Skip to content

Boole Learn — Propositional Learning Framework

Learn any subject fast by finding its foundational propositions, testing them, and mapping their logical structure. Three phases that build expert-like mental schemas before touching any material.

The Method

Phase 1: Foundational Propositions

Find the 5-7 load-bearing claims the entire field is built on. Not facts or definitions — falsifiable claims about reality.

Phase 2: Falsification Testing

For each proposition, find what would break it. Rate as: rock solid / contested / provisional / paradigm-dependent.

Phase 3: Logical Mapping

Map how propositions connect: dependency, independence, mutual support, tension. The generative tensions are where open questions live.

Provenance

Synthesis of three intellectual traditions:

  1. George Boole, The Laws of Thought (1854): "All sciences consist of general truths, but of those truths some only are primary and fundamental, others are secondary and derived." The insight that knowledge is hierarchical and reducible to foundational propositions.

  2. Karl Popper, The Logic of Scientific Discovery (1934): The falsification principle — a proposition you can't challenge isn't knowledge.

  3. David Ausubel, Educational Psychology (1968): "The most general ideas of a subject should be presented first." Empirically validated. The 5-7 count aligns with Sweller's cognitive load theory.

The viral "MIT student using Boole" story is a creative reinterpretation. Boole didn't prescribe a study method — he described the structure of thought. The falsification step is Popper, not Boole. But the combination works because Boole gives structure, Popper gives stress testing, and Ausubel confirms this matches how cognition builds understanding.

Where It Works Best

  • Well-established sciences (physics, chemistry, biology, CS)
  • Social sciences with strong theoretical frameworks
  • History and law (propositions become interpretive frameworks)
  • Software architecture (tested on Claude Code — see propositional map)

Where It Needs Adaptation

  • Skill-based domains (music, sports) — add a procedural knowledge layer
  • Aesthetic domains — propositions become principles, falsification becomes "when does this produce bad results?"
  • Pre-paradigmatic fields — be honest about instability
  • Emergent/complex systems — add "whole > sum of parts" caveat

Implementation

Available as a CC Sam skill: ~/dev/feral-cc-bots/sam/skills/boole-learn/SKILL.md

See Also