Site Logo

STPrism

16. Can you repeat this exercise for the authors but based on the trade-offs for observer structure variety uncertainty

5 min read

This analysis maps the intellectual landscape of the sources based on four critical trade-offs: Observer (Ontology vs. Epistemology), Structure (Mechanism vs. Constraint), Variety (Reduction vs. Absorption), and Uncertainty (Optimization vs. Resilience).

1. The Observer: Objective Reality vs. Constructed Model

This trade-off determines whether the “system” is an objective thing in the world or a mental construct of the investigator.

• Constructivist/Epistemic (High): Peter Checkland, Humberto Maturana, and Paul Cilliers argue that systems do not exist in the world; they are “intellectual devices” or “frames” used by an observer to make sense of a situation[1]. James Wilk takes this further, describing reality as a “symposium of points of view” defined by the questions we ask[4][5].

• Realist/Ontic (High): James Ladyman and Dave Snowden posit that complexity is an objective property of the world (e.g., the difference between a Ferrari and a rainforest)[6][7]. Fred Emery views the environment as an “objective reality” with a “causal texture” that can be directly known[8][9].

2. Structure: Decomposable Mechanism vs. Recurrent Constraint

This trade-off explores whether a system is defined by its independent parts or by the relationships and constraints that bind them.

• Mechanistic/Part-Centric: Herb Simon utilizes “near-decomposability,” breaking systems into hierarchical “boxes-within-boxes” to make them manageable for bounded minds[10][11]. TRIZ and Stafford Beer (in early VSM) often focus on functional organs and their specific inputs/outputs[12][13].

• Relational/Constraint-Centric: Alicia Juarrero and Ross Ashby define structure as the “presence of constraints”—patterns where parts co-determine each other’s futures[14][15]. Robert Pirsig views structure as “Static Quality,” a set of “labyrinthine fortifications” that preserve order against the flux of “Dynamic Quality”[16][17].

3. Variety: Complexity Reduction vs. Variety Absorption

Following Ashby’s Law (“Only variety can destroy variety”), authors differ on whether to simplify the system or boost the controller’s capacity[18][19].

• Reduction (Attenuation): James Wilk focuses on “filtering complexity” using logarithmic bisection to find a single idiosyncratic leverage point[20][21]. Herb Simon uses “satisficing” to ignore the infinite variety of the world in favor of a “good enough” model[22][23].

• Absorption (Amplification): Max Boisot and Dave Snowden advocate for “Distributed Intelligence,” using human sensor networks to absorb environmental variety that no central hierarchy could handle[24][25]. Stafford Beer emphasizes “Variety Engineering”—using autonomy and technology to amplify management’s response[26][27].

4. Uncertainty: Predictive Optimization vs. Adaptive Resilience

This trade-off addresses whether the goal of inquiry is to reach a stable “end-state” or to maintain “viability” in flux.

• Optimization/Goal-Seeking: Hard Systems Thinking and Operational Research traditionally seek the “which” action is best to reach a defined goal[28][29]. Isak Bukhman (TRIZ) seeks the “Ideal Final Result” through mathematical precision[30][31].

• Resilience/Muddling Through: Nassim Taleb argues for “Antifragility”—structuring systems to benefit from volatility rather than trying to predict it (Via Negativa)[32][33]. John Flach and Roger James advocate for “muddling through”—making small moves and monitoring feedback in “wicked” situations[34][35].

--------------------------------------------------------------------------------

Summary Table of Author Trade-offs

Author / GroupObserver FocusStructure LogicVariety StrategyUncertainty Stance
AshbySelection of variables[36]Organization as Constraint[15]Requisite Variety[19]Deterministic behavior[37]
ChecklandWeltanschauung (Subjective)[38]Human Activity Systems[39]Structured Debate[40]Learning Cycle[41]
SimonBounded Rationality[22]Near-decomposability[10]Heuristic Search[42]Satisficing[22]
JuarreroContext-dependent/Relational[43]Context-Dependent Constraints[44]Modulating Constraints[45]Path-dependence/History[46]
LadymanScale Relativity[47]Informational Real Patterns[48]Algorithmic Compression[49]Robustness/Probability[50]
SnowdenMulti-ontology (Cynefin)[51]Enabling Constraints[52]Human Sensor Networks[25]Safe-to-fail Probes[53]
TalebRelative surprise (Turkey)[54]Fragility vs. Antifragility[32]Skin in the Game[55]Via Negativa[33]
WilkQuestion-relativity[4]Flux-and-Constraint[56]Logarithmic Filtering[21]Minimalist Nudge[35]
Churchman”Eyes of another” (Inclusive)[57]Teleology (Purpose)[58]“Sweeping in” variables[59]Securing Improvement[60]

--------------------------------------------------------------------------------

Plot of Systems Thinking Trade-off Extremes

graph LR
    subgraph Order_Regime["ORDER REGIME (Mechanistic)"]
        A1[Objective System/Ontology]
        B1[Decomposable Hierarchy]
        C1[Variety Reduction/Simplification]
        D1[Goal Optimization/Prediction]
    end

    subgraph Complexity_Regime["COMPLEXITY REGIME (Systemic)"]
        A2[Subjective Model/Epistemology]
        B2[Recurrent Constraints/Holism]
        C2[Variety Absorption/Amplification]
        D2[Adaptive Resilience/Muddling]
    end

     Link to Transition Authors
    A2 ---|Checkland / Maturana / Cilliers| A2
    B2 ---|Ashby / Juarrero / Pirsig| B2
    C2 ---|Beer / Boisot / Snowden| C2
    D2 ---|Taleb / Flach / Revans| D2

    style Order_Regime fill:#f9f,stroke:#333,stroke-width:2px
    style Complexity_Regime fill:#bbf,stroke:#333,stroke-width:2px

The sources suggest that “Mastering the Muddle” requires the ability to switch between these poles. For example, Nassim Taleb suggests remaining “stupid” (reducing variety/complexity) regarding prediction while being “antifragile” (absorbing variety) regarding payoffs[61]. Similarly, Tim Allen advises treating the lower level as a simple mechanism (Order) to understand the focal level’s complexity[62].

Which of these trade-offs—such as the decision to filter complexity like Wilk or absorb it like Boisot—is most critical for your current project?

Graph View