For an organisational system to be suitable for both self-organisation and reflection, it needs to have a flat and decentralised structure, individuals need to be able to interact easily with others, and all parts of the system are necessary and sufficient for the operation of the whole but are not necessarily required to produce or replicate each other. The environment also needs to be conducive to reflective learning and cooperation between the parties. Seibert (1999) suggests the conditions that cultivate this environment are:
autonomy – ample freedom and discretion to structure one’s work as one sees fit;
feedback – information on the results of one’s actions – information is the raw material of reflection;
interactions with other people, entailing access to others – encounters with skilled and knowledgeable people; connection to others – at least one caring interpersonal relationship; and stimulation by others – encounters with people who provide new ideas and perspectives;
momentary solitude – periodic, brief occasions at or away from work to process new information alone.
Within an organisation that has these attributes, there will be individuals who, at any particular time, will become aware of a problem, a complex or unusual situation or perhaps an opportunity for improvement in their line of work. At this point, the individual conducts an analysis by examining the current situation, challenging their espoused theory of action, contemplating the ideal outcome and using their own experience and knowledge as well as knowledge from other sources (if it is reflection-on-action) to propose viable alternatives. The individual then tests each alternative to arrive at the optimal solution. The interaction of the individual with the information creates a spatio-temporal structure that has minimal interference from external sources. The iterative testing of alternatives within this structure creates a pattern from which cues are emitted and then perceived by other individuals in the organisation who are pursuing a similar purpose. The existence of patterns and emission of cues cause unintentional attraction between the individuals. These cues trigger an emergence of order – the creation of a cluster based on the common purpose. Within the cluster, links between individuals are strengthened as they work towards attaining a common meaning assigned to the problem or a modified purpose of this newly generated system.
This emergence of order initiates the second level. Clusters, based on various meanings are formed – each has its own purpose. Knowledge and information needs of each cluster may become part of the internal environment (stigmergy) or may form actual communication patterns (channels within the organisation) that are used to trigger the coordination and construction of activities targeting problem solving. Actors within the cluster work together, reflecting on each other’s suggestions and comparing them with the ideal, often coming to a satisfactory conclusion. However, there may not be sufficient information available within the cluster, necessitating further exploration – possibly from another cluster. One important factor to consider, and that is often missed in the self-organising literature, is that cooperation is a necessary condition for the existence of a social system and therefore for the presence of clusters. Cooperation facilitates the interaction among clusters and ensures information flow so that each cluster’s issues and concerns are known across the organisation. Under these conditions, clusters are better informed in a problem situation as their multiple interactions support further generation of information and knowledge, which is then embedded in the organisational environment.
Faced with a problem, the purpose of one particular cluster in the system will have an impact on other clusters, and vice versa (i.e. organisational networks are affected by each cluster’s reflection on information and ideal). These trigger self-organisation of clusters at a different level with a much more informed image of the problem-solving ideal. Because of the cooperation among clusters, reflection from different perspectives arises, allowing the emergence of new alternatives.
At this point it is worth noting that clusters constantly accommodate in order to target best alternatives. They can therefore structure themselves, generating new patterns, stimuli, actions, and eventually new clusters. The clusters come together because of problems at an organisational level. This initiates the third level. The cooperative system carried on from the second level strengthens the availability of information from past experience, facilitating the forecasting and reflection on business trends.
At this level, an organisation is in a position to reflect upon complex environmental changes and put itself in a position where blueprints and information systems templates are no longer required. Consistent reflective processes allow the system to monitor and be aware of changes in the environment and are therefore an important asset for the organisation as they assist further development of capabilities and help ensure market position.
As pointed out earlier, decentralisation is one of the key characteristics of self-organised systems, and one that enhances reflection. A decentralised organisation does not follow a hierarchical system. In fact, they focus on a more linear structure. Here, actors are empowered to act based on their own knowledge and perceptions, and no longer have to go through extended decision channels. They start to offer solutions as soon as they are faced with a problem and they are no longer inhibited or held back waiting for instructions from superiors. There is now an internal culture for constant information-knowledge development. Being able to self-reflect indicates that the organisation has placed itself in an advantaged position by establishing a circle of learning, auto-analysis and multiple ideals for problem solving.
With each level of problem solving, there are three opportunities for reflection (Raelin, 2001):
Anticipatory reflection occurs prior to the relevant experience (often at the planning stage).
Contemporaneous reflection occurs at the moment of the relevant experience (as with Schon’s reflection-in-action).
Retrospective reflection looks back at the relevant experience (as with reflection-on-action).
Mathiassen (2002) explains that, in systems development, the timing of reflection depends on whether the task is constructive, evolutionary or an intervention. When the task is constructive, requirements need to be taken into account to design the system required, and therefore reflection needs to be done before construction or action takes place. When dealing with evolutionary systems, the situation is already ongoing, and probably not stable, so the developer needs to reflect as the development takes place – it requires reflection-in-action. With intervention, the current situation and the problems associated with it need to be explored, alternatives suggested and implemented. This requires both reflection-in-action and reflection-on-action.
Dewey urges that reflection take place ad infinitum. He believes one can never be certain of the ‘truth’ and therefore should continually search in an attempt to invalidate an idea. Each time the reflection supports the belief, it is strengthened, but reflection should never stop.
Self-organisation should also be perceived as a continual process. The systems thinking literature (e.g. Checkland, 1981) highlights the importance of flexibility within a system. Reflection supports self-organising systems’ flexibility by not imposing time in the actual process of analysis and pattern formation. In a self-organising system, the flexibility of actors to adjust to the environment and anticipate problems and their solutions can only be possible by reflecting:
before the construction of actions takes place (i.e. prior problem recognition and awareness);
while the actions are taking place (i.e. during emergence of order); and
after actions have impacted the environment and system (i.e. learning from actions).
Actors should be able to benefit from reflecting at different times. A continuous reflecting process would guarantee the argumentation and negotiation required among actors for self-organisation to emerge. It supports the creativity needed for problem solving and helps the understanding of the underlying assumptions in complex environments. Reflection leaves actors in the position of reading the environment at any time and able to record and retrieve information and knowledge to address the ongoing requirements of the system. This type of flexibility, as described by Checkland (1981), results in the effectiveness, efficiency and efficacy of the system. Furthermore, it can support the reorganisation of systems while new patterns evolve. Success for a range of soft systems applications, issues such as complexity, variables and long-term system monitoring, are therefore addressed by the reflection rooted in self-organising systems.