Herbert Butterfield

Cambridge historian whose 1931 book The Whig Interpretation of History named "Whig history": writing the past "with reference to the present," sorting past actors into winners and losers by how well they anticipated where things ended up. The vault treats this as the general-case label for the pattern several AI-history corrections independently catch from inside their own domain — crediting Werbos's 1974 thesis with "spatial credit assignment" language it never uses, or reading a 1986 journal table of contents as a deliberate editorial pairing. First flagged as an entity candidate in the 2026-07-18 deep-lineage-myth capture, though his own claim-note and citing notes predate that capture by a week.

References

• claim-butterfield-1931-coined-whig-history-as-present-referenced-past
• observation-whig-history-splits-vault-myth-ledger-into-defensible-and-anachronistic
• claim-mahoney-warned-against-whig-history-of-computing
• moc-backpropagation-origins
• Captures: 2026-07-11-hop-whig-history-of-ai, 2026-07-18-does-the-deep-lineage-myth-displacing-contingent-truth (entity-candidate mention only)
• 2026-08-18: A direct primary read of Michael Mahoney's 2005 essay "The Histories of Computing(s)" found it never names Butterfield or uses the term "Whig" — the connection listed above is the vault's own bridge between two independently-arrived-at critiques, not a citation Mahoney himself made (claim-mahoney-2005-essay-omits-whig-and-butterfield)
• 2026-09-04: Wikipedia's summary of Butterfield's own critique, re-fetched and quoted directly, names hero/villain sorting ("good guys ... bad guys") as its own explicit target — distinct from, and broader than, the anachronistic-concept-importing reading the vault's myth-ledger split had been using as his side of the binary (claim-weinberg-and-butterfield-both-explicitly-reject-hero-villain-narratives)