Autopoiesis

Consider a situation in which advanced technology is introduced—through official development assistance—to drill wells in regions struggling to secure drinking water. If the external support—funding, engineers, spare parts—eventually disappears, the well will become nothing more than a useless relic. For the well to function sustainably, the local community must develop its own autonomous “system” capable of operating and maintaining it. Without a mechanism within the community itself that continually reproduces the people responsible for securing water, the well cannot be sustained.

The logic underlying such a system is called autopoiesis. In Aristotle’s philosophy, poiēsis (productive activity) referred to the realization of a purpose within a thing through technical intervention. Against this background, the Chilean theoretical biologists Humberto Maturana and Francisco Varela proposed the concept of autopoiesis to describe the autonomous logic of living systems. They defined it as the organization of a living system as a machine, formulated as follows:

An autopoietic machine is a machine organized (defined as a unity) as a network of processes of production (transformation and destruction) of components that produces the components which: (i) through their interactions and transformations continuously regenerate and realize the network of processes that produced them.　

In this framework, the cell is the fundamental unit of life. A cell constitutes a network that continuously reproduces the very components that make up that network. Maturana and Varela argue that whether something qualifies as a living system depends on whether it possesses this kind of self-maintaining organization—one that sustains itself through the ongoing reproduction of its own operations.

Such a self-producing system is autonomous. An ordinary machine—an air conditioner, for example—may regulate room temperature through a thermostat and thereby achieve a kind of homeostasis. Yet it is produced heteronomously by external systems such as factories and machine tools, and it does not reproduce itself. For an air conditioner to be autopoietic, it would have to generate not only its functional processes but also the very components and structure that constitute it. In that case, homeostasis would no longer be merely an effect produced by the system; it would extend to the system’s very constitution, making the entire system autopoietic.

An autopoietic system also has no inputs or outputs in the conventional sense. Consider the visual-motor system. From a conventional perspective, light enters from the external world, is converted into electrical signals by the retina, transmitted to the brain, and then produces action as output. But if we understand the visual-motor system as a single autopoietic unity, it is simply a network that produces its own components—photoreceptor cells, neural pathways, muscles—and is, in this sense, operationally closed. Perception is not the “input” of light, but rather a perturbation of the system. What appears, from an external observer’s standpoint, as disturbance and internal change is, from within the system, the maintenance of its own organization in relation to its environment.

What distinguishes autopoiesis from classical poiēsis is that it does not locate purpose within the entity itself—hence its characterization as a “machine.” Classical poiēsis involved bringing an inherent purpose to completion and was governed by teleology. By contrast, a self-producing system simply operates as a network that continually reproduces itself through self-reference. There is no internal “purpose” guiding its activity. Teleology, in this view, is nothing more than a projection imposed by an external observer onto a system that operates without intrinsic goals.

If autopoiesis describes the logic by which living systems sustain themselves, then attempts to simulate artificial life—such as through self-referential computational processes—can be understood as examples of autopoietic design. Yet the concept becomes especially significant in its social dimension. Whenever the aim is to build autonomous and sustainable systems—in community development, international aid, or architectural planning—the logic of autopoiesis becomes indispensable. In such cases, design shifts toward shaping the conditions for the system’s continued operation, while entrusting its outcomes to self-producing processes.

On this point, the philosopher Hideo Kawamoto, in his interpretation of Maturana and Varela, describes the design of future societies as follows. In conventional social planning, one first imagines an ideal future and then works backward to determine present actions. This tends to produce a hierarchical relationship in which planners (the elite) conceive of the future while others merely execute it, reducing them to instruments. In contrast, an autopoietic approach to society establishes patterns of communication as an ethical code, such that the consistent enactment of this code gives rise to a particular form of society.

To offer a simple example: suppose we adopt “being kind to the person in front of you” as a basic ethical principle and construct a communicative network that continually reproduces this behavior. In that case, acts of kindness are not merely means to an end; through their repetition, they become the very processes that constitute the system itself. Any supposedly objective evaluation of whether a desirable future society has been achieved is an external perspective. From within the system, what matters is the lived intensity of each moment of kind interaction. Here, means and ends are inseparable, distinctions between designer and participant, or between planning and execution, dissolve. In fact, such distinctions were never truly there to begin with.

When we consider sustainability, endogenous development, intrinsic motivation, spontaneous creativity, or forms of ownership that foster self-efficacy—or when we reflect on the relationship between individual design actions and broader design ecosystems—the logic of autopoiesis becomes crucial. At the same time, self-reproducing systems can also appear in negative forms: capitalism, patriarchy, and cycles of harassment can all be understood as autopoietic systems.

→ See “Systemic Design”

(Toru Koga)

Related Classes

• Philosophy of Design

Reference

• Humberto R. Maturana and Francisco J. Varela, Autopoiesis and Cognition — The Realization of the Living, D. Reidel Publishing Company, 1980.