Based on the provided sources, Michael Mainelli’s risk-case-study—specifically his “Environmental Consistency Confidence” (ECC) and “Predictive Key Risk Indicators” (PKRI)—can be interpreted through T.F.H. Allen’s complexity theory as a collision between modernist control and post-normal complexity.
Allen’s framework reinterprets Mainelli’s attempts to predict financial risk not as discovering objective facts, but as a dynamic interaction between an observer and a reflexive system (the “Other”).
1. Simple vs. Complex Systems: The Limits of Prediction
Mainelli posits that if incidents and losses can be predicted via correlation, the risk is manageable; if not, it is random or the data is wrong[1].
• Allen’s Interpretation: Allen would classify Mainelli’s “predictable” zone as a simple system (or at best, a complicated one). For Allen (citing Robert Rosen), a system is only “simple” if it can be fully captured by a formal model or simulation[2],[3].
• The “Lie” of the Model: Allen argues that models are “lies” or “simulacra” that freeze the world to make it tractable[4],[5]. When Mainelli uses statistics to predict risk, he is not measuring an independent reality; he is forcing a complex system into a “simple” equivalence class to achieve control[6]. Mainelli’s “unknown-unknowns” (Black Swans)[7] are what Allen defines as true complexity—the aspect of the system that has no simulable model and resides in the “Other,” beyond the observer’s definitions[8],[9].
2. The Observer and “The Other”: Interpreting Goodhart’s Law
In Mainelli’s case study of a European Investment Bank, the IT department changed the definition of “downtime” because they were being measured against it, skewing the risk data. Mainelli cites this as Goodhart’s Law (“when a measure becomes a target, it ceases to be a good measure”)[10].
• Allen’s Interpretation (Reflexivity): Allen views this not just as data corruption, but as a fundamental property of biosocial systems (systems containing living/social entities). Unlike billiard balls, the components of a social system (the IT staff) possess their own models of the observer (the risk managers)[11],[12].
• Anticipation: The IT department is an anticipatory system; it adjusted its behavior based on a predictive model of the risk managers’ actions[13]. For Allen, this is the “Other” reasserting itself. The risk model failed because the system it modeled (the bank) evolved its “essence” in response to the observation[14],[15].
3. High Gain vs. Low Gain: The Commodities Case
Mainelli describes a Global Commodities Firm where “high volume and high complexity days in… a high stress environment indicated relatively high forthcoming losses”[16].
• Allen’s Interpretation (Gain): Allen would interpret this trading environment as a High Gain system. High gain systems are driven by steep thermodynamic gradients (in this case, massive flows of money and information acting as energy) and are rate-dependent[17],[18].
• Instability: High gain systems are inherently unstable and prone to “accident” or collapse because they rely on the rate of flux rather than structural constraints[19]. Mainelli’s solution—to restrict complex trades during high stress—is an attempt to impose a Low Gain constraint (structure/rules) to dampen the thermodynamic instability of the high-speed trading floor[20],[21].
4. Epistemology: Science vs. Narrative
Mainelli asserts that “Science is facts” and advocates a “scientific management” approach based on statistical correlation[22],[23].
• Allen’s Interpretation (Post-Normal Science): Allen explicitly rejects this “modernist” view, arguing that science is not about objective truth but about “commensurate experience”—getting independent observers to agree on a story[24],[25].
• The Role of Narrative: Allen argues that when statistical models (like Mainelli’s correlations) encounter contradictions or “fat tails” (rare events where models fail), the analyst must switch to narrative[26],[27]. While Mainelli views the failure of statistics as a data problem[28], Allen views it as a necessity for post-normal science, where uncertainty is irreducible and stakeholders (traders, regulators, risk managers) must negotiate a shared story rather than calculating a single “correct” number[29],[30].
Summary Table
| Mainelli’s Concept | Allen’s Interpretation |
|---|---|
| Statistical Prediction | Turning a Complex system into a Simple one by freezing definitions[6]. |
| IT Dept changing data (Goodhart’s Law) | Reflexivity in biosocial systems; the observed system (the “Other”) adapting to the observer’s model[14]. |
| High Volume/Stress Losses | High Gain instability; thermodynamic flux overwhelming structural constraints[18]. |
| Risk “Management” | Adaptive Management; iterating between the formal model and the changing narrative of the system[31]. |
| Unknown-Unknowns | Emergence; the point where the “Other” breaks the observer’s model[32],[33]. |
References
[1] Risk-case-study.md [2] Allen 2017 - Mapping degrees of complexity, complicatedness, and emergent complexity.pdf [3] Allen 2003 - Distinguishing Ecological Engineering from Environmental Engineering.pdf [4] Allen - The loss of narrative - paradigms models mechanisms and analogy.pdf [5] Allen - Narratives and transdisciplines for a post-industrial world.pdf [6] Allen 2006 - The nature of ecological complexity.pdf [7] Risk-case-study.md [8] Allen 2006 - The Epistemological predicament associated with purposive qunatitative analysis.pdf [9] Allen 2006 - The nature of ecological complexity.pdf [10] Risk-case-study.md [11] Integrating_economic_gain_in_biosocial_s.pdf [12] Allen - Narratives and transdisciplines for a post-industrial world.pdf [13] [Book] Allen - Toward a Unified Ecology.pdf [14] Allen 2014 - Holons creaons genon environs in hoerarchy theory where we have gone.pdf [15] [Book] Allen - Toward a Unified Ecology.pdf [16] Risk-case-study.md [17] Allen - Narratives and transdisciplines for a post-industrial world.pdf [18] Integrating_economic_gain_in_biosocial_s.pdf [19] Allen 1999 - Supply side sustainability.pdf [20] Allen 2003 - Distinguishing Ecological Engineering from Environmental Engineering.pdf [21] Allen 2015 - The Observer in Complex Systems the Gateway Papers.pdf [22] Risk-case-study.md [23] Risk-case-study.md [24] Allen - Narratives and transdisciplines for a post-industrial world.pdf [25] Allen 2006 - The nature of ecological complexity.pdf [26] Allen - The loss of narrative - paradigms models mechanisms and analogy.pdf [27] Allen 2006 - The nature of ecological complexity.pdf [28] Risk-case-study.md [29] Kay 1999 - An ecosystems approach for sustainability addressing the challenge of complexity.pdf [30] [Book] Allen - Toward a Unified Ecology.pdf [31] [Book] Allen - Toward a Unified Ecology.pdf [32] Allen - The loss of narrative - paradigms models mechanisms and analogy.pdf [33] Allen 2006 - The Epistemological predicament associated with purposive qunatitative analysis.pdf