Look for structural parallels across inputs during synthesis, not topical overlaps — topic matching produces aggregation, structural matching produces genuine synthesis

Why This Is a Rule

There's a critical distinction between aggregation (collecting related items under a shared topic) and synthesis (creating new understanding by identifying shared deep structures). Aggregation asks: "What do these inputs have in common topically?" Answer: "They're all about leadership." That's not a new idea — it's a category. Synthesis asks: "What structural pattern appears across these inputs?" Answer: "Leadership, ecosystem management, and software architecture all use the same feedback-loop dynamic for stability." That's a new insight — a transferable pattern that wasn't visible from any single input alone.

The search for structural parallels rather than topical overlaps produces synthesis because structure transfers across domains while topic does not. "Negative feedback loops produce stability" applies to thermostats, ecosystems, leadership, and software systems. This cross-domain structural pattern is the synthesis — it reveals a deep mechanism that explains surface-level phenomena across multiple fields.

The default search pattern during reading and note review is topical: "This note is about productivity, and that note is also about productivity — they go together." This produces aggregation (a bigger pile of productivity notes) but no synthesis (no new understanding). Redirecting attention to structure ("This productivity note describes a bottleneck dynamic, and this ecology note describes the same bottleneck dynamic — what's the general pattern?") produces genuine synthesis.

When This Fires

• During synthesis sessions when laying out multiple notes (Lay out multiple notes in parallel visual access for synthesis — sequential reading prevents the simultaneous comparison that synthesis requires) and looking for connections
• When connections between notes feel forced or obvious — likely topical rather than structural
• When you want to produce genuinely new insights rather than reorganized summaries of existing ideas
• Complements Lay out multiple notes in parallel visual access for synthesis — sequential reading prevents the simultaneous comparison that synthesis requires (parallel layout) with the cognitive search strategy for what to look for

Common Failure Mode

Topical clustering disguised as synthesis: "I synthesized my notes on habits, productivity, and time management." If the "synthesis" is just a summary of what all three topics say, it's aggregation. Genuine synthesis would identify a structural pattern that runs through all three: "All three domains share the principle that environmental design outperforms willpower-based approaches — the mechanism is reducing decision load at the point of action."

The Protocol

(1) When scanning notes in parallel (Lay out multiple notes in parallel visual access for synthesis — sequential reading prevents the simultaneous comparison that synthesis requires), consciously shift your search pattern from "What topics overlap?" to "What structures, dynamics, or mechanisms appear across multiple notes?" (2) Structural parallels to look for: Shared mechanisms (same causal process in different domains), Shared dynamics (same pattern of change over time), Shared constraints (same limiting factor appearing in different contexts), Shared failure modes (same type of breakdown across different systems). (3) When you spot a structural parallel, write it as a cross-domain principle: "Both X and Y exhibit [pattern] because [shared underlying mechanism]." (4) Test the synthesis: does it produce an insight that wasn't available from any single input alone? If yes, it's genuine synthesis. If it's just a summary of one input applied to another's topic, it's aggregation. (5) The most valuable syntheses connect notes from different domains — that's where structural parallels are most surprising and most informative.