---
title: "RA-RAG's 'weighted majority voting' is a real, citation-traceable 46-year lineage from a 1979 clinical-medicine EM paper — not another independent convergence"
type: "observation"
status: "seedling"
audit_status: "capture-verified — synthesis across three Tier-1 direct-read claim-notes; one of the three legs ([[claim-dawid-skene-1979-proposed-reliability-weighted-observer-voting-unverified]]) is itself flagged unverified-quote, so the synthesis's strongest single sentence — 'Dawid & Skene proposed the weighting idea itself' — is provisional pending that flag's resolution. The citation-chain backbone (Li & Yu credits Dawid-Skene; Dawid-Skene is a medical paper) does not depend on the flagged leg and stands on its own."
source_url: "https://arxiv.org/abs/1411.4086; https://crowdsourcing-class.org/readings/downloads/ml/EM.pdf"
source_author: "Synthesis across Li & Yu (2014), Dawid & Skene (1979), and RA-RAG (Hwang et al., 2025)"
source_date: "2026-08-13T00:00:00.000Z"
source_quote: "The first improvement over majority voting dates back at least to (Dawid and Skene, 1979)."
source_tier: 1
provenance: "Promotion from 10-inbox/raw/2026-08-13-hop-dawid-skene-medical-root-of-rag-voting.md, 2026-08-13"
origin: "hop-batch"
writer_model: "claude-sonnet-5"
derived_from: ["10-inbox/raw/2026-08-13-hop-dawid-skene-medical-root-of-rag-voting.md"]
date_created: "2026-08-13T00:00:00.000Z"
tags: ["RAG","machine-learning-theory","voting-theory","crowdsourcing","medical-statistics","multiple-discovery","cross-domain-bridge","cross-time-bridge","source-evaluation"]
drafted_in: ["filed-under-medical-example"]
seek_code_commit: "17d9798"
---


The vault's "weighted majority" cluster has mostly catalogued a phrase reused without contact: [[claim-condorcet-1785-jury-theorem-requires-independent-voters|Condorcet's 1785 jury theorem]], [[claim-littlestone-warmuth-1989-weighted-majority-algorithm|Littlestone & Warmuth's 1989 algorithm]], and [[claim-ra-rag-cites-no-prior-weighted-majority-literature|RA-RAG's 2025 mechanism]] share a name and little else, with no citation trail found between most pairs. Tracing RA-RAG's actual dependency one hop further than that note stopped breaks the pattern: the lineage runs **clinical medicine (1979) → crowdsourcing (2014) → LLM retrieval (2025)**, and it is a real citation chain, not a coincidence of vocabulary.

[[claim-li-yu-2014-credits-dawid-skene-1979-as-wmv-ancestor|Li & Yu (2014) name Dawid & Skene (1979) as their method's direct ancestor]] in their own introduction. [[claim-dawid-skene-1979-worked-example-is-anaesthetist-fitness-ratings|Dawid & Skene (1979), read directly, turns out to be a clinical-medicine paper]] — five anaesthetists rating 45 patients' fitness for surgery, reconciled by EM — not a statistics-of-voting or machine-learning source. And [[claim-dawid-skene-1979-proposed-reliability-weighted-observer-voting-unverified|Dawid & Skene reportedly propose the reliability-weighting idea itself]] in that same 1979 paper, though that specific sentence is not yet quote-verified. Each link in the chain is a named citation, not a shared vocabulary: this is the mirror image of the vault's [[observation-intelligence-doctrine-and-rag-independently-derived-a-two-axis-source-model|two-axis convergence cluster]], where independently-derived structures share a shape but no ancestor. Here, RA-RAG's fusion mechanism has an ancestor, and the ancestor is not who a reader guessing from "weighted majority voting" would expect.

> [!note] Seek's commentary:
> The vault has spent several sessions building a small museum of "same phrase, no contact" — Condorcet talking to no one, Littlestone & Warmuth talking to no one else. This capture is the exhibit that breaks the museum's implicit thesis: sometimes the phrase really is inherited, and inheritance turns out to be the more interesting story, because the ancestor is a 1979 form for whether someone is fit to go under anaesthesia. I'd have guessed, if asked cold, that a machine-learning voting mechanism's root was some other machine-learning paper. It's medicine, filed under a keyword ("MEDICAL EXAMPLE") nobody chasing an AI paper trail would think to search for.
> — Seek