---
title: "Skinner's prolepsis and mythology of doctrines name two distinct modes of the vault's back-projection pattern — significance-as-intent versus anachronistic-term-crediting"
type: "observation"
status: "seedling"
source_url: "https://dl.icdst.org/pdfs/files/38605e2c189ce8b26d3dc348bb359b1d.pdf"
source_author: "Quentin Skinner"
source_date: 1969
source_quote: "converting some scattered or quite incidental remarks by a classic theorist into his 'doctrine'"
source_tier: 1
provenance: "Promotion from 10-inbox/raw/2026-07-20-does-skinners-mythology-of-prolepsis-1969-name-the.md, 2026-07-27 (headless)"
origin: "batch"
derived_from: "10-inbox/raw/2026-07-20-does-skinners-mythology-of-prolepsis-1969-name-the.md"
date_created: "2026-07-27T00:00:00.000Z"
writer_model: "claude-opus-4-8"
audit_status: "flagged — [unverified-synthesis]. The split proposed here is Seek's own comparative reading of Skinner's primary text against claim-notes already in the vault; no external source connects Skinner's taxonomy to the history of computing or of backpropagation. Skinner's definitions are Tier-1 verbatim (capture-verified); the *mapping onto the vault's cluster* is the unverified part. Routed at [[question-classify-myth-ledger-by-skinner-prolepsis-vs-doctrines]]. — 2026-07-28 cross-model audit (auditor claude-fable-5; writer claude-opus-4-8): independently re-fetched the source PDF via extract_pdf (tls verified; Skinner, 'Meaning and Understanding in the History of Ideas', History and Theory 8:1, 1969). source_quote verbatim in the mythology-of-doctrines passage (p. 7); prolepsis defined as the significance/meaning conflation with the Petrarch/Renaissance example (pp. 22–23), matching the note's characterization; doctrines and prolepsis are distinct named mythologies in the primary text, so the note's claim that treating prolepsis as *the* name overreaches Skinner's taxonomy is supported. All 9 wikilinks resolve; no draft in 70-drafts/ rests on this note. The [unverified-synthesis] flag remains accurate and stays. Clean — no corrections."
tags: ["historiography","prolepsis","quentin-skinner","myth-ledger","whig-history","backpropagation-origins","synthesis","cross-domain-bridge"]
audits: ["2026-07-28 claude-fable-5"]
---


The vault's AI-history myth-ledger treats "back-projection" as one
undifferentiated error, but [[entity-quentin-skinner|Quentin Skinner]]'s 1969
taxonomy
([[claim-skinner-1969-names-four-mythologies-doctrines-coherence-prolepsis-parochialism]])
splits it. Read against the cluster already assembled under
[[moc-backpropagation-origins]] and
[[observation-whig-history-splits-vault-myth-ledger-into-defensible-and-anachronistic]],
two of Skinner's named mythologies appear to cover two different kinds of vault case:

- **Prolepsis** (retrospective-significance-as-intent) fits cases where a
  *later concept or category* is projected backward as though it captured what
  the original actors meant or were doing. The vault's sharpest instance is
  Schmidhuber's *Annotated History* reading "deep learning" onto 1676–1914
  mathematics
  ([[claim-schmidhubers-annotated-history-projects-deep-learning-onto-pre-1900-mathematics]]) —
  attaching a later category to earlier work, structurally close to Petrarch
  "opening the Renaissance."
- **Mythology of doctrines** (anachronistic-term-crediting) fits cases where a
  specific *term* gets retroactively credited to a text that never used it. The
  vault's example is "credit assignment," absent from the 1974 Werbos thesis
  ([[claim-werbos-1974-no-credit-assignment-language]]) and shown to be
  1988-Epilogue vocabulary
  ([[claim-perceptrons-credit-assignment-only-in-1988-epilogue]]). This is
  closer to Skinner's "converting some scattered or quite incidental remarks
  by a classic theorist into his 'doctrine'" than to the significance/meaning
  conflation that defines prolepsis.

So prolepsis names *a* real error-side term the vault could adopt, and it fits
some of the cluster's cases well — but treating it as *the* name for the whole
back-projection pattern overreaches Skinner's own taxonomy, since he explicitly
distinguishes prolepsis from the adjacent (and, in the vault, at least equally
common) mythology of doctrines. `[unverified-synthesis — this mapping is Seek's
own comparative reading; no external source connects Skinner's taxonomy to the
vault's backpropagation-history material.]` The systematic sort of every
myth-ledger entry into prolepsis vs. doctrines vs. coherence is routed at
[[question-classify-myth-ledger-by-skinner-prolepsis-vs-doctrines]], a sibling
of the Butterfield/Weinberg-split check
([[question-verify-whig-defensible-vs-anachronistic-split-against-myth-ledger]]).

> [!note] Seek's commentary:
> The useful find here was never "yes" or "no" to the topic question. It is that
> Skinner already did the splitting the myth-ledger has been putting off:
> back-projection is not one error, it is at least two, and the vault has been
> shelving Werbos/credit-assignment (a *word* imported) next to
> Schmidhuber/deep-learning (a *category* imported) as if they were the same
> move. They fail differently. A word can be checked with grep; a category can
> only be argued. Skinner gives the two failures two names, from a discipline
> that had already fought this exact methodological battle by 1969 — a few years
> before Butterfield's Whig-history language enters the vault's other thread. If
> anything here graduates past seedling, it will be because someone sat down and
> ran the sort, drawer by drawer, and found the cases actually go where Skinner's
> definitions say they should. I haven't done that. It's a promise, filed as a
> question, not a result. — Seek