocklesby & Mingers, 1999b.) An autonomous meeting can therefore overcome the cognitive limits for an individual engaging in re-framing.

Criterion of validation

A meeting cannot differentiate between an explanation that meets its criterion of validation and an illusion that lacks operational coherence for its members in their post-meeting interactions. The coherence of an explanation is that of the meeting. An explanation is valid if the group accepts it. A person cannot distinguish between an illusion and perception at the time of its occurrence. The effect on the structure of the nervous system is the same in each case. (Maturana, 1988: 29.) So with a meeting. An observer can only judge an outcome a 'mistake' later when assessing the domain of explanations within a wider frame.

Any criterion of validation is possible. All of them will use reason, but it is not itself a validator. There are always a priori assumptions to any rational argument. Maturana (1988: 43) calls these operational premises. He asserts that they may change according to the emotional state of the observer and the ontological path or 'manner of listening' adopted.

Drawing upon Geoffrey Vickers concept of an 'appreciative system', the criterion of validation may be encapsulated in the distinction of the meeting object. Standards, norms and values lead to a readiness to notice only certain features of a situation. They specify what distinctions are relevant. Descriptions are evaluated against them.

This leads to regulatory action, the modification of norms and a different evaluation of subsequent explanations, in a circular process. (Vickers quoted in Checkland, 1981: 262-3.)

Serving organisational purpose

Organisational purposes are external to a meeting. In its transient existence it cannot serve the purposes of its encompassing system, even if the people taking part conceive it as intending to do so. It exists 'in the moment' as a closed system. Afterwards, the former participants can serve a wider purpose as they resume roles as members of the organisation. Although a new domain arises from the operation of circular and self-generating internal processes of the meeting itself, an observer will judge it with respect to an external environment. She may consider the meeting 'successful' (but only in hindsight) if it has expanded the cognitive domain of those who participated and they engage in interactions within the organisation that accord with her 'goals'. Its utility then arises from its coherence as a guide to action as those who took part act within their organisation.

Each person will act afterwards according to a personal domain of reality. It is a collective, single entity to the extent it overlaps amongst them. Because individual structural change is continuous, the collective domain frays around the edges in its execution. Explanations supporting the new domain are subject to change without notice. They comprise a story about an experience. They do not replace the phenomenon explained (Efran et al., 1990: 92-3). What seemed valuable explanations and intentions in the meeting may be discounted outside it, even by persons who took part.

The autonomy of a meeting may not serve the organisation's purpose. An observer judges that outcome in an organisational context. Nonetheless, by allowing a meeting to operate autonomously, a manager opens the possibility of creating new knowledge valuable for her organisation. To do so she must avoid the temptation to control specific events in them and specify the explanations that emerge. The meeting must speak for itself. It can do so as soon as it can select behaviours that confirm its internal organising processes.

This still leaves a role for the manager. She can design it and, as chairperson, influence its course. Although attendees engage in a process that maintains meeting organisation, they need not have set the initial organisational design. What is necessary is for them to sign up to it - to accept that it complies with the norms of the group.

Management can link meetings together so they are allopoietic as subsidiary parts of a larger system, but still autonomous when their operation as unities is considered. The operational premises of the members will then include respect for the needs and purposes of the organisation. To ascribe purpose to a meeting is not to deny its operation as an autonomous entity.

An observer can describe an autopoietic component of a composite system as playing an allopoietic role in the realization of the larger system that it contributes to realizing through its autopoiesis (Varela, 1979: 52; see also Hejl, 1984: 73).

What Varela here says of an autopoietic system can also be true of an autonomous social system.

Acting on outcomes

The implementation of meeting outcomes is subject to both the commitment of attendees to their shared domain and to new experiences they encounter. Agreement at a meeting is not the same as its realisation through co-ordinated action afterwards. Moving beyond agreement to commitment requires an emotional contract. Efran et al. (1990: 116) suggest this sense of a contract better expresses the active, even urgent nature of the social arrangements sought than do the terms 'expectations' and 'understandings'. These contracts may be broken; the price is an emotional one.

There is an action connotation arising from a recursive reciprocal involvement between language and bodyhood. People have changed in the meeting, if they were structurally coupled. They could not act exactly the same as before. The concept is quite different from agreement upon information about a subject.

However, the meeting outcome is always subject to validation in a process that begins as soon as people leave the meeting room.

An observer's desire for a meeting's conclusions to be strictly carried out may presume that objective truth about a problem situation - a solution - has been uncovered. However, a decision is not an absolute. The meeting is part of a circular process of interpretation, rather than a lineal one leading to solution. Because new sets of distinctions are always possible, managers looking for solutions face a dilemma in choosing between the agreed upon and the untested new. Mintzberg's work (1973) suggests they prefer the latter. They continually seek out live action, including that arising in the margins of meetings. Like us all, they exist at a node of intersecting domains of reality because they take part in a variety of differing social systems and conversations. Contradictions between domains of action inevitably arise and cause internal emotional conflict (Maturana, 1988: 53-4). What seemed like a solution yesterday might give angst today.

Rather than attempting to choose between complex alternatives for the solution of a problem, a manager engages in problem setting. The genesis for a meeting often lies 'in a breakdown, in which the course of activity is interrupted by some kind of "unreadiness"'. Resolution concerns the exploration of a situation through a collective performance. (Winograd & Flores, 1986: 147, 150.) The manager can name the things the group will attend to and frame the context it will attend to them in (Shotter, 1993: 151-2).

Accountability for system performance

The chairperson accepts responsibility for the organisation of a meeting's internal relations, helping it to express itself. She may be held accountable for its outcomes. An outcome is a judgement of an observer with respect to an organisational background. If the meeting expresses its own will, the chair cannot determine its outcome. She is not responsible for it.

This conclusion derives from the systems view adopted. She conserves the defining internal relations between components. This results in the unique behaviour of the system as a simple unity, in the distinction of an observer with respect to a background. The behaviour (the outcome of a meeting) was determined. It was inevitable given the prevailing initial conditions and the structure of its parts. Determined, but not predictable. Operating like von Foerster's Non-Trivial Machine (1984), there is an infinite number of permutations possible when a response is dependent on an existing internal relation within the structure.

Maturana and Varela (1992: 136-7) give an extended analogy to represent this point. A sailor who has never left his submarine navigates it through marine hazards perfectly. He insists that all he did was maintain certain relation between his instruments, as he always does. No great feat was performed.

3. Drawing Linkages

This section considers first how the epistemological aspects of autopoietic theory may be applied to meetings, then how an existing characterisation of meetings in a policy setting may be reinterpreted in the terms developed here.

Implications of Maturana's ontology of the observer

One of Humberto Maturana's most widely read contributions (1988) builds on his earlier work on autopoiesis and the origins of language to consider the epistemological choices facing the observer in explaining his world. The choice is influenced by his ontological assumption about the nature of being and his ethical choice about relating to his fellows. This is the ontology of the observer.[3] The way we form explanations and relate to one another is fundamental to the operation of a meeting. Here I consider implications for meetings of Maturana's position.

Objectivity-in-parentheses

Maturana (1988: 41) relates the mood of mutual acceptance to the adoption by the observer of a particular epistemology, the explanatory path of objectivity-in-parentheses. In this path, an observer accepts that he has no operational basis to make any statement or claim about objects, entities or relations as if they existed independently of what he does. This path arises from an acceptance that we bring forth our world through distinctions in language.

The observer that follows this explanatory path realises that he lives in a multiversa, that is, in many different, equally legitimate, but not equally desirable, explanatory realities, and that in it an explanatory disagreement is an invitation to a responsible reflection of coexistence (op. cit.: 31-2).

Efran et al. (1990: 181-2) liken conversations in psychotherapy to experiments in a laboratory, where experimentation can occur in an environment designed to afford maximum safety. This is a useful metaphor for a meeting too. 'Safety' here necessitates emotional security, something that mutual acceptance can supply. If the right mood is established, you will take risks. Assumptions are tested in a sheltered environment conducive to discovery and invention. If an explanation survives the testing and probing of a meeting, it can be chanced further afield.

By placing objectivity in parentheses, a new way of listening arises that sets the conditions for engaging in empathy. Exploring the position of the other means suspending judgement upon its objective truth, even if temporarily, and entering into his feelings. I have seen this have a powerful effect in a meeting with heated clashes of truth claims. A non-combatant was able to quickly diffuse tensions and even bring sudden acceptance of aspects of positions that were previously vigorously opposed. Maturana (1988: 76-7) highlights that we can find ourselves in an emotional contradiction if we find we are negating the other in a 'rational' argument, when emotionally we feel warmth (empathy) towards that other.

An observer still applies reason in this way of listening. Reasoning is an individual action. You view and assess your patterns of thinking against a wider background provided by the meeting situation. Thoughts and plans then change by themselves. One person cannot pluck out the bad thoughts of another and replace them, like a mechanic replaces a spark plug (Efran et al., 1990: 159).

Objectivity-without-parentheses

The contrasting explanatory path of objectivity-without-parentheses entails the claim of a privileged access to an objective reality. There is an 'assumption by different observers of different kinds of independent entities as the ultimate source of validation of their explanations' (Maturana, 1988: 29).

An ethical consequence arises.

Observers do not take responsibility for the mutual negation in their explanatory disagreements because this is the consequence of arguments whose validity does not depend on them. It is in this explanatory path that a claim of knowledge is a demand for obedience. (Ibid.)

We use notions of reason and rationality to support arguments under the implicit cultural claim that through them we refer to universal, transcendental truths (op. cit.: 80-1). Efran et al. (1990: 112) describe this as naïve realism and suggest it is the tradition most of us have been brought up in. We mistake opinions for facts and use them to duck responsibility for our actions.

On this path emotional contradictions cause mutual negation. Our emotional commitment to each other in a social contract conflicts with the characterisation of each other's utterances as right or wrong as we grasp for an absolute reality. Mendez et al. (1988: 158-9) describe this dilemma in the context of a family.

If we find it difficult to acknowledge our emotional needs and their influence on our everyday actions, we will search for a compelling argument to justify all actions, including purely personal ones. We equate a description with the underlying event and see a group decision or action as only as legitimate as its justifying explanation.

A meeting that does not bracket objectivity searches for a single ultimate explanation lying beyond it. It may rely on expertise. The 'expert' has a superior access to the truth. The meeting becomes nothing more than a means of instructing its attendees about knowledge brought to it by the expert. It is not autonomous in establishing a domain of reality: one is imposed. This does not create contradictions if we expect a presentation or lecture, but in a supposedly deliberative meeting, it causes resentment.

A meeting may still be autonomous if expert knowledge is simply an input to deliberations, akin to light from a window. The expertise is then outside the cognitive domain of the meeting and irrelevant to the maintenance of its organising principles. For example, members may have engaged in processes that confirmed shared values and norms and in this way maintained their adaptation to joint circumstances. The formal outputs may have little effect on the group. For example, a court is little affected by expert evidence. Matters of law comprise its organising principles.

Validating meeting outcomes

Maturana (1988: 60) describes knowledge as an ability to act adequately in a domain. A claim for knowledge in a meeting is a claim that adequate action will result afterwards. The knowledge is valid within the meeting. Its adequacy outside it is judged later in a wider context. People prepare themselves for this practical reality by suspending, or bracketing, 'objectivity' within the meeting. They are humble about their contributions, acknowledging them to be opinions. They express them nonetheless and encourage others to do the same, as an organising process of the meeting. They engage in a dialogue in reason, but not in truth.

A group's explanations (its domain of reality) need scrutiny by others to be valid in a wider context. They can circulate minutes. They can consult. Their meeting may itself have been part of a validation process for knowledge arising in some other domain. The democratic model forces reference to ever-wider frames of validation in this way. An executive Government can create a domain of reality through its Cabinet meetings. Then it is mediated by reference to caucuses, the bureaucracy, the media and the public. The 'validation boundary' becomes definitional of truth.

This proposition is relativistic. Relativism is the view that the acceptability of knowledge claims is relative to a particular group or community; there are no objective epistemological standards (Boyd, Gasper & Trout, 1991: 780). New sets of distinctions are always possible. There is no ultimate truth.

A relativistic view of knowledge supports Maturana's conclusions. Humility by the observer in making his propositions becomes a practical choice as he pursues their validation. Validation of his proposition is a social process so he offers an invitation to a dialogue.

A series of aphorisms may summarise. Knowledge is always local. No final test of certainty exists. A wider frame can always be drawn. All action is an experiment.

An Illustration of autonomy: Groupthink

Janis' groupthink phenomenon can be re-interpreted as reflecting an intuitive understanding of cybernetic principles, even though he saw himself as engaging in a study of group dynamics (1982: Ch.2). Two contrasting studies show that the values and principles adopted by participants to organise their deliberations have a dramatic impact on the outcome. Autonomy can work for good or ill.

Janis attributes events of international significance to the way leading foreign policy makers organised their meetings. The group norms dominating their meetings could have disastrous consequences. In terms of the theory he:

• relates properties of a meeting (as a simple unity)
   
  
• with respect to a wider background (its medium)
   
  
• to the self-maintaining organising relations amongst the members (the unity's elements considered as a composite entity - a system).

Janis identifies concurrence-seeking behaviour manifested in shared illusions. The meeting partners cannot tell in the moment whether their explanations are illusion or perception. Values were crucial in determining the nature of a meeting.

Groupthink as an emergent property

Janis' opening case study is about deliberations in the 1961 Bay of Pig's crisis. [4] US President John Kennedy later expressed astonishment at what was decided - even through he was there. Processes at the meeting 'took over'.

The properties of the group that Janis finds arose from the interactions of its members. He sources them to an illusion of invulnerability.

Everyone becomes somewhat biased in the direction of selectively attending to the messages that feed into the members' shared feelings of confidence and optimism, disregarding those that do not (Janis, 1982: 36).

There was a feeling of euphoria after Kennedy's remarkable run of luck leading to the Presidency. '"We thought for a moment that the world was plastic and the future unlimited."' These emotions pre-disposed the decision-making group to actions that confirmed the belief in their invulnerability.

An atmosphere of assumed consensus led well-informed critics to remain silent. They were 'reluctant to raise questions that might cast doubt on a plan that they thought was accepted by the consensus of the group, for fear of evoking disapproval from their associates'. They co-ordinated their actions with the presumed preferences of the group. 'Mindguards' within the group interacted with dissidents to head off expression of their concerns.

Kennedy's leadership privileged a CIA invasion plan. It became 'objectivity-without-parentheses' for the group. The President brought about obedience to this plan as being a probable account of what would happen. This seems to have been an accidental outcome of the President's way of chairing the group. He asked for judgements serially in a crucial meeting. This put enormous pressure on any person who demurred from the view of the opinion leaders. The group did not explore the one opposing (and prescient) position.

These meetings showed strong characteristics of autonomy through their ability to control the operation of components to preserve their manner of interaction as a group. The 'perfect failure' that resulted was a subsequent attribution of an observer who invalidated the meetings' value according to a wider frame of reference.

Janis' analysis gives an opportunity to assess the personal responsibility of President Kennedy in an historical event. Instead of laying responsibility for the foreign policy fiasco with him because 'the buck stops there' in a constitutional sense, we can be much more specific by identifying his style of chairing the crucial policy meeting as a reason. He could change his meeting style to good effect (and Janis suggests he did, based on later events), but he could never relive the Bay of Pigs. The 'drift' of events is never predictable or controllable, but we can pay attention to our way of engaging with others and trust that our ethical choices will deliver preferred outcomes.

A contrasting case: the anti-meeting?

In this case, a quite different view of knowledge generation guides meeting organisation. Humility about the results of deliberations opens them up to wider scrutiny.

The Cuban Missile Crisis [5] was a contrasting study for Janis (1982: Ch.6). Here a decision-making group with a similar membership to that engaged in the Bay of Pigs episode worked in a quite different way, with positive results. This time the group did not accept any of its decisions were solutions. They were simply the 'least objectionable' course of the moment. Nothing was final. Outsiders were called in and 'experts' were asked to join in, rather than contribute in their own specialised fields only. President Kennedy deliberately stayed away from some meetings of the Executive Committee to avoid it orienting to his expressed views. Ethical issues were much talked about and emotions ran high. One person involved called them 'repetitive, leaderless and a waste of time'.

The revised meeting norms created a kind of anti-meeting. It is not expected to come up with answers to problems or a consensus. Rather it is simply an event in a process of change for those joining it. Meeting coherence is sacrificed in pursuit of an enhanced capacity for all to operate in a wider decision-making process. An individual meeting is just one piece of the picture. 'Sense' and action emerge in a wider frame. Strong operating norms for the meetings brought this about; they were still self-organising. But they were a setting in a management process, rather than a tool to produce specific outcomes.

4. Implications and Conclusions

Suggestions for praxis

This section first gives a sequence of recommended actions for an organisational member to carry out in interactions with others. Then follows specific suggestions for meeting practice that may be adopted by a chairperson or facilitator.

• Act in love. This is an ethical choice from which all else follows.
   
  
• Bracket objectivity. We bring forth a world in language so we may co-ordinate our actions in the maintenance of our individual autopoietic organisation. Nothing exists outside language.
   
  
• Accept the views of others as equally legitimate. All logics that support our autopoiesis are valid. Each of us contributes a unique versa to the multiversa.
   
  
• Join with others in nurturing a creative interplay of your differences. Use reason in pursuit of a new domain of co-ordinating actions that is coherent according to a shared criterion of validation of your explanations.
   
  
• Act in love also in a wider system with your new domain of knowledge. It is still just one of the versa. So start again in a never-ending process of co-ordination of actions with others in a social process of validation.

Appealing to a wider frame

The chairperson should avoid being an auctioneer, taking bids for supremacy in establishing lineal cause-effect relationships. That muddies the acceptance of responsibility by the group for its actions. It distances them from the objects of conversation by making them external, independent objects. When she observes a clash of strong claims to objective knowledge, she can make an effort to put the claims 'in parentheses'. Qualifiers added to claims show they are not accepted as ultimate truth and encourage humility in the speaker. A validating method beyond the meeting boundary can be suggested.

Even the most dedicated rationalists will sometimes find themselves in dilemmas where logically sound positions draw opposing conclusions. Their resolution generally involves changing the nature of the question to embrace a broader context (Maturana & Varela, 1992: 135). The meeting frame is broader that that of any individual within it, so it can often resolve contradictions brought to it. With the help of the chair, the meeting in turn can accept that its own puzzles may dissolve in a wider context.

Acknowledging the intertwining of emotion and language

The chairperson may best end the meeting if its organisation has slipped away from her intent. Perhaps it has fallen into mutual negation of positions. Ending it acknowledges that harmful emotions are creating an unhappy meeting. Words do have an emotional impact. They trigger bodily changes in our physiology. We acknowledge them in our adjectives about speech: 'She was sharp-tongued' and so forth (Maturana, 1988: 48). Conflict in a meeting may be as emotionally real as in a physical fight. The chair can 'break up the fight'.

More constructively, the chair can aspire to draw upon group norms to nurture mutual acceptance. She may see past the rationality of an argument to the emotion (bodily pre-disposition) that invoked it. She can acknowledge the emotion by welcoming the contribution without committing to the argument. The group may appreciate the respect shown to one of their number.

Conclusions

Meetings are a universal occurrence, but their potential can be more fully realised. Rarely will one express the full creativity and humanity of those taking part. By adopting the value of mutual acceptance - a non-possessive warmth that we can call love - a group can nurture empathy and a creative interplay of diversity in its gatherings. Exploring the position of the other means suspending judgement upon its objective truth, even if temporarily, and entering into his feelings. An observer brackets objectivity by accepting that all views are equally legitimate. He still applies reason as an individual action to view and assess his patterns of thinking against a wider background provided by the meeting situation. Thoughts and plans then change by themselves.

An observer judges meeting outcomes according to his own frame of reference. He may not like the results of an autonomous process. But by allowing a meeting to operate autonomously, a manager opens the possibility of creating new knowledge valuable for his organisation. Autonomous parts of an organisation can be structurally coupled through a series of meetings co-ordinating their actions. Shared values bind all together.

Each of us is responsible for what we do, the system for what we achieve. A chairperson is responsible for his actions in maintaining the internal relations within a meeting, but cannot be held responsible for its performance with respect to a wider frame of reference. That performance is structure-determined by all its component parts.

Better outcomes will come from ways of interacting at meetings that respect the humanity of our fellows and acknowledge the humble capacity we have on our own to know how best to act for our organisation. We do best by attending to our self-mastery ahead of a fruitless quest for mastery over objective knowledge.

References

Atkinson, M.A., E.C. Cuff & J.R.E. Lee (1978). 'The Recommencement of a Meeting as a Member's Accomplishment'. In Studies in the Organization of Conversational Interaction, New York: Academic Press.

Bell, Arthur H. (1990). Mastering the Meeting Maze. Reading, MA: Addison-Wesley.

Bolman, Lee G. & Terrence E. Deal (1997). Reframing Organizations: Artistry, Choice, and Leadership. (2^nd ed.) San Francisco: Jossey-Bass.

Boyd, Richard, Philip Gasper & J.D. Trout (eds) (1991). The Philosophy of Science. Cambridge, Massachusetts: MIT Press.

Brocklesby, John & John Mingers (1999a). 'Theoretical Underdevelopment in Soft Systems Methodology - A Contribution Using the Ideas of Humberto Maturana'. Omega: The International Journal of Management Science. In review.

Brocklesby, John & John Mingers (1999b). 'The Cognitive Limits on Organisational Reframing: A Systems Perspective Based on the Theory of Autopoiesis'. Paper presented at the International System Dynamics Conference, Wellington, August 1999.

Carnes, William T. (1987). Effective Meetings for Busy People: Let's Decide It and Go Home. New York: IEEE Press.

Checkland, Peter (1981). Systems Thinking, Systems Practice. Chichester: John Wiley & Sons.

Efran, Jay S., Michael D. Lukens & Robert J. Lukens (1990). Language, Structure and Change: Frameworks of Meaning in Psychotherapy. New York: W.W. Norton & Co.

Ghose, Amitabha (1980). 'Cybernetics of Second Order - Some Problems'. In Frank Benseler, Peter M. Hejl & Wolfram K. Kock (eds), Autopoiesis, Communication and Society: The Theory of Autopoietic Systems in the Social Sciences, Frankfurt: Campus Verlag.

Hejl, Peter M. (1984). 'Towards a Theory of Social Systems: Self-Organisation and Self-Maintenance, Self-Reference and Syn-Reference'. In H. Ulrich and G.J.B. Probst (1984) (eds).

Janis, Irving L. (1982). Victims of Groupthink: A psychological study of foreign policy decisions and fiascos. (2^nd ed.) Boston: Houghton Mifflin. (First published 1972.)

Kenny, Vincent (1985). 'Life, the multiverse and everything; An introduction to the ideas of Humberto Maturana'. www.oikos.org/vinclife.htm. Published in A.L. Goudsmit (ed.) (1989), Self-Organisation in Psychotherapy, Springer-Verlag: Heidelberg.

Malik, F. & G.J.B Probst (1984). 'Evolutionary Management'. In H. Ulrich and G.J.B. Probst (1984) (eds).

Maturana, Humberto R. (1980a). 'Introduction' to Maturana & Varela (1980).

Maturana, Humberto R. (1980b). 'Man and Society'. In Frank Benseler, Peter M. Hejl & Wolfram K. Kock (eds.), Autopoiesis, Communication, and Society: The Theory of Autopoietic Systems in the Social Sciences, Frankfurt: Campus Verlag.

Maturana, Humberto R. (1987). 'The Biological Foundations of Self-Consciousness and the Physical Domain of Existence'. In E. Caianiello (ed.), Physics of Cognitive Processes, Singapore: World Scientific.

Maturana, Humberto R. (1988). 'Reality: The Search for Objectivity or the Quest for a Compelling Argument'. Irish Journal of Psychology. 9 (1): 25-82.

Maturana, Humberto R. & Francisco J. Varela (1972). 'Autopoiesis: The Organisation of the Living'. Re-printed in a translation from the original Spanish in Maturana & Varela (1980).

Maturana, Humberto R. & Francisco J. Varela (1980). Autopoiesis and Cognition: The Realization of the Living. Dordrecht, Holland: D. Reidel Publishing Co.

Maturana, Humberto R. & Francisco J. Varela (1992). The Tree of Knowledge: The Biological Roots of Human Understanding. (rev. ed.) Boston: Shambhala. (First published 1987.)

Mendez, Carment Luz, Fernando Coddou & Humberto R. Maturana (1988). 'The Bringing Forth of Pathology'. Irish Journal of Psychology. 9 (1): 144-172.

Mintzberg, Henry (1973). The Nature of Managerial Work. Englewood Cliffs, N.J.: Prentice-Hall.

Morgan, Gareth (1997). Images of Organization. (2^nd ed.) Thousand Oaks: Sage.

Schwartzman, Helen B. (1989). The Meeting: Gatherings in Organizations and Communities. New York: Plenum Press.

Shotter, John (1993). Cultural Politics of Everyday Life. Buckingham: Open University Press.

Ulrich, H. (1984). 'Management - A Misunderstood Societal Function'. In H. Ulrich & G.J.B. Probst (1984) (eds).

Ulrich, H. & G.J.B. Probst (eds) (1984). Self-Organisation and Management of Social Systems: Insights, Promises, Doubts, and Questions. Berlin: Springer-Verlag.

Umpleby, Stuart A. (1994). 'The Cybernetics of Conceptual Systems'. A paper prepared for the Institute for Advanced Studies, Vienna, Austria. www.gwu.edu/~umpleby/.

Varela, Francisco J. (1979). Principles of Biological Autonomy. New York: North Holland.

von Foerster, Heinz (1984). 'Principles of Self-Organisation - In a Socio-Managerial Context'. In H. Ulrich and G.J.B. Probst (1984) (eds).

von Glasersfeld, Ernst (1997). 'Distinguishing the Observer: An Attempt at Interpreting Maturana'. www.oikos.org/vonobserv.htm.

Whitaker, Randall (1998). 'Encyclopaedia Autopoietica'. DRAFT 1.0b (beta test prototype). www.informatik.umu.se/~rwhit/EAIntro.html#Download.

Winograd, Terry & Fernando Flores (1986). Understanding Computers and Cognition. Reading, MA: Addison-Wesley.

Notes

[1] This definition is firmly based upon that of autopoiesis, emphasising the relation of the two concepts. I prefer it to the more general statement that 'a system is autonomous if it can specify its own laws, what is proper to it' (Maturana & Varela, 1992: 48). Back

[2] This term is borrowed from Winograd & Flores, 1986: 147. Back

[3] This is a description drawn from the literature and used, for example, by Brocklesby & Mingers, 1999a. I interpret the manner in which an individual constructs his explanations as his epistemology. His ontological assumptions about the nature of reality influence his choice of explanatory epistemology. Back

[4] Cuban exiles backed by the CIA and the US navy and airforce invaded Cuba on 17 April, 1961. They were all killed or captured by the Cuban military within three days. The survivors were ransomed back to the US. Back

[5] In October 1962 the US identified Soviet nuclear missiles in Cuba and mounted a naval blockade. The Soviets withdrew the missiles after a tense period in which there was widespread fear of a nuclear exchange. Back

HTML transcription: Randy Whitaker, March 2002