Over recent days, I've contributed to a couple of wide-ranging discussions on self-organization, in response blog posts by Bas Reus and Stephen Billing. This post and the next one summarize my current thoughts on the topic, from an informal coalitions perspective, to anchor them in this 'home' blog. I'm working on the basis of Weick's principle (in Sensemaking in Organizations) of "How do I know what I think until I hear [see] what I say?"
In the book, I describe organizations as dynamic networks of self-organizing conversations. Organizations exist because people can’t achieve alone what they need or want to achieve. So at the core of organization is the notion of people in interaction. And people use ‘talk’ (in the broadest sense of the word) as their medium of communication – both with others and with themselves (i.e. thinking). This means that self-organization isn’t a strategy designed and implemented by managers (as opposed, say, to command and control) it is a natural dynamic of all organizations.
Self-organization is not an optional way of managing
Often we hear management writers and commentators advocating the principles of self-organization as an alternative way of managing organizations. Typically, they position it as a more enlightened view of leadership – more empowered, in contrast to command and control. What they overlook is that the unalloyed ‘commanders and controllers’ are themselves active participants in the self-organizing processes of interaction within their organizations and beyond. Their inputs to these interactions may well be conditioned by their command-and-control mindset and behaviours. But the outcomes that arise will be no less subject to the principles and processes of emergence, self-organization and co-creation than if they were unconditional disciples of complexity thinking.
Outcomes are unavoidably co-created
I’ve mentioned co-creation here and a similar confusion has arisen around this related concept. Many books, consultants and so on talk of co-creation as a deliberate act of collaboration and joint problem solving, in which decisions are worked out jointly rather than being imposed from above. But all outcomes are unavoidably co-created, as people come together to perceive, interpret, evaluate and act upon the events, issues, edicts and so on that make up their world (more here). It is the self-organizing interplay of this myriad of local (i.e. one-to-one and small-group) conversations across the organization and beyond that leads to the global (e.g. organization-wide) outcomes. It is important to recognize that this does not imply a free-for-all. The dynamics of conversation and interaction are such that people are both enabled and constrained by their interactions with others.
Organizations are not "living systems"
The critical thing for me, then, is that it is the conversational interactions that are self-organizing. And, the dynamics of these are affected by such things as shifting power relations, the identities and self-interests of participants (which themselves emerge through interaction), the capacities of people for self-awareness and self-reflection, and so on. This also means that all interactions are necessarily political in nature. So it is misleading to talk of organizations as “living systems”, as is popular in many quarters today. None of the systems that are used as examples of organizational dynamics exhibit these uniquely human characteristics, which are played out through conversation.
... nor Complex Adaptive Systems
I would argue that it is also misleading to talk of organizations as Complex Adaptive Systems (CAS) and to import lessons from the laboratory-researched behaviour of these directly into the organizational world. For example, the latter rely on an outside observer/controller (the programmer), who sets down a small number of rules governing the local behaviour of agents in the CAS. Research shows that global (system-wide) patterns emerge as a result of the local, rule-based interaction of these agents. This has led advocates of the CAS model of organization to call for managers to set a few rules, within which it is presumed people will self-organize into organizationally beneficial ways of working. Unfortunately (or fortunately?), as Ralph Stacey says, people don’t tend to follow rules in the mechanistic way that this implies. Also, as suggested above, managers are active participants; that is, they are ‘on the pitch, playing’, they are not ‘sitting in the stands’, so to speak, objectively observing and controlling other people’s actions.
... nor systems of any kind
A CAS has, by definition, a system boundary. And I do not view organizations in these terms. The dynamic network of conversations is as wide and unbounded as the ‘most distant’ conversations that it spawns and that impact upon it. In other words, no organizational ‘thing’ exists outside these conversations other than more conversations. Clearly, a whole host of physical artefacts come into being over time as a result of some of these conversations (such as structures, strategies, processes, information systems, procedures, value statements and so on). But, in terms of organizational dynamics, these just provide an imprint of past conversations, which continue to impact upon the ongoing conversational life of ‘the organization’.
These artefacts often serve, by accident or design, to institutionalize certain power relationships and intended ways of acting. And they inevitably both enable and constrain ongoing interactions. But the ways in which they do so, and the outcomes that result from them, again depend wholly upon the nature and content of the conversations that they trigger. The physical artefacts may be very visible and tangible, but it is in the ongoing negotiation of their meaning that we find the essence of organization.
A self-organizing view of organizational culture
One final point worth mentioning in the context of self-organization is how I conceptualize what I call the “deep culture” of organization(more here). The more that people make sense of the world in particular ways, through their day-to-day conversations and interactions with others (both in their ‘immediate circle’ and beyond) the more likely it is that they will continue to make sense of things in similar ways going forward. That is, patterns of meaning emerge which tend to channel ongoing sense making down these same mental, emotional and behavioural ‘pathways’. And this deepens the metaphorical ‘channels of meaning’ still further.
Here again then, meaning cannot be handed down by ‘management’. It is determined in the moment of interaction whilst, at the same time, tending to be imperceptibly self-organized by the patterns of past sense making. This means that, whilst the possibility always exists for novelty to emerge, the dynamics of self-organization make it more likely that existing patterns of meaning will be reinforced.
The consequence of all of this for leadership is that managers can act with intention but with no certainty of outcome. If meaning, action and outcomes arise in the moment of people’s interaction, it means that it is on these day-to-day interactions that managers need to focus their attention.