Dave's reading notes

chaos-and-prediction

Properties1
tagsbanks, complexity, concept-epistemology, epistemology, prediction

Chaos and Prediction

Summary: In chaotic systems, running many simulations produces individually meaningful results that collectively add up to nothing — no pattern, no lesson, no obvious course.

Sources: Raw/Banks-The Hydrogen Sonata (via Clippings/Banks-The-Hydrogen-Sonata.md)

Source pages: The Hydrogen Sonata

Last updated: 2026-05-02

The Chaos Problem

From The Hydrogen Sonata by Iain M. Banks (source: Clippings/Banks-The-Hydrogen-Sonata.md):

“The Chaos Problem meant that in certain situations you could run as many simulations as you liked, and each would produce a meaningful result, but taken as a whole there would be no discernible pattern to them, and so no lesson to be drawn or obvious course laid out to pursue; it would all depend so exquisitely on exactly how you had chosen to tweak the initial conditions at the start of each run that, taken together, they would add up to nothing more.”

This is a fictional label for a real phenomenon: in systems sensitive to initial conditions, simulation volume does not reduce uncertainty. You can produce an arbitrarily large set of plausible futures without narrowing the range of what might actually happen.

What this challenges

The Chaos Problem cuts against several common assumptions:

  • More data = better prediction: Not if the sensitivity is in the initial conditions, which can never be measured exactly.
  • Ensemble models reduce uncertainty: They can characterise a distribution, but if the distribution has no useful structure, the ensemble is noise.
  • Rational planning under uncertainty: If the space of outcomes has no learnable pattern, decision-making must proceed without the anchor of expected value.

Relation to narrative bias

The mind’s instinct (narrative-bias) is to extract a pattern and build a story from it. The Chaos Problem is a category of situation where that instinct produces false confidence: each simulation looks like a story, but the stories contradict each other and none is more valid.

The Jackpot as applied Chaos Problem

Gibson’s Agency (gibson-agency) presents a close cousin: the Jackpot is not a single catastrophe but a decades-long accumulation of compounding crises (pandemics, climate, economic collapse, political decay) that no actor successfully predicted or prevented. No single feedback loop was decisive; the collapse emerged from their interaction. The Chaos Problem explains why: each crisis looked like a discrete problem with tractable interventions, but the system’s sensitivity to initial conditions meant that each intervention spawned new instabilities.

Graph View

Backlinks