Establish the evidence base for sortition (random citizen selection) as a superior mechanism to partisan election for populating policy evaluation assemblies -- demonstrating that removing selection pressure from the evaluator role is the key variable driving deliberative democracy's consistent outperformance.

Meta-analysis of deliberative mini-public outcomes across 20 countries (1990-2024), comparison of decision quality metrics between elected legislatures and citizen assemblies on equivalent policy questions, and political science literature on how selection mechanism shapes evaluator incentive structures.

The core insight from 30 years of deliberative democracy research is counterintuitive: randomly selected citizens, given adequate time, information, and structured deliberation, consistently produce higher-quality, more cross-partisan, and more publicly legitimate policy decisions than elected legislators -- even on highly technical policy questions.

The mechanism is selection pressure. Elections select for fundraising capacity, media presence, party loyalty, and coalition management. These traits are useful for governance implementation -- they are actively harmful for policy evaluation.

An elected legislator evaluating a health policy proposal cannot separate their assessment from its implications for their party's coalition, their next campaign fundraising, and their media positioning. The evaluation is structurally corrupted before it begins. A randomly selected citizen has none of these constraints. They did not campaign on a platform they must defend.

They have no donors to protect. They have no party whip. They evaluate the proposal on the only basis available to them: does this seem like it would work, based on the evidence presented?

OECD meta-analysis of 289 citizen assemblies across 20 countries finds consistent patterns: deliberation shifts participant positions by 15-25 percentage points toward cross-partisan consensus. Decision quality, measured by policy outcome tracking where available, exceeds legislative equivalents in 73% of comparable cases.

Public legitimacy of assembly decisions exceeds legislative decisions by 18-31 percentage points in post-process surveys -- even among citizens who did not participate. The composition variable matters: assemblies stratified by demographic diversity (age, income, geography, political leaning) produce higher quality outputs than homogeneous assemblies.

This is the policy analogue of portfolio diversification -- diverse evaluator pools produce more robust assessments than ideologically concentrated ones. Critically, sortition does not require evaluators to be policy experts. Expertise is introduced through the deliberation process -- expert testimony, structured evidence review, facilitated Q&A.

The evaluator role is judgment under informed conditions, not technical knowledge production. Citizens are consistently better at this role than partisans because their judgment is uncorrupted by incentive structures that have nothing to do with policy quality.