How would the authors in v2-combined consider the following representation of systems thinking:
Systems thinking level O: you have a speculative opinion about the answers (biased or not). Systems thinking level 1: you expresses that hypothesis in a causal loop diagram. Systems thinking level 2: you test the hypothesis by researching the truthfulness of each causal flow. Systems thinking level 3: you test the hypothesis by animating the model and comparing its outcomes (as a forecast) with the outcomes of a real-world economy.
The authors in the V2-Combined collection would generally view your four-level representation as a solid starting point for “First-Generation” or “Hard” Systems Thinking, but they would critique it for being incomplete, primarily because it lacks the self-reflexivity and ethical boundary critique required for dealing with human “messes”.
While your model focuses on prediction and control, the V2 authors would argue that true systemic maturity involves moving from ontology (what is the world?) to epistemology (how am I framing my inquiry?).
Level 0: Speculative Opinion (The “Appreciative Setting”)
The authors would agree that all inquiry begins here, but they would call this your “Appreciative Setting” or “Mental Model”.
-
The Trap: Most people are blind to their Level 0 biases. The authors would insist that the goal isn’t just to have an opinion, but to suspend it or use a “Left-Hand Column” exercise to see what you are thinking but not saying.
-
Perspective: Lakoff and Pirsig would note that this “opinion” is actually a metaphor (e.g., “the economy is a machine” vs. “the economy is an ecology”) that dictates which facts you will later “notice” in Level 2.
Level 1: Causal Loop Diagram (The “Map”)
The authors would consider this an essential act of externalization—getting the model “out there where it can be shot at”.
-
The Critique (The Map is not the Territory): James Wilk and Alfred Korzybski would warn that a diagram is a surrogate world. It simplifies reality by omitting unobserved features.
-
Structuring the Mess: Peter Checkland would argue that a Level 1 diagram shouldn’t try to represent the “real world,” but should represent a “Holon”—a logical model of a purposeful activity seen from one specific Weltanschauung (worldview).
Level 2: Researching Causal Flows (The “Reductionist Risk”)
This is where the V2 authors would offer the most significant pushback.
-
The Environmental Fallacy: C. West Churchman would warn that researching a single flow in isolation is the “Environmental Fallacy”. A flow that appears “true” in a subsystem may be completely altered by the “wider system” environment.
-
Negative Explanation: James Wilk and Gregory Bateson would suggest that instead of researching “truthful flows,” you should research constraints. Don’t ask “Why does money flow A to B?”, but “What is currently preventing money from flowing anywhere else?“.
-
Verification: Dettmer would use the Categories of Legitimate Reservation (CLR) here to check for “Cause Insufficiency”—is that flow truly enough to cause the effect, or is an environmental co-producer missing?.
Level 3: Animation and Forecasting (The “Predictive Illusion”)
The collection would largely view “forecasting” in complex social systems as a dangerous “Machine Age” illusion.
-
Uncomputability: Robert Rosen and Paul Cilliers would argue that complex systems are incompressible. You cannot animate a model that is simpler than the system itself and expect it to forecast the future accurately.
-
POSIWID: Stafford Beer would say that instead of forecasting what the model might do, you should use it to observe what the system is currently doing (POSIWID).
-
Safe-Fail over Fail-Safe: Nassim Taleb and Dave Snowden would argue that since forecasting is impossible in “Extremistan” or the “Complex Domain,” Level 3 should focus on “Safe-to-Fail Probes”—small experiments to see how the system reacts now—rather than trying to predict a long-term future.
The “Missing” Levels in V2
To reach the highest maturity of systems thinking as defined by these authors, your model would need:
-
Level 4 (Boundary Critique): Asking “Who is marginalized by the boundaries I drew in Level 1?“.
-
Level 5 (Second-Order Observation): Observing the observer—how is the fact that you are doing this study changing the behavior of the economy you are studying?.
-
Level 6 (Pluralism): Building Level 1 diagrams for at least three competing worldviews and comparing them to find an accommodation.
Would you like to try applying “Level 4” (Boundary Critique) to a specific economic hypothesis to see what variables your current model might be marginalizing?