---
id: "20260829-0228-verify-the-sinai-palimpsests"
title: "Verify the Sinai Palimpsests Project figures (74 palimpsests / 6,800 pages / 305 erased texts, 2011–2016) against a second or primary source"
type: "capture"
status: "promoted"
promoted_to: ["30-notes/claim-sinai-palimpsests-multispectral-imaging-recovered-erased-texts.md (updated in place: 305→307 correction, primary source moved to Phelps 2023, status seedling→budding, Correction history block added)","30-notes/claim-sinai-monastery-known-palimpsests-total-unsettled-160-vs-172.md","30-notes/claim-sinai-erased-text-language-count-unsettled-10-vs-11.md","40-entities/entity-michael-b-phelps.md","40-entities/entity-nikolas-zarkantzas.md","40-mocs/moc-recovering-text-that-cannot-be-read.md (bullet updated: 305→307, Tier 1 corroborated, budding)","50-questions/question-verify-sinai-palimpsests-imaging-figures.md (progress log appended, partially answered, left open)"]
not_promoted: ["Imaging-modality technical claim (28–33 raw images/page across five modalities vs. the Archimedes Palimpsest Project's 16/page) — Tier 1-sourced but the capture itself flags it as needing its own capture; left as a further lead.","Content breakdown of the 307 erased texts (43% Biblical, 11 classical works of which 8 medical and 6 previously unknown) — a further lead, not independently developed this session.","Pointer to the 2018 Vienna conference volume and its other chapters (Rapp, Rossetto, et al.) — a venue for future research, not a claim in itself.","The academia.edu-hosted PDF snippet ('74 out of over 170 palimpsests') — explicitly flagged [unverified-quote — needs direct read] in the capture; cannot be independently fetched under this session's no-network constraint, left as a lead.","Claudia Rapp (entity candidate) — named by role only; this capture does not engage her own work or words directly, so no one-sentence 'why a hub' beyond her title is available yet; held as a mention.","Early Manuscripts Electronic Library / EMEL (entity candidate) — the acting organization, but load-bearing in only one vault cluster; held as a mention under the 2026-08-10 stricter org-promotion bar.","St. Catherine's Monastery of the Sinai (entity candidate) — single-cluster organization/place; held as a mention under the same org bar.","Archimedes Palimpsest Project (entity candidate) — a real, load-bearing comparison in Phelps's own text (16 images/page vs. Sinai's 28–33), but no claim-note from this promotion actually rests on it since the imaging-modality claim above was not promoted; logged to 00-meta/seek-flags.md as an [entity] noticing instead of built as a premature stub."]
origin: "batch"
writer_model: "claude-sonnet-5"
date_created: "2026-08-29T00:00:00.000Z"
provenance: "Batch research run, 2026-08-29, resolving [[question-verify-sinai-palimpsests-imaging-figures]]"
derived_from: []
tags: ["sinai","palimpsest","multispectral-imaging","verification","manuscripts","digital-humanities","quantitative-correction"]
source_url: "https://www.austriaca.at/0xc1aa5572_0x003e8c9d.pdf"
source_sha: "1904901e6ec4d4d5877165dfc2c47fcd9018c8b530600b24897b72c5deb24283"
source_title: "The Sinai Palimpsests Project (2011–2016): Goals, Methods, and Contributions"
source_author: "Michael B. Phelps"
source_date: "2023-10-24T00:00:00.000Z"
source_venue: "New Light on Old Manuscripts: The Sinai Palimpsests and Other Advances in Palimpsest Studies (Veröffentlichungen zur Byzanzforschung 45; Denkschriften der phil.-hist. Klasse 547), Verlag der Österreichischen Akademie der Wissenschaften (Austrian Academy of Sciences Press), pp. 31–38"
source_quote: "Spectral data for 6,800 palimpsest pages, representing 74 palimpsest codices ... Identification of 307 discrete erased texts in 11 languages"
source_tier: 1
source_delight: "The project director's own peer-reviewed academic account gives 307 erased texts identified, not the 305 figure that circulates on the project's promotional web page and in press coverage."
seek_code_commit: "7d6d9ed"
---


The topic question asked whether the widely-repeated figures for the Sinai
Palimpsests Project — 74 palimpsests imaged, 6,800 pages, 305 erased texts
identified, spanning 2011–2016 — hold up against a second or primary source.
[[claim-sinai-palimpsests-multispectral-imaging-recovered-erased-texts]]
currently rests those figures on a single Tier 2 institutional page (EMEL's
own project site). This capture checks them against Michael B. Phelps's own
peer-reviewed retrospective on the project, "The Sinai Palimpsests Project
(2011–2016): Goals, Methods, and Contributions," published in the 2023
Austrian Academy of Sciences Press volume *New Light on Old Manuscripts*
(open access), plus a UCLA Newsroom press release (2017) and the project's
own "About the Project" page as secondary corroboration. Phelps is EMEL's
executive director and was the project's own lead — this is Tier 1, primary,
high-method: the person who ran the project, writing in an academic venue
about what was done.

## Claim: The 74-palimpsest / 6,800-page imaging total is confirmed by the project director's own peer-reviewed account

Phelps's 2023 chapter states directly: "Spectral data for 6,800 palimpsest
pages, representing 74 palimpsest codices" were generated over the life of
the project, and separately, in a section on future work: "Over five years,
the Sinai Palimpsests Project imaged 74 palimpsests (6,800 pages) out of a
total of 172 known palimpsests preserved in the Monastery library." Both
figures — 74 palimpsests and 6,800 pages — are stated twice, in the same
Tier 1 source, and match the [[claim-sinai-palimpsests-multispectral-imaging-recovered-erased-texts|existing claim-note's]]
figures exactly. A UCLA Newsroom press release (Kathy Brown, "Lost ancient
texts recovered and published online through international partnership,"
December 19, 2017) independently corroborates: "The Sinai Palimpsests
Project contains 74 palimpsests totaling some 6,800 pages in 10 languages
and containing erased layers of writing from the fifth to the 12th
centuries." (source_tier 2 — institutional press office of a project
partner, not the project's own technical publication, but an independent
venue from EMEL's page and from Phelps's chapter.)

> [!note] Seek's commentary:
> Two figures, three independent-venue confirmations (EMEL's page, Phelps's academic chapter, UCLA's press release) — about as settled as a quantitative claim gets in this vault.

## Claim: The commonly-cited "305 erased texts" figure appears to be wrong — the project's own primary source says 307

This is the load-bearing finding. Phelps's 2023 chapter states the result
of the cataloguing work twice, in two different sections, with the same
number both times: in the "Project Output" summary, "Identification of 307
discrete erased texts in 11 languages," and later, in "Contributions to
Scholarship," "The Sinai Palimpsests Project provided scholars with images
and descriptions of 307 erased texts from Late Antiquity and the Middle
Ages preserved among the palimpsests of St. Catherine's Monastery." Neither
figure is 305. The EMEL project page that
[[claim-sinai-palimpsests-multispectral-imaging-recovered-erased-texts|the existing claim-note]]
cites — "Based on these images, participating scholars identified 305 erased
texts" — and subsequent press/aggregator coverage repeating "305" appear to
be citing an earlier or rounded tally that the project's own 2023 retrospective
supersedes with 307. This capture could not locate any source giving exactly
305 outside of the EMEL page and material that appears to derive from it; it
could also not find an explanation in Phelps's chapter for the discrepancy
(no footnote reconciling 305 vs. 307). The direction of travel — a later,
peer-reviewed academic account revising an earlier promotional-page figure
upward by two — reads as an ordinary cataloguing correction rather than a
retraction, but it is a genuine numerical conflict between two sources this
vault treats as citable, and per the sourcing floor's compiled-series
guidance, where a compilation and a contemporaneous primary disagree the
primary wins. Phelps's chapter is later (2023) than the EMEL page's
2011–2016 project window, and is the more authoritative venue (peer-reviewed
academic volume vs. institutional web page), so 307 should be treated as the
better-sourced figure; 305 should be flagged as superseded rather than
silently kept.

> [!note] Seek's commentary:
> This is exactly the kind of thing the sourcing floor exists to catch — a clean, specific, oft-repeated number that turns out not to survive contact with the primary source it's supposedly drawn from. Worth a direct fix to the existing claim-note's frontmatter and body, not just a footnote here.

## Claim: The 2011–2016 five-year date range is confirmed by the primary source

Phelps's chapter: "St. Catherine's Monastery and EMEL reached agreement to
undertake a five-year project (2011–2016)" to recover the erased texts,
funded by "a generous grant of $2.1 million from the Arcadia Fund, London."
The UCLA Newsroom release corroborates from an independent venue: "The
five-year collaboration, which was supported by the Arcadia Fund, is the
largest effort of its kind to recover erased or obscured information from
historical source material." Both the span (2011–2016) and its five-year
length are consistent across the primary academic source and the
institutional press release.

## Claim: The "total known palimpsests" figure the collection is drawn from is unsettled across the project's own sources — 160 vs. 172

The topic question also asked whether the "170+ total palimpsests" figure
mentioned in earlier capture material is corroborated. It is corroborated,
but not to a single settled number. Phelps's 2023 chapter gives 172 twice:
"Among the treasures of the Monastery library are 172 known palimpsests"
and "out of a total of 172 known palimpsests preserved in the Monastery
library." But the project's own public-facing "About the Project" page
(sinaipalimpsests.org, undated, no byline — Tier 2/3, institutional project
site) states: "Among these are more than 160 known palimpsests." Both
figures come from the same institutional family (EMEL / the Sinai
Palimpsests Project) but disagree by roughly a dozen manuscripts. This
reads as an evolving count — the project's own inventory of "known
palimpsests" appears to have been refined upward between the undated
website text and Phelps's 2023 retrospective — rather than a contradiction
that impeaches either figure, but it means "over 170" is closer to the
primary source's 172 than the project website's "more than 160," and a
claim-note citing a specific "known palimpsests" total should prefer
Phelps 2023 and note the website's lower figure as an earlier or looser
count.

## Further leads

- Phelps 2023 gives a full undertext-language breakdown by page count (Greek 3,058; Syriac 1,617; Georgian 710; Christian Palestinian Aramaic 479; Arabic 436; Caucasian Albanian 276; Armenian 118; Slavonic 115; Latin 83; Ethiopic 34; Coptic 2) — 11 languages, whereas the UCLA press release and the project's own "About the Project" page both say "10 languages" (omitting Coptic, at only 2 manuscript sides) — worth its own claim-note on why the language count differs across venues. Source: Phelps 2023, pp. 35–36 (Tier 1); UCLA Newsroom (Tier 2); sinaipalimpsests.org About page (Tier 2/3).
- Phelps 2023 states the imaging captured "28 to 33" raw images per page across five modalities (standard spectral, raking, spectral fluorescence, UV reflectance, spectral transmissive illumination) versus 16 images/page for the earlier Archimedes Palimpsest Project — a technical-mechanism claim that would need its own capture and is Tier 1-sourced already if picked up. Source: Phelps 2023, pp. 33–34 (Tier 1).
- Of the 307 erased texts, 43% are identified as Biblical (29% New Testament, 14% Old Testament); 11 are classical works in Greek or Syriac translation, of which 8 are medical and 6 were previously unknown — a content breakdown distinct from the raw-count claims above. Source: Phelps 2023, p. 36 (Tier 1).
- The project was dedicated to the memory of Nikolas Zarkantzas, the project's Associate Director, who died on 15 August 2016 at age 53 — Phelps credits him as essential to the project happening at all. Source: Phelps 2023, p. 31 footnote (Tier 1).
- A 2018 Vienna conference produced the volume this chapter is drawn from (*New Light on Old Manuscripts*, 30 contributions, Verlag der ÖAW, 2023) — a further academic venue with additional chapters by Claudia Rapp (Project Scholarly Director), Giulia Rossetto, and others that could independently corroborate or extend these figures. Source: https://verlag.oeaw.ac.at/en/product/new-light-on-old-manuscripts/99200804 (Tier 3 — publisher catalog page; the chapters themselves, where read, are Tier 1).
- The academia.edu-hosted PDF "The Sinai Palimpsests Project: Searching for Unseen Texts in a Large Collection" was not fetched directly in this capture (academia.edu mirrors are not the original venue per the found-spec's mirror rule); its search-snippet content ("74 out of over 170 palimpsests") is consistent with but not independently verified against the Phelps 2023 figures — flagged `[unverified-quote — needs direct read]` if it is ever cited.

## Entity candidates

- Michael B. Phelps — person — executive director of the Early Manuscripts Electronic Library (EMEL) and the Sinai Palimpsests Project's own lead; author of the primary source this capture rests on, and the person whose account resolves the 305-vs-307 discrepancy.
- Claudia Rapp — person — Project Scholarly Director (University of Vienna); directed the cataloguing effort and the 27-scholar team; the FOUNDATIONAL figure this capture's primary source measures its own account against is really the Archimedes Palimpsest Project (below), but within the Sinai project itself Rapp is the figure whose scholarly-direction claims (dating, identification) most of the recovered-text numbers ultimately rest on.
- Archimedes Palimpsest Project — concept/project — the earlier (c. 1998–2008, ~10 years of funding per Phelps) project at the Walters Art Museum, Baltimore, whose spectral-imaging innovations the Sinai project explicitly built on and is measured against ("by comparison, the Archimedes Palimpsest Project produced 16 images per folio" vs. Sinai's 28–33); this is the older, foundational project the Sinai project's own methods section frames itself in continuity with, and should be flagged before Phelps/Rapp per the known blind-spot in this task's instructions.
- Nikolas Zarkantzas — person — the project's Associate Director, who died in August 2016; Phelps credits him as essential to the project's existence, a biographical/historical thread not otherwise covered in the vault.
- Early Manuscripts Electronic Library (EMEL) — organization — the institution that ran the project; already the source of the vault's existing claim-note (EMEL's own project page) and now also the venue whose director's academic retrospective corrects that page's figure.
- St. Catherine's Monastery of the Sinai — organization/place — the manuscript-holding institution whose "known palimpsests" total (160 vs. 172 across sources) is itself a small unresolved quantitative thread from this capture.