The sources offer diverse perspectives on the nature of complexity, ranging from it being an inherent physical property to a purely psychological sensation. Consequently, the advice on how to manage it varies from technical engineering to shifting mental metaphors.
1. Defining Complexity: Objective Property vs Observer Phenomenon
The sources are divided on whether complexity exists in the “territory” or merely in the “map” of the observer.
- Complexity as an Observer Phenomenon: Many sources argue that complexity is not an inherent property of a system but a result of human cognitive limits 1, 2. Ackoff and Wilk suggest that as familiarity with a situation increases, perceived complexity decreases; for Wilk, complexity is even described as a perceptual “bug” caused by using blurry abstractions instead of concrete details 1, 3, 4. Similarly, Hoverstadt and the Open University (OU) notes describe it as purely relative to the observer, who decides which differences are significant 5, 6.
- Complexity as a Quantifiable Property: Conversely, Stafford Beer and Triz provide technical definitions. Beer defines it as variety—the total number of possible states a system can exhibit 7, 8. Triz identifies it as a measurable Number of Complexity (NC) based on the expenditure of spatial and temporal resources required for a system to function 9.
- The Middle Ground (Systemic Property): Some sources, like Rosen, define complexity as an intrinsic property that is characterised by the necessity for multiple, non-equivalent descriptions 10. In this view, a system is complex if it cannot be fully captured by a single formal model or algorithm 10, 11.
2. Strategic Philosophies: Matching vs Reducing Complexity
A primary point of contrast in the sources is whether one should attempt to match the complexity of the environment or deliberately ignore most of it.
- Matching via Requisite Variety: Drawing on Ashby’s Law, several sources (Beer, Boisot, Hoverstadt, Jackson) argue that “only variety can absorb variety” 12-15. This suggests that for a manager to maintain control, their regulatory capacity must match the complexity of the system they manage 14, 15.
- Complexity Reduction: Luhmann provides a contrasting view, stating that systems must not attempt to match environmental complexity point-for-point 16. Instead, they must use “selectivity” to reduce environmental stimuli to only what is relevant to their own internal processes 16. Boisot similarly describes a “cognitive strategy” aimed at simplifying stimuli to economise on resources 13.
- Dissolution vs Solution: Ackoff offers a unique distinction: while “solving” a problem seeks the best outcome within a model, “dissolving” it involves redesigning the system so the problem is removed entirely 17.
3. Analytical Frameworks: Synthesis, Narrative, and Constraints
The sources provide various “tools for thought” to navigate complex “messes.”
- Synthesis over Analysis: Ackoff, Dettmer, and Mitroff all warn against traditional analysis (breaking problems into parts), arguing that in complex systems, the whole is not equal to the sum of its parts 17-19. They advocate for synthesis, which focuses on the functional interactions between components within a larger containing whole 18, 20.
- Narrative and Metaphor: Allen and Lakoff emphasize that because models are often internally inconsistent, narrative is the most robust tool for handling complexity as it can juxtapose contradictory components 21, 22. Lakoff further advises shifting between different metaphors to gain a comprehensive understanding, as any single model will inevitably hide some aspects of reality 23.
- The Weakest Link: Dettmer simplifies complexity by viewing a system as a chain; management should focus exclusively on the system constraint (the weakest link) where effort produces disproportionate benefits, rather than trying to improve everything at once 24.
4. Practical Action: Nudges, Trimming, and Infrastructure
Advice on physical or operational intervention ranges from “strategic nudges” to total system redesign.
- Minimalist Interventions: The Mastering the Muddle (TOG) source advises against “boiling the ocean” and instead suggests “strategic nudges”—finding the minimal changes necessary for maximum impact 25. This aligns with Jackson’s idea of “trojanmice,” or small, well-focused actions that produce significant improvements 15.
- Trimming and Modularisation: Triz offers a concrete engineering approach called “Trimming,” which involves eliminating problematic components and redistributing their functions to remaining parts to simplify the design 26. It also suggests modularity to reduce interdependencies between subsystems 27.
- Collective Infrastructure: Warfield argues that individuals cannot solve complexity alone due to cognitive limits 2. He recommends a “Situation Room” with large visual displays and a disciplined “Work Program of Complexity” to pool the beliefs of a group and reach a collective diagnosis 28, 29.
- The “Black Box” and “Video Descriptions”: Beer suggests treating exceedingly complex systems as “black boxes,” managing them by observing inputs and outputs rather than trying to understand internal details 30. In contrast, Wilk advises “climbing down the ladder of abstraction” to “video descriptions”—verifiable, concrete facts—to disperse the conceptual smokescreen of complex-sounding terminology 31.