I want you to develop a framework which encompasses the breadth of the different views presented by the various authors whish is ready for use to bring a fresh perspective onto problematic situations
To bring a fresh, actionable perspective to problematic situations, we must first abandon the linear, reductionist “Machine Age” mindset that attempts to neatly solve problems by breaking them into isolated parts. Real-world organizational and social challenges are not tame “puzzles” with definitive true/false solutions; they are “wicked problems” or “messes”—complex systems of highly interacting, inseparable issues governed by human subjectivity and unpredictable dynamics.
Drawing on the vast synthesis of systems thinking and complexity science, I have developed the Polyocular Framework for Complex Messes. This framework integrates the ontological realities of complexity (how the world actually behaves) with the epistemological tools of systems thinking (how we structure our understanding of it).
The framework operates through four iterative phases: Framing, Mapping, Sensemaking, and Intervention.
Phase 1: Epistemological Framing (Defining the Boundaries)
Before analyzing data, you must establish how you are looking at the mess and where you draw the boundaries. Complexity often lies in the mind of the observer (cognitive overload) just as much as it does in the world.
- Acknowledge “Conscious Not-Knowing”: Assume a stance of epistemic humility. Because complex systems are non-linear and historically dependent, they are mathematically “incompressible”—meaning any model you create is flawed and leaves out vital information.
- Apply POSIWID: To understand what the system currently does, use Stafford Beer’s heuristic: The Purpose Of A System Is What It Does (POSIWID). Ignore the stated mission, rhetoric, or intentions of the organization, and look strictly at its actual emergent outputs.
- Conduct a Boundary Critique: Systems do not exist in nature; they are boundaries drawn by observers. Use C. West Churchman’s and Werner Ulrich’s emancipatory questions to interrogate the ethics of your boundary: Who is the actual beneficiary of this system? Who is the decision-maker? Who is the “witness” (the victim affected by the system but excluded from its design)?.
- Separate System from Environment: Define the “System” as the variables you can actually control, and the “Environment” as the variables you cannot control but which co-produce your outcomes.
Phase 2: Structural & Dynamic Mapping (Uncovering the Architecture)
Do not treat the mess as a sequence of isolated events. You must map the deep structure and constraints governing the behavior.
- Shift to Systemic Causation: Discard linear “A causes B” thinking. Look for “Systemic Causation” by mapping the feedback loops (reinforcing loops driving runaway growth, and balancing loops resisting change) and the inevitable delays between actions and consequences.
- Distinguish Detail vs. Dynamic Complexity: Ensure you are not fighting dynamic complexity (where cause and effect are distant in time and space) with detail complexity (building massive, rigid spreadsheets with thousands of variables).
- Look for Constraints, Not Causes: Using James Wilk and Alicia Juarrero’s cybernetic logic, stop asking “What caused this?” and instead ask: “Why is this happening rather than something else?” Find the invisible constraints (the environment, rules, or limits) that are making alternative behaviors impossible.
- Audit for Requisite Variety: Apply W. Ross Ashby’s Law of Requisite Variety (“only variety can absorb variety”). Does your management team possess enough internal flexibility and diverse responses to match the chaotic variety being generated by the external environment? If not, the system will fail.
Phase 3: Pluralistic Sensemaking (Handling Human Diversity)
Messes are sustained because different human actors hold entirely different interpretations of reality. You cannot force a single, “objective” consensus.
- Embrace the Symmetry of Ignorance: Recognize that in a wicked problem, there are no definitive experts. Knowledge is distributed among all stakeholders.
- Apply the T-O-P Lenses: Use Ian Mitroff’s multiple perspectives. Analyze the mess through the Technical lens (science, data, optimization), the Organizational lens (power, standard operating procedures, culture), and the Personal lens (intuition, leadership, individual morality).
- Model Weltanschauungen (Worldviews): Use Peter Checkland’s Soft Systems Methodology (SSM) to build different conceptual models based on different stakeholders’ worldviews. Structure these views using the CATWOE mnemonic (Customers, Actors, Transformation, Worldview, Owners, Environment) to make subjective biases logically explicit.
- Seek Accommodation, Not Consensus: Do not water down opinions to find a weak compromise. Cultivate “essential friction” and seek an accommodation—a proposed change that is logically desirable and culturally feasible, which all conflicting parties can “live with” to allow action to proceed.
Phase 4: Pragmatic Intervention (Acting in the Unknown)
In a complex system, you cannot fully understand the problem until you attempt to intervene in it. Planning must transition from trying to predict the unpredictable to building resilience.
- Probe, Sense, Respond: Using Dave Snowden’s Cynefin framework, recognize that in complex domains, “best practices” and top-down engineering fail. You must manage the present by launching multiple, parallel “safe-to-fail” experiments (probes). If a pattern succeeds, amplify it; if it fails, dampen it.
- Shift from “Fail-Safe” to “Antifragile”: Stop trying to build rigid systems optimized to withstand predicted shocks—they will shatter when hit by an unpredictable “Black Swan”. Build “safe-fail” systems (modularity, redundancy) that actively learn and benefit from volatility and error.
- Dissolve the Mess (Idealized Design): Instead of treating symptoms (resolving or solving), attempt to dissolve the mess using Russ Ackoff’s Idealized Design. Work backward: if the current organization were completely destroyed today, what ideal replacement would you design from scratch? This bypasses self-imposed historical constraints and allows you to redesign the environment so the problem can no longer exist.
- Action Learning (P + Q): Finally, implement Reg Revans’ ecological learning equation: (Learning must equal or exceed the rate of environmental Change). Abandon reliance entirely on “Programmed Knowledge” (P - past facts and expert best practices). Elevate “Questioning Insight” (Q), forcing leaders to take personal risk, act in the face of ignorance, and learn continuously from the feedback generated by the system.