I have in a previous post, considered a possible origin of the idea of self-organizing team in Agile methodologies, concluding that I had not found the philosophical source for the practice. Continuing research has taken me back through a number of sources as illustrated below, with the lighter boxes being sources I have not yet studied in detail.
The earliest resource I’ve considered to date is Morgan’s Images of Organization. The book is a study of the use of metaphor as an approach to describing organizational behavior. It reviews several metaphors, like organizations as mechanical mechanisms, as organisms, as cultures, and as human brains, reviews how those metaphors provide potentially useful insights into the properties of organizations, and then critiques the use of those metaphors. The concept of self-organization appears most centrally in the discussion of the brain metaphor.
As illustrated in the following diagram, Morgan developed a complex network of inferences, metaphors, and assumptions based on the initial premise that organizations are like human brains. Chaining the use of metaphors, Morgan reviews research modeling the brain as an information processing system (cybernetics), identifying brain properties that the model suggests. Additionally, he reviews the use of holograms as a model for additional properties of the brain. He also provides research examples leading to the description of the brain as “self-organizing,” defined in this context as being capable of reorganizing internal structure and function as they learn to meet the challenges posed by new demands. He next assumes that the properties identified in the cybernetics and holographic models are the supporting conditions for the ability to self-organize and uses that to conclude that it is possible to design a self-organizing organization by creating the conditions suggested by the models. Noting that some organizations viewed as innovative have the property of being self-organizing, he concludes that following the proposed design will enable innovation in organizations of that design.
Based on this network of reasoning, the conditions that support self-organization are:
- Redundancy in function – “each part is able to engage in a range of functions…”
- Requisite variety – internal diversity must match the variety and complexity of the environment
- Minimum critical specification – managers should primarily adopt a facilitating or orchestrating role, creating enabling conditions that allow an organization to find its own form. One should specify no more than is absolutely necessary for a particular activity to occur.
- Double-loop learning (learning to learn) – placing primary emphasis on being open to inquiry and self-criticism
This would seem to be a likely candidate as the basis for the Agile 11th Principle, “The best architectures, requirements, and designs emerge from self-organizing teams.” From this hypothesis we can see the possible origins of having Agile team members work in different roles (redundancy of function), of having all disciplines represented in Agile teams (requisite variety), of servant leaders (minimum critical specification), and the use of retrospectives (learning to learn). However, the derivation of holographic design of organizations is complex, highly speculative, with little supporting evidence, and makes assumptions based on correlation rather than causation. The key to the reasoning, that the properties identified in the cybernetic and holographic models are necessary and sufficient to create self-organization, is largely unsupported.
The previous post focused on Highsmith’s Adaptive Software Development, concluding that while he certainly researched and thought deeply about self-organization and emergence as elements of managing software projects, his definitions diverged from those in the prior literature and ultimately, he abandoned the term as a negative influence on Agile practice. Morgan’s Images of Organization seems to provide a much closer source to the definition of self-organizing and its use in describing Agile organizations. However, its design for realizing innovation by creating self-organizing organizations is built on speculation and unsupported reasoning. Perhaps in my next attempt I will find a better source for the theory behind Agile’s 11th principle.
