Communication and diversity secure system-wide performance
In our competitive view of the world, we often think that the most capable are those who are the most competitive, and accordingly that competition creates and secures efficiency. It may be that high performance is incorrectly attributed to competition and is more a result of diversity, self-organizing communication and non-competitive processes of collaboration.
Competitive processes lead to the handicapping of the higher-level system that these processes are part of. This is because competitive selection leads to exclusion: something is left outside. Leaving something out always means a reduction of diversity. The resulting less diverse system is efficient in the short term, but always at the expense of flexibility. Agility and complex problem solving require diversity. Everything goes fine if nothing changes and if there are only easy problems to take care of!
Self-organizing, non-competitive processes are about interdependent individuals and groups solving problems in a shared context. Interaction creates capability beyond what could ever be predicted just by looking at the performance of the individuals involved. The higher performance and robustness are emergent properties of interaction. They are not attributable to the parts of the system. Social networks provide problem-solving capability that results directly from the amount of communication and level of diversity of communication.
Most organizations would soon fail if all their employees thought alike or had little or no contact. There are two new challenges. The first is to understand the need for networking with views and values that are different. The second challenge is even bigger because of the mainstream reductionist thinking: our assumption has been that by understanding the parts in detail, we understand the whole. This is simply not possible! What happens in interaction between the parts is more important than the parts. The whole is the emergent pattern of that interaction, not the sum of the parts.
Diversity here means the degree of unique information in the network. If all contribute the same information, then diversity is low. If each agent contributes relevant, unique information that is not shared by others, then the diversity measure is high.
Networks with a wide spectrum of filtering information are resilient to noise. This facilitating effect of diversity is critical when dealing with difficult problems where false information can lead to expensive consequences. Higher system performance and robustness occur through the simple collaborative combination of the different experiences of individuals, even though each individual takes part in communicative interaction from their own limited perspective.
The importance of self-organization and diversity is unfortunately still greatly underestimated today, particularly in centralized, monoculture systems – like firms. One of the great societal promises of social media is that interaction in wide-area networks, with enough diversity, can solve problems beyond the awareness of the individuals involved.
Background: On serendipity
Thank you Stuart Kauffman, Sari Baldauf and Norman Johnson