How paradigms contribute to different realities
People think political conflict is mostly about clashing moral principles. Usually it isn’t. In real disputes—especially the explosive ones—communities share the same basic ethics but disagree about what they believe actually happened. They inhabit different factual worlds because they divide the space of explanations differently.
@DrTimBarber and I worked out the math behind this three decades ago (citation at end). Each community has a paradigm: the set of distinctions it bothers to notice. These distinctions partition the hypothesis space into equivalence classes. Since priors are assigned uniformly to these classes, different paradigms create different starting plausibilities—even with the same data and the same moral principles.
One broad category to consider is when a paradigm notices hierarchy or power differential. The more structured actor simply has more internal relations to notice: leaders, captains, subordinates, chains of command. If the paradigm includes that relation—is the immediate boss of—then explanations involving the stronger actor get carved into many small equivalence classes, while explanations involving the weaker side collapse into one. The result is a large prior tilt against the more structured actor, even without ideological intent. (This is one reason why those with an anti-colonial bent are prone to believing “the U.S. or Israel did it!”)
A tiny example shows this. Suppose the Castle and the Villagers each have exactly seven individuals. Under a uniform prior that notices only “who did it,” each has 50% probability. But a paradigm that notices just one relation—is the immediate boss of—splits the Castle into three classes (leader, captains, privates) while the hierarchy-free villagers collapse into one. There are four classes total, each with one quarter of the prior. The Castle now carries 75% of the starting probability. The leader alone gets 1/4, a sevenfold increase relative to any villager’s 1/28.
No evidence changed. No morality changed. Only the conceptual carving changed. And this is how communities with shared principles and shared data can still end up in incompatible factual universes—because their paradigms differ in what distinctions they treat as relevant.
Changizi, M. A., & Barber, T. P. (1998). A paradigm-based solution to the riddle of induction. Synthese, 117, 419–484
See also chapter 3 of my first book, The Brain from 25000 Feet.