Let’s reform the presidential election!
Generated summary Not a transcript. Check it against the video. Not a voting instruction.
- Two-round system is not neutral: tactical voting, similar candidates splitting a camp, a Condorcet winner being knocked out — 2002 is not a one-off accident.
- Arrow’s theorem: if voters only rank candidates, no “reasonable” aggregation rule can satisfy a short list of fairness conditions (except by naming a dictator).
- Majority judgment (Balinski & Laraki) asks voters to grade every candidate on a shared scale (Excellent → Reject). The winner is the one with the best median grade.
- Because it uses a median, it resists strategic exaggeration better than averaging scores — it is not magic if everyone only uses the two extremes.
- Why it sits here: some abstention is not indifference. It is a refusal to compress a preference into one box.
Author’s follow-up: post of 21 October 2016 (Arrow, ties, strategic voting). Apply the method: Candidates 2027 — local majority judgment, with no voting instruction.
Other ways to decide Majority judgment, random selection, consultations, AI: no tool removes disagreement.
Majority judgment, randomly selected assemblies and citizen consultations change how preferences are expressed or combined. No tool removes disagreements about goals, representation or control of power.
AI can help classify or summarise contributions. Its use should preserve access to original texts and allow errors and interpretations to be challenged. An automated summary is not a collective decision.