---
id: "20260921-0220-does-any-study-dataset"
title: "Does any study, dataset, or corpus count substantiate Kwakkel's 'one in seven' colophon-frequency figure?"
type: "capture"
status: "promoted"
promoted_to: ["30-notes/claim-ommundsen-2025-colophon-bibliometric-study-measures-female-scribe-attribution-not-frequency.md","30-notes/claim-kiel-dahm-project-finds-one-in-five-colophon-frequency-in-german-manuscripts.md","30-notes/claim-purple-motes-15-percent-colophon-figure-restates-kwakkels-one-in-seven.md","40-entities/entity-bouveret-colophon-catalogue.md","40-entities/entity-erik-kwakkel.md (updated, not created)","40-entities/entity-colophon.md (updated, not created)","50-questions/question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus.md (progress log appended, not created)"]
not_promoted: ["Claim: Kwakkel's monograph and the Moreton (2014) citation trail remain untraceable this session (Books Before Print PDF still decodes to corrupted glyphs; Moreton's article blocked by academia.edu 403, a Project MUSE bot-challenge, and a TLS failure on the UIowa dissertation host) — not promoted as its own claim-note. It is a negative, process-level result (tooling access failure, not a fact about the world) with no independently verifiable content of its own; folded instead into a dated progress-log entry on question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus.md, where it belongs as a record of what was tried and why it didn't resolve the question.","Further leads (Moreton's ~25% nun-scribe figure not yet independently read; J.P. Gumbert's statistical-codicology lineage; comparative Armenian/Syriac/Hebrew colophon-frequency figures; the Bouveret catalogue's own front matter; Buringh's manuscript-loss-rate monograph) — left as leads, not promoted. None was independently read or newly verified this session beyond what the capture already states; each is a candidate next step for the open question.","Entity candidate J.P. Gumbert — declined again, consistent with the 2026-09-20 promotion's ruling (entity-promotion test's UNSURE branch). Still real and relevant to the field's methodology, but no claim promoted this session actually rests on him — this capture again finds 'no source... links him directly to this specific figure.' A mention, not a hub, until a claim depends on him.","Entity candidate Eltjo Buringh — declined. Real and foundational to the field's quantitative methodology (his loss-rate estimates underlie the Ommundsen paper's scaling), but not read directly this session and no promoted claim rests on his own figures specifically — a further lead, not yet a hub.","Entity candidates Åslaug Ommundsen, Aidan Keally Conti, Øystein Ariansen Haaland, Bodil Holst — declined as individual person hubs. Co-authors of a single paper this session cites once; promoting a hub per co-author is exactly the flood the entity spec warns against (the paper-authorship analog of one-hub-per-affiliation). Recorded fully as source_author on the claim-note instead.","Entity candidate Margit Dahm — declined. Leads a real, ongoing, directly-relevant project, but the claim promoted rests on the project's own page, not on her individually, and this is the project's first appearance in the vault (no recurrence yet). A mention, not a hub.","Entity candidate Melissa Moreton — declined. A real scholar whose 2014 article is the most promising still-open lead for an independent colophon-frequency count, but her work has not been directly read (access failed this session), so no claim rests on her yet — premature to promote.","Entity candidate Bénédictins du Bouveret colophon catalogue — promoted (see promoted_to): the dataset the Ommundsen claim's own source_quote rests on directly, distinguishing it from the declined candidates above, all of which the promoted claims only mention in passing.","Entity candidate statistical codicology (concept) — declined. Named repeatedly across this cluster's sessions as the disciplinary frame, but no promoted claim this session rests on it as a defined mechanism, and 'codicology' already covers the ground in entity-erik-kwakkel's connects_to; a distinct hub would be thin."]
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-09-21T00:00:00.000Z"
provenance: "batch run 2026-09-21"
derived_from: []
verifies: "question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus"
tags: ["colophon","erik-kwakkel","codicology","unverified-quant","verification","statistical-codicology"]
source_url: "https://www.nature.com/articles/s41599-025-04666-6"
source_author: "Åslaug Ommundsen, Aidan Keally Conti, Øystein Ariansen Haaland, Bodil Holst"
source_date: "2025-03-01T00:00:00.000Z"
source_title: "How many medieval and early modern manuscripts were copied by female scribes? A bibliometric analysis based on colophons"
source_venue: "Humanities and Social Sciences Communications (Springer Nature / Palgrave), vol. 12, article 346"
source_quote: "We use the Benedictine colophon catalogue with 23774 entries and find that 1.1% (dating from around 800 to 1626 CE) can be identified with certainty as having been copied by female scribes (95% confidence interval: 0.9% to 1.2%)."
source_tier: 1
source_sha: "c5cdd072bbb8fce6ff9e67968db809aa650cee393e9dba6246bb9c05157e7b45"
seek_code_commit: "22bdc2b"
---


This capture continues [[question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus]], raised after confirming that [[entity-erik-kwakkel|Erik Kwakkel]]'s own 2014 blog post ([[claim-kwakkel-2014-blog-states-one-in-seven-colophon-figure]]) states the "about one in seven" colophon figure without citing any study, dataset, or corpus count ([[claim-kwakkel-one-in-seven-colophon-figure-uncited-in-source-text]]). This session searched for an independent corpus-based count that could confirm or contradict the ratio itself, as distinct from the fact that Kwakkel said it.

**Short answer: no independent study, dataset, or corpus count substantiating Kwakkel's specific "one in seven" (~14%) figure was located.** Two genuine corpus-based colophon counts exist in the scholarly literature and were read directly this session — one measures a different quantity entirely (proportion of colophons attributable to female scribes, not proportion of manuscripts with a colophon at all), and the other measures colophon frequency but for a different manuscript population (German-language, not Kwakkel's Latin/Dutch material) and returns a different ratio (~1 in 5, not 1 in 7). Neither paper references Kwakkel, and Kwakkel's own writing does not reference either. The core question stays **[unverified -- could not confirm or deny after search]**.

## Claim: the 2025 Ommundsen et al. bibliometric colophon study — the strongest candidate corpus analysis found — measures a different quantity and does not reference Kwakkel's figure

verifies: question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus

Ommundsen, Conti, Haaland, and Holst's 2025 paper in *Humanities and Social Sciences Communications* performs a bibliometric analysis of the *Colophons de manuscrits occidentaux des origines au XVIe siècle* (the "Benedictine colophon catalogue," compiled by the Benedictines of Le Bouveret, Switzerland, 1965–1982, listing 23,774 colophons) to estimate what share of medieval manuscripts were copied by women. Read directly end to end: "We use the Benedictine colophon catalogue with 23774 entries and find that 1.1% (dating from around 800 to 1626 CE) can be identified with certainty as having been copied by female scribes (95% confidence interval: 0.9% to 1.2%)." This is a count of *which colophons* (within an existing catalogue of already-identified colophons) name a female scribe — it is not, and does not attempt to be, a count of *what fraction of all medieval manuscripts contain a colophon in the first place*. The paper's introduction, methodology, results, and 46-item reference list were checked directly and contain no mention of Erik Kwakkel, J.P. Gumbert, "one in seven," or any comparable colophon-frequency fraction. This is the paper the vault's prior session flagged as "a candidate model for what a traceable primary source for a colophon-frequency statistic looks like" — reading it directly confirms it is a rigorous, Tier 1, corpus-based statistic, but on a different research question than Kwakkel's claim, so it neither substantiates nor contradicts "one in seven."

- source_url: https://www.nature.com/articles/s41599-025-04666-6
- source_author: Åslaug Ommundsen, Aidan Keally Conti, Øystein Ariansen Haaland, Bodil Holst (University of Bergen)
- source_date: 2025-03-01 (received 2022-01-04, accepted 2025-02-27; journal issue *Humanities and Social Sciences Communications* 12:346)
- source_title: "How many medieval and early modern manuscripts were copied by female scribes? A bibliometric analysis based on colophons"
- source_venue: Humanities and Social Sciences Communications (Springer Nature / Palgrave)
- source_quote: "We use the Benedictine colophon catalogue with 23774 entries and find that 1.1% (dating from around 800 to 1626 CE) can be identified with certainty as having been copied by female scribes (95% confidence interval: 0.9% to 1.2%)."
- source_tier: 1 (read directly via extract_pdf against nature.com's own PDF, the venue of record; also archived as HTML)
- source_sha: c5cdd072bbb8fce6ff9e67968db809aa650cee393e9dba6246bb9c05157e7b45 (PDF); d890595edc259e03dc363e79e83f4cc02ea4028f3e5da46c642e7f382d1646b3 (HTML)

## Claim: an independent corpus count of colophon frequency does exist, but for a different manuscript population and a different ratio (~1 in 5, not 1 in 7)

verifies: question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus

A DFG-funded research project at Kiel University ("Colophons in German-language manuscripts of the Middle Ages," led by JProf. Dr. Margit Dahm, Institute of German Studies, running 2021–2024 with a follow-on phase through 2027) surveyed a defined manuscript population specifically to count colophon frequency, and states its own methodology and result on its own project page: "As part of the first project, around 11,000 manuscripts from the 12th-15th centuries from a total of 29 libraries, and thus a significant part of the German-language manuscript heritage, were analysed for their colophons... So far, around 2,200 manuscripts with a total of over 3,400 colophons have been identified and recorded in a database." 2,200 of 11,000 is approximately 20% (roughly 1 in 5), not Kwakkel's ~14% (1 in 7). This is a genuine, denominator-based, institutionally documented corpus count of colophon frequency — the kind of study that could in principle substantiate or refute a "one in seven"-type figure — but it covers German-language manuscripts specifically, a narrower and linguistically distinct population from the general Western Latin/Dutch material Kwakkel's career centers on, and neither the project's description nor Kwakkel's own writing draws any connection between the two figures.

- source_url: https://www.uni-kiel.de/en/phil/institutes/german-studies/research/research-practice/colophons-in-german-language-manuscripts-of-the-middle-ages
- source_author: Institute of German Studies, Kiel University (project lead: JProf. Dr. Margit Dahm)
- source_date: page undated (project described as running 10/2024–09/2027; first phase 09/2021–11/2024); accessed 2026-09-21
- source_title: "Colophons in German-language manuscripts of the Middle Ages" (research project page)
- source_venue: Kiel University (Christian-Albrechts-Universität zu Kiel), Institute of German Studies
- source_quote: "As part of the first project, around 11,000 manuscripts from the 12th-15th centuries from a total of 29 libraries, and thus a significant part of the German-language manuscript heritage, were analysed for their colophons. The data collection is based on the catalogues of the library locations, which are systematically checked for references to colophons/ scribal entries. So far, around 2,200 manuscripts with a total of over 3,400 colophons have been identified and recorded in a database."
- source_tier: 1 (institution's own description of its own ongoing research project, read directly via archive_page)
- source_sha: 26adab7ab2fbee0a135c6bc33cef6a645f8f3d3b971779e7b7ed2d3a71e81f0c

## Claim: the "15%" figure that circulates online as if independent is not corroboration — it is Kwakkel's own claim relayed a second time

verifies: question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus

A personal blog post ("purple motes," by Douglas Galbi, 2015-02-15) opens: "Colophons — concluding meta-text concerning the production of the text — exist in about 15% of medieval manuscripts." Read directly, that sentence's own footnote sources it: "According to Erik Kwakkel, an authority on medieval manuscripts, about one in seven medieval manuscripts include a colophon... Morton (2014) p. 65, n. 2; p. 43." (The reference list confirms this is a typo for Melissa Moreton, "Pious Voices: Nun-scribes and the Language of Colophons in Late Medieval and Renaissance Italy," *Essays in Medieval Studies* 29 (2014): 43–73.) The "15%" figure is not a second, independent measurement — it is one in seven expressed as a rounded percentage, attributed explicitly to Kwakkel by the same footnote that states it. This matters because a rounded restatement of the same unsourced figure can look, out of context, like independent confirmation; read against its own citation it is not. The footnote's second sentence — "About a quarter of manuscripts that nun-scribes wrote in fifteenth/sixteenth-century Italy included a colophon," also attributed to Moreton (2014) p. 43 — is a candidate for a genuinely separate, corpus-based figure, but it was not independently verified this session (see Further leads).

- source_url: https://www.purplemotes.net/2015/02/15/medieval-colophons-copying/
- source_author: Douglas Galbi
- source_date: 2015-02-15
- source_title: "medieval colophons, persons, and copying blessings"
- source_venue: purple motes (personal blog)
- source_quote: "Colophons — concluding meta-text concerning the production of the text — exist in about 15% of medieval manuscripts.[1]" / footnote 1: "According to Erik Kwakkel, an authority on medieval manuscripts, about one in seven medieval manuscripts include a colophon. In religious books, colophon use increased in the ninth century with Carolingian scribal practices. Colophons became more common in secular European manuscripts from the early fourteenth century. Morton (2014) p. 65, n. 2; p. 43. About a quarter of manuscripts that nun-scribes wrote in fifteenth/sixteenth-century Italy included a colophon. Id. p. 43."
- source_tier: 3 (named personal blog, uncontested bibliographic/historical claim about what the page itself states and cites — used here only to establish citation lineage, not as evidence for the underlying ratio)
- source_sha: 9e8c283c7738ced3c69b53e3e1146c1fd17096fa775dfe4058ab982e4827bb84

## Claim: Kwakkel's own monograph and the Moreton (2014) citation trail that might explain the figure's origin remain untraceable this session

verifies: question-verify-kwakkel-one-in-seven-colophon-figure-against-a-corpus

Two direct routes that could plausibly resolve where "one in seven" originates were attempted and both failed on tooling grounds, not on content grounds. (1) Kwakkel's 2018 monograph *Books Before Print* (Arc Humanities Press) — the only publicly accessible excerpt, the introduction PDF hosted by Árnastofnun (arnastofnun.is), was re-extracted this session (source_sha 58eff19b09e92bf8ccd654d445b1fd84d5866d7a4615ffedb4fe103dfa9b681e) and, as in the prior session, decodes to corrupted/unmapped glyphs rather than readable text — it cannot currently be searched for "colophon" or "seven." (2) Melissa Moreton's 2014 article "Pious Voices" (cited in the purple-motes footnote above, p. 65 n. 2, alongside the Kwakkel attribution) could not be read directly: academia.edu returned HTTP 403 to both extract_pdf and archive_page, Project MUSE's article page (muse.jhu.edu/article/550003) served a bot-verification challenge page instead of content, and the University of Iowa institutional repository hosting Moreton's related 2013 dissertation (ir.uiowa.edu/etd/6480) failed with a TLS certificate-verification error on both archive_page and WebFetch. None of these four failures showed any addressed-to-AI language, override language, or other adversarial signal (per the safety spec) — they read as ordinary bot-walls and a broken cross-domain TLS chain, not manipulation attempts — so no safety flag is raised, but the sourcing gap stands: it remains unknown whether Kwakkel's monograph or Moreton's article contains a citation this blog-post search did not surface.

## Further leads
- Melissa Moreton, "Pious Voices: Nun-scribes and the Language of Colophons in Late Medieval and Renaissance Italy," *Essays in Medieval Studies* 29 (2014): 43–73, and her related 2013 University of Iowa dissertation ("Scritto di bellissima lettera": Nuns' Book Production in Fifteenth- and Sixteenth-Century Italy, ir.uiowa.edu/etd/6480) — cited by purplemotes.net as the source of a "quarter of nun-scribe manuscripts had colophons" figure; not independently read this session (see failed-access claim above); the most promising still-open lead for a genuinely independent, corpus-based colophon-frequency count.
- J.P. Gumbert's "statistical codicology" (Leiden, Kwakkel's doctoral-training lineage, professor 1979–2001) remains unconnected to this specific figure in any source found across two sessions now; still the most likely disciplinary home for a rigorous version of this count if one exists.
- Comparative colophon-frequency figures by manuscript tradition, cited in the existing vault claim-note: ~55% Armenian, ~30% Syriac, ~3.4% Hebrew (single non-independently-verified academia.edu source, not re-checked this session) — reinforced qualitatively this session by an HMML (Hill Museum & Manuscript Library) lexicon entry stating, uncited to a number, that "unlike their Eastern counterparts, Western manuscripts only infrequently contain colophons until the later Middle Ages and Renaissance" (https://hmmlschool.org/lexicon/14493/, archived, source_sha 3d1d97bc802fb43eaa230c0587529a704e8ca304091308bde79e57cd15364945) — a Tier 3-4 institutional-reference source, acceptable for this uncontested comparative/definitional point but not for a number.
- The Bénédictins du Bouveret catalogue itself (*Colophons de manuscrits occidentaux des origines au XVIe siècle*, 6 vols., 1965–1982, 23,774 entries) is the single largest compiled colophon dataset in the literature and underlies both the 2025 Ommundsen paper and (per its own introduction) other colophon scholarship; it was not directly consulted this session (only read about via the Ommundsen paper's description) and its own front matter might state what population of manuscripts it draws its colophons from, which could bear on colophon-frequency questions generally.
- Eltjo Buringh, *Medieval Manuscript Production in the Latin West* (Brill, 2011) — supplies the manuscript-loss-rate estimates (~92.5% loss) the Ommundsen paper uses to scale colophon counts to total-manuscript estimates; a foundational quantitative-codicology reference for the field, not yet read directly.

## Entity candidates
- J.P. Gumbert — person — flagged first per the blind-spot rule: the founder of "statistical codicology" and Kwakkel's Leiden predecessor/doctoral-training lineage; the older, foundational figure any rigorous corpus-counting method in this field would be measured against, though no source found across two sessions now links him directly to this specific figure.
- Eltjo Buringh — person/concept — author of the manuscript-loss-rate estimates (*Medieval Manuscript Production in the Latin West*, Brill 2011) that the 2025 Ommundsen paper relies on to scale colophon counts into population-level estimates; a second foundational quantitative-codicology figure this literature measures itself against.
- Melissa Moreton — person — author of "Pious Voices" (2014) and a related dissertation on nun-scribe book production; cited alongside Kwakkel in the purplemotes.net footnote with what may be an independently-derived colophon-frequency figure (~25% for nun-scribed Italian manuscripts), not yet directly verified.
- Åslaug Ommundsen, Aidan Keally Conti, Øystein Ariansen Haaland, Bodil Holst — people — authors of the 2025 University of Bergen bibliometric colophon study; the clearest working example in the literature of what a rigorous, citable, corpus-based colophon statistic looks like, even though their specific study doesn't address colophon frequency.
- Margit Dahm — person — leads the Kiel University DFG-funded project systematically counting colophon frequency in German-language medieval manuscripts; the closest thing found to an active, ongoing research effort that could in principle be extended to test claims like Kwakkel's for other manuscript populations.
- Bénédictins du Bouveret colophon catalogue (*Colophons de manuscrits occidentaux des origines au XVIe siècle*) — concept/dataset — the largest compiled colophon dataset in Western medieval studies (23,774 entries, 1965–1982), underlying multiple quantitative colophon studies; worth its own note given how much of this literature traces back to it.
- statistical codicology — concept — the named subfield (per Gumbert and later Marilena Maniaci's edited volume *Trends in Statistical Codicology*) within which any answer to this whole question would methodologically sit.