The Problem of Induction and Causation in the Complex Domain
Taleb’s functional definition of complexity focuses on interdependence, reinforcing feedback, and nonlinear effects that can produce fat tails. He argues that these features limit Gaussian models and complicate causal analysis.
THE IDEA
What Is Complexity?
I will simplify here with a functional definition of complexity—among many more complete ones. A complex domain is characterized by the following: there is a great degree of interdependence between its elements, both temporal (a variable depends on its past changes), horizontal (variables depend on one another), and diagonal (variable A depends on the past history of variable B). As a result of this interdependence, mechanisms are subjected to positive, reinforcing feedback loops, which cause “fat tails.” That is, they prevent the working of the Central Limit Theorem that, as we saw in Chapter 15, establishes Mediocristan thin tails under summation and aggregation of elements and causes “convergence to the Gaussian.” In lay terms, moves are exacerbated over time instead of being dampened by counterbalancing forces. Finally, we have nonlinearities that accentuate the fat tails.
So, complexity implies Extremistan. (The opposite is not necessarily true.)
As a researcher, I have only focused on the Extremistan element of complexity theory, ignoring the other elements except as a backup for my considerations of unpredictability. But complexity has other consequences for the conventional analyses, and for causation.
When to use it
Risk management: check whether interdependence and feedback make extreme losses more likely than the chosen financial model allows.
Policy-making: When designing interventions for epidemics or climate change, account for nonlinear feedback and interdependence.
Scientific research: Avoid overreliance on linear causality and average effects when studying ecosystems, economies, or social networks.
Engineering: In designing resilient systems, incorporate safeguards against cascading failures driven by feedback loops.
FROM THEORY TO EVERYDAY LIFE
In practice
Stock market crashes
Small initial shocks can cascade through interconnected financial instruments, with past price changes influencing future ones (temporal interdependence) and correlations between assets creating horizontal interdependence, leading to extreme crashes (fat tails) that defy Gaussian predictions.
Epidemic spread
Disease transmission depends on current infections, contact networks, and the timing of interventions. These relationships can produce rapid growth and make simple extrapolations unreliable. Growth alone does not establish that outbreak sizes follow a fat-tailed distribution.
A MOMENT TO REFLECT
Put it to work.
Identify a system in your life or work that exhibits temporal or horizontal interdependence. How might it produce fat tails?
How does the presence of positive feedback loops challenge the use of standard statistical models based on the Central Limit Theorem?
In what ways does complexity complicate causal inference compared to domains with thin-tailed distributions?
Often confused with
CONTINUE THE CONVERSATION
Source texts
- The Black Swan The Impact of the Highly Improbable With a new section On Robustness and Fragility (Nassim Nicholas Taleb)