CONCEPT
Near-Decomposability
Simon's 1962 principle that complex systems are buildable by bounded minds because their components interact strongly within subsystems and weakly between them — and that AI is dissolving this property faster than new architectures can replace it.
Near-decomposability is the architectural property that makes complex systems buildable by bounded
minds. Simon introduced it in '
The Architecture of Complexity' (1962) using the parable of two watchmakers: Hora assembles watches in a hierarchy of stable sub-assemblies; Tempus assembles watches as single sequences of a thousand steps. When interruptions occur, Hora loses only the current sub-unit; Tempus loses all prior progress. The parable illustrates why nearly decomposable structures — systems in which interactions within subsystems are strong and interactions
between subsystems are weak — are ubiquitous in biology, organizations, and software. Such structures allow each subsystem to be understood, assembled, and maintained by a bounded mind without requiring that mind to hold the entire system in view. AI weakens near-decomposability by enabling individuals to operate across subsystem boundaries — the backend engineer building frontend interfaces, the designer writing features, the product manager prototyping — producing both integration gains (fewer handoff losses) and cognitive costs (evaluation spread across