---
slug: the-one-that-broke-off
passed: 2026-10-14
by: the assay line (research/assay-line/cited-1014b/), Codex fixer directed by claude-assaying-boyle-1014; source record by Codex, checked by Codex
checked: 152
confirmed: 123
wrong: 29
unverifiable: 0
observed: 0
stale: 0
---

This was the 2026-10-14 directed repair pass for the-one-that-broke-off at source pin f5e5f84f34, using research/assay-line/cited-1014b/the-one-that-broke-off/triage.json. The fixer re-read the evidence for all 31 triage items, corrected 20 WRONG and nine MINOR findings (X29 was already corrected in the working tree), and rejected X26 and X27 with evidence. The 121 CONFIRMED lines below retain the checker's source re-reads from N2-codex.md, explicitly attributed as such; they are checks of the pinned statements, with context retained where only a clause was confirmed. The wrong count includes the nine MINOR lines. Eight dated correction lines address the eight distinct WRONG errors and are mirrored in the layer markdown; MINOR fixes have no new page correction lines. The earlier gear-inventory correction was preserved. Generated data and the Corrections note were rebuilt with this layer's build script.

## Claims

- MINOR | X1: the most famous mathematical object to survive from the ancient world | https://jmhammond.github.io/history/Robson2002.pdf | Robson 2002, p. 105 describes one of the most famous ancient mathematical artefacts, without ranking a unique first.
- WRONG | X2: the single digit that would settle the argument broke off the left edge | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X3: What is missing, on the broken left edge, is whether that fraction was preceded by a 1. | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X4: The clay that would have told us apart is gone. | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X5: fourteen hundred correct digits | git show f5e5f84f34:research/plimpton-322/page.json | Recounted the pinned page.json rawDigits, falling back to digits: columns I to III have 137 sexagesimal places including restorations, not 1400 correct surviving digits. Adding row labels and leading ones gives 167 positions; wedge counts are 919 or 1000 with those additions.
- MINOR | X6: Two of them are the same slip twice. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Re-read Robson 2001, p. 175, Table 2. Row 2 is classified as copying and row 8 as arithmetical; their similar numerical effect does not make the mechanisms identical.
- WRONG | X7: a mistake you can only make in a place-value system where the places are written side by side with nothing between them | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Re-read Robson 2001, pp. 174-175: the notation discussion and Table 2 describe historical slips, not an exclusive condition. Explicitly separated 45:14 can still be mistakenly replaced by 45 + 14 = 59 in notation with zero symbols.
- WRONG | X8: a mistake you can only make in a place-value system where the places are written side by side with nothing between them | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Re-read Robson 2001, pp. 174-175: the notation discussion and Table 2 describe historical slips, not an exclusive condition. Explicitly separated 45:14 can still be mistakenly replaced by 45 + 14 = 59 in notation with zero symbols.
- MINOR | X9: with a tolerance of ten or fifteen degrees. | https://jmhammond.github.io/history/Robson2002.pdf | Re-read Robson 2002, p. 113: the 10 to 15 degree estimate is qualified by "probably".
- WRONG | X10: The angle numbers are ours and are exact; | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs | Re-read reconstruct.mjs: Math.atan2 converts BigInt inputs to Number and uses Math.PI; these degree angles are floating-point approximations. Integer and rational tablet calculations use BigInt exactly.
- MINOR | X11: Robson 2002, p. 117. | https://jmhammond.github.io/history/Robson2002.pdf | Re-read the quotation on the existing image of Robson 2002, printed p. 118; all words match the page quotation.
- MINOR | X12: A transcription slip would have broken both. | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs | Re-read reconstruct.mjs and independently changed row 1 column I from 59 00 15 to 58 00 15 in memory: the triple, generator and angle remain unchanged, while colIMatchesWithOne becomes false.
- WRONG | X13: Every number here is a BigInt rational. | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs | Re-read reconstruct.mjs: Math.atan2 converts BigInt inputs to Number and uses Math.PI; these degree angles are floating-point approximations. Integer and rational tablet calculations use BigInt exactly.
- WRONG | X14: No floating-point value appears anywhere in the reconstruction | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs | Re-read reconstruct.mjs: Math.atan2 converts BigInt inputs to Number and uses Math.PI; these degree angles are floating-point approximations. Integer and rational tablet calculations use BigInt exactly.
- WRONG | X15: 175 offline checks plus a browser pass over this page. | git show f5e5f84f34:verify-the-one-that-broke-off.mjs | Ran the original verifier before editing: 213 offline assertions passed. Its sections 1-8 give 175 numerical/data assertions, section 9 gives 38 HTML assertions. The verifier blob at its first commit a2f78c86d18 equals the pinned f5e5f84f34 blob, so 175 was never the whole offline total.
- MINOR | X16: fails if this page and the reconstruction disagree. | git show f5e5f84f34:verify-the-one-that-broke-off.mjs | Re-read HTML.includes checks and browser section. In-memory replay: original HTML 213/0; changing only the prose 25921 to 25922 still 213/0; replacing every 25921 gives 212/1. The old browser section did not test this sentence.
- WRONG | X17: is not open access. | https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Read the public full text and ResearchGate's statement that Steven Shnider uploaded it on 30 April 2018. The recorded upload date and present access do not establish uninterrupted public availability since then. Re-opened the publisher page: subscription content. The recorded upload predates the layer; free reading does not establish an open reuse licence.
- MINOR | X18: acute angle 33.85° | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; https://jmhammond.github.io/history/Robson2002.pdf | Re-read reconstruct.mjs, generator and display. atan2(161,240) gives 33.855026128294085 degrees, rounding to 33.86; the old intermediate 33.855 displayed as 33.85. Visually re-read Robson Figure 4: row 13 is 33.86.
- MINOR | X19: they reproduce Robson 2002 Figure 4 to two decimal places. | https://jmhammond.github.io/history/Robson2002.pdf ; git show f5e5f84f34:research/plimpton-322/reconstruct.mjs | Re-read reconstruct.mjs, generator and display. atan2(161,240) gives 33.855026128294085 degrees, rounding to 33.86; the old intermediate 33.855 displayed as 33.85. Visually re-read Robson Figure 4: row 13 is 33.86.
- WRONG | X20: a mistake only a place-value notation with no zero can make | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Re-read Robson 2001, pp. 174-175: the notation discussion and Table 2 describe historical slips, not an exclusive condition. Explicitly separated 45:14 can still be mistakenly replaced by 45 + 14 = 59 in notation with zero symbols.
- WRONG | X21: the leading 1 that decides between reading column I as d squared over l squared or s squared over l squared broke off the clay | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X22: The leading 1 that decides between two readings of the tablet broke off the clay | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X23: The leading 1 that decides between two readings of the tablet broke off the clay | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X24: the digit that would settle what the tablet is broke off the edge. | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- WRONG | X25: the digit that would settle what the tablet is broke off the edge. | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | Re-read Britton, Proust and Shnider 2011, p. 524: traces of leading ones are visible on the original. Re-read Mansfield and Wildberger 2017, p. 398: inspection supports original leading ones while the purpose remains interpreted. The identity d²/l² - s²/l² = 1 still holds.
- CONFIRMED | X26: a missing digit no calculation can restore. | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | REJECTED with evidence. The first three leading ones are editorial restorations in tablet.mjs. The two reconstructed ratios differ by exactly 1 and share fractional digits, so those digits alone cannot choose the leading place. Mansfield and Wildberger 2017, p. 398 identifies ones by physical inspection. That does not refute the isolated calculation limit.
- CONFIRMED | X27: a missing digit no calculation can restore. | https://jmhammond.github.io/history/Mansfield2017.pdf ; https://www.researchgate.net/publication/226483872_Plimpton_322_A_review_and_a_different_perspective | REJECTED with evidence. The first three leading ones are editorial restorations in tablet.mjs. The two reconstructed ratios differ by exactly 1 and share fractional digits, so those digits alone cannot choose the leading place. Mansfield and Wildberger 2017, p. 398 identifies ones by physical inspection. That does not refute the isolated calculation limit.
- WRONG | X28: the tablet's whole selection rule is that the generating number be regular | https://arxiv.org/pdf/1004.0025 ; git show f5e5f84f34:verify-the-one-that-broke-off.mjs | Re-read and reran the verifier enumeration: 28 regular generators with at most three fractional sexagesimal places lie in range, 15 are used and 13 unused. Regularity does not determine that subset.
- MINOR | X29: every astronomical claim of the Antikythera mechanism is recomputed from its tooth counts | git show f5e5f84f34:src/content/strata/the-machine-made-of-months.md | Re-read the pinned related-layer markdown: reconstructed mean ratios come from tooth counts, while day conversions and astronomical comparisons use separately supplied values. The requested narrower wording was already present in this layer when this pass began.
- WRONG | X30: two sexagesimal places added into one could only happen in a notation that wrote them side by side with no zero between. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Re-read Robson 2001, pp. 174-175: the notation discussion and Table 2 describe historical slips, not an exclusive condition. Explicitly separated 45:14 can still be mistakenly replaced by 45 + 14 = 59 in notation with zero symbols.
- WRONG | X31: check √(d² − s²) | git show f5e5f84f34:public/strata/the-one-that-broke-off/index.html | Re-read cellFor and renderTable: raw operands changed but row.l stayed corrected. Recomputed row 15: 53² - 56² = -327, while the retained 45 is the corrected-row long side.
- CONFIRMED | Around 1800 BCE, somewhere near Larsa in southern Iraq, someone ruled a clay tablet into four columns and filled it with fifteen rows of numbers. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [001 date-provenance-format] p. 111 dates the comparable Larsa tables to 1822-1784 BCE and places the author near Larsa around 1800 BCE; pp. 106-107 confirm four columns and fifteen rows.
- CONFIRMED | Around 1800 BCE, somewhere near Larsa in southern Iraq, someone ruled a clay tablet into four columns and filled it with fifteen rows of numbers. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [002 four-columns-fifteen-rows] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | the argument about what it is for has run for eighty years. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [004 eighty-years] The abstract dates publication by Neugebauer and Sachs to 1945. About eighty years to the page's August 2026 date is supported.
- CONFIRMED | Columns II and III of Plimpton 322 hold two whole numbers per row. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [007 body-01-2] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | Call them s and d, the short side and the diagonal of a right triangle. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [008 body-01-3] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | Nothing on the tablet gives you the third side. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [009 body-01-4] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | You can get it anyway: subtract the squares and take the root. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [010 body-01-5] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | Every single time, the root comes out a whole number. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [011 body-02-2] The long excerpt supports the early date and regular-number discussion but does not itself prove fifteen integer roots. Robson 2002 Figure 3 and an independent exact recount supply that missing support for the corrected rows.
- CONFIRMED | The tablet is a list of Pythagorean triples, written more than a thousand years before Pythagoras, and it says so in a way you can check yourself in the next thirty seconds. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [012 body-02-3] The long excerpt supports the early date and regular-number discussion but does not itself prove fifteen integer roots. Robson 2002 Figure 3 and an independent exact recount supply that missing support for the corrected rows.
- CONFIRMED | Numerals below are drawn from the data, not photographed: each digit is a count of ten-wedges and unit-wedges, exactly as base sixty writes it. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [015 wedge-values] p. 2 describes narrow unit wedges and wide ten wedges, including their grouping into digits 1-59. The renderer implements those counts.
- CONFIRMED | The left edge of the tablet is broken away, and the greyed leading places are the editor's restoration, not the clay. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [017 broken-restoration] Abdulaziz p. 2 describes the broken edge and the older leading-1 uncertainty. Conditional ratio arithmetic checks out; later physical evidence is treated in the specific WRONG lines rather than inferred from this older excerpt.
- CONFIRMED | Sexagesimal places are written with spaces between them, as editors of these tablets do. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [018 editor-spacing] p. 107 explicitly uses spaces between sexagesimal places and a semicolon between integer and fractional parts.
- CONFIRMED | Column I is a ratio, and it is the reason the tablet is argued about. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [020 body-05-1] Abdulaziz p. 2 describes the broken edge and the older leading-1 uncertainty. Conditional ratio arithmetic checks out; later physical evidence is treated in the specific WRONG lines rather than inferred from this older excerpt.
- CONFIRMED | Its surviving digits are a fraction. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [021 body-05-2] Abdulaziz p. 2 describes the broken edge and the older leading-1 uncertainty. Conditional ratio arithmetic checks out; later physical evidence is treated in the specific WRONG lines rather than inferred from this older excerpt. This applies to the fractional part; X3 corrects the separate assertion that evidence of a leading one is absent.
- CONFIRMED | With the leading 1, column I reads d²/l², the square of the diagonal over the square of the long side. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [023 body-06-2] p. 107 gives the two conditional readings d²/l² and s²/l². Their difference is exactly 1 by the Pythagorean identity; the arithmetic is distinct from the physical evidence for leading 1s.
- CONFIRMED | Without it, the same digits read s²/l², the square of the short side over the same denominator. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [024 body-06-3] p. 107 gives the two conditional readings d²/l² and s²/l². Their difference is exactly 1 by the Pythagorean identity; the arithmetic is distinct from the physical evidence for leading 1s.
- CONFIRMED | And because d² = s² + l², those two quantities differ by exactly one, in every row, always. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [025 body-06-4] p. 107 gives the two conditional readings d²/l² and s²/l². Their difference is exactly 1 by the Pythagorean identity; the arithmetic is distinct from the physical evidence for leading 1s.
- CONFIRMED | The digits after the point are identical under both readings. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [026 body-06-5] p. 107 gives the two conditional readings d²/l² and s²/l². Their difference is exactly 1 by the Pythagorean identity; the arithmetic is distinct from the physical evidence for leading 1s.
- CONFIRMED | Take any row and compute x = (d + s)/l. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [030 body-07-1] p. 108 gives s/l = (x - 1/x)/2 and d/l = (x + 1/x)/2. Rearrangement gives x = (d+s)/l and its reciprocal (d-s)/l. Each corrected row satisfies these identities.
- CONFIRMED | Then 1/x is exactly (d − s)/l, and the two multiply to 1 on the nose. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [031 body-07-2] p. 108 gives s/l = (x - 1/x)/2 and d/l = (x + 1/x)/2. Rearrangement gives x = (d+s)/l and its reciprocal (d-s)/l. Each corrected row satisfies these identities.
- CONFIRMED | Every row of this tablet is one half of a reciprocal pair, run through s/l = (x − 1/x)/2 and d/l = (x + 1/x)/2. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [032 body-07-3] p. 108 gives s/l = (x - 1/x)/2 and d/l = (x + 1/x)/2. Rearrangement gives x = (d+s)/l and its reciprocal (d-s)/l. Each corrected row satisfies these identities.
- CONFIRMED | Reciprocal pairs are not a modern reconstruction imposed on the tablet. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [033 reciprocal-background-1] Read the image of p. 113: the first multiplication-table section was thirty reciprocal pairs learned by heart, and division used reciprocals. This concerns mathematical table training, not a claim that it preceded every literacy lesson.
- CONFIRMED | They are the first thing a trainee scribe in this time and place learned by heart, because division in base sixty was multiplication by a reciprocal. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [034 reciprocal-background-2] Read the image of p. 113: the first multiplication-table section was thirty reciprocal pairs learned by heart, and division used reciprocals. This concerns mathematical table training, not a claim that it preceded every literacy lesson.
- CONFIRMED | And here is the constraint that makes the reading bite: in base sixty, 1/x terminates only when x is built from the primes 2, 3 and 5 and nothing else. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [035 reciprocal-background-3] p. 3 defines regular numbers by primes 2, 3 and 5, allowing negative exponents. Recreated all fifteen row generators to verify the individual application, not just the general theorem.
- CONFIRMED | A scribe could write down the reciprocal of 2, 3, 4, 5, 6, 8, but never of 7. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [036 reciprocal-background-4] The surrounding discussion limits this to exact, finite sexagesimal expressions. p. 4 immediately discusses approximations to 1/7; read in that explicit finite-expansion context, the page's contrast is supported.
- CONFIRMED | All fifteen generating numbers are of exactly that kind. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [037 body-09-1] p. 113 says five standard pairs and at most four sexagesimal places. The source's count is quoted accurately; the separately recorded four-pair recount is preserved as an observation. Numerical ordering and bounds also reproduce.
- CONFIRMED | They run downward without a single break, from 2;24 in row 1 to 1;48 in row 15, and not one of them exceeds four sexagesimal places. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [038 body-09-2] p. 113 says five standard pairs and at most four sexagesimal places. The source's count is quoted accurately; the separately recorded four-pair recount is preserved as an observation. Numerical ordering and bounds also reproduce.
- CONFIRMED | In row 2, column I should run … 14 50 06 15; the clay has … 14 56 15. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [044 body-11-2] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | In row 8 it should run 41 33 45 14 03 45; the clay has 41 33 59 03 45. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [045 body-11-3] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Look at what happened: 50 and 06 became 56. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [046 body-11-4] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | And 45 and 14 became 59. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [047 body-11-5] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Two adjacent places were added together into one. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [048 body-11-6] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column II should read 2 41, which is 161. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [051 body-12-2] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | The clay reads 7 12 01, which is 25921. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [052 body-12-3] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | That is 161 squared, exactly. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [053 body-12-4] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | The scribe was working from squares and, in this one row, wrote down the square instead of taking its root. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [054 body-12-5] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Row 15 is the honest hard case, and it does not resolve. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [056 body-13-1] Table 2 footnote a gives 56, 1 30, 1 46 and 28, 45, 53, attributes the alternatives to Neugebauer/Sachs and Friberg, and states Robson's preference. In the table's error-for-correction notation, "53 for 1 46" means raw 53 should be 106, not the reverse.
- CONFIRMED | The clay reads 56 and 53. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [057 body-13-2] Table 2 footnote a gives 56, 1 30, 1 46 and 28, 45, 53, attributes the alternatives to Neugebauer/Sachs and Friberg, and states Robson's preference. In the table's error-for-correction notation, "53 for 1 46" means raw 53 should be 106, not the reverse.
- CONFIRMED | Either the short side is wrong and the triple is 28, 45, 53, or the diagonal is wrong and the triple is 56, 1 30, 1 46. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [058 body-13-3] Table 2 footnote a gives 56, 1 30, 1 46 and 28, 45, 53, attributes the alternatives to Neugebauer/Sachs and Friberg, and states Robson's preference. In the table's error-for-correction notation, "53 for 1 46" means raw 53 should be 106, not the reverse.
- CONFIRMED | Both are consistent with everything else on the tablet. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [059 body-13-4] Table 2 footnote a gives 56, 1 30, 1 46 and 28, 45, 53, attributes the alternatives to Neugebauer/Sachs and Friberg, and states Robson's preference. In the table's error-for-correction notation, "53 for 1 46" means raw 53 should be 106, not the reverse.
- CONFIRMED | Robson prefers the first, following Friberg; Neugebauer and Sachs preferred the second. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [060 body-13-5] Table 2 footnote a gives 56, 1 30, 1 46 and 28, 45, 53, attributes the alternatives to Neugebauer/Sachs and Friberg, and states Robson's preference. In the table's error-for-correction notation, "53 for 1 46" means raw 53 should be 106, not the reverse.
- CONFIRMED | This is the reading the tablet is famous for outside the field, and it was revived in 2017 to a great deal of press. | https://www.unsw.edu.au/news/2017/08/mathematical-mystery-of-ancient-babylonian-clay-tablet-revealed- | Checker re-read retained from N2: [062 2017-revival] The dated UNSW release and the 2017 paper confirm the public revival of a trigonometric interpretation. Its angle-free ratio theory should not be confused with the older one-degree description.
- CONFIRMED | In its usual form the claim is that column I is tan² or 1/cos², and that the rows are arranged so the acute angle falls by roughly one degree a line. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [063 trigonometric-description] Read p. 107 Figure 3 and its following paragraph: Robson describes approximately one-degree steps as an older trigonometric reading. The computed uneven steps test that description; they do not refute Mansfield and Wildberger's angle-free 2017 theory. The page makes the former limitation explicit.
- CONFIRMED | A table built to step through angles is not what this looks like. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [068 angle-table-inference] Read p. 107 Figure 3 and its following paragraph: Robson describes approximately one-degree steps as an older trigonometric reading. The computed uneven steps test that description; they do not refute Mansfield and Wildberger's angle-free 2017 theory. The page makes the former limitation explicit.
- CONFIRMED | Measured angle, in the sense the claim requires, is not attested in Mesopotamian mathematics of this period. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [069 angle-history-1] pp. 112-113 argue that measured-angle trigonometry lacks the relevant conceptual framework. The page explicitly attributes this historical inference to Robson and distinguishes it from proof.
- CONFIRMED | Reading these fifteen rows as a trigonometric table asks the tablet to be doing something for which there is no other evidence anywhere in the corpus, in order to explain something reciprocal pairs already explain using only what we know these scribes were taught. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [071 angle-history-3] pp. 112-113 argue that measured-angle trigonometry lacks the relevant conceptual framework. The page explicitly attributes this historical inference to Robson and distinguishes it from proof.
- CONFIRMED | what Robson calls wrong angles | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [072 wrong-angles] p. 113 discusses gradient ratios and calls non-right configurations "wrong angles". Its estimated angular leeway is qualified by "probably", addressed separately.
- CONFIRMED | the historical claim about attested angle measurement is Robson's, and it is an argument from the surviving record, which is a strong argument but not a proof. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [074 angle-attribution] pp. 112-113 argue that measured-angle trigonometry lacks the relevant conceptual framework. The page explicitly attributes this historical inference to Robson and distinguishes it from proof.
- CONFIRMED | And the uneven steps do not by themselves refute a trigonometric purpose. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [075 uneven-inference-0] Read p. 107 Figure 3 and its following paragraph: Robson describes approximately one-degree steps as an older trigonometric reading. The computed uneven steps test that description; they do not refute Mansfield and Wildberger's angle-free 2017 theory. The page makes the former limitation explicit.
- CONFIRMED | They refute one common description of it. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [076 uneven-inference-1] Read p. 107 Figure 3 and its following paragraph: Robson describes approximately one-degree steps as an older trigonometric reading. The computed uneven steps test that description; they do not refute Mansfield and Wildberger's angle-free 2017 theory. The page makes the former limitation explicit.
- CONFIRMED | Robson's reading, which is the one the evidence outside the numbers supports best, is unglamorous and much more human than a lost trigonometry. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [077 school-interpretation-0] p. 117 lists reciprocal pairs, cut-and-paste geometry, completing the square and division by regular factors as school techniques. The page labels the teacher reading probable and explicitly says the purpose is not stated.
- CONFIRMED | The tablet's format, its landscape shape, its column headings, and its descending first column all match a class of administrative tables written in Larsa between about 1822 and 1784 BCE. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [078 school-interpretation-1] p. 111 dates the comparable Larsa tables to 1822-1784 BCE and places the author near Larsa around 1800 BCE; pp. 106-107 confirm four columns and fifteen rows.
- CONFIRMED | Its methods, reciprocal pairs and completing the square and dividing out regular factors, are the ordinary furniture of a scribal school. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [079 school-interpretation-2] p. 117 lists reciprocal pairs, cut-and-paste geometry, completing the square and division by regular factors as school techniques. The page labels the teacher reading probable and explicitly says the purpose is not stated.
- CONFIRMED | And its structure, one set-up repeated fifteen times with different well-behaved numbers, matches a teacher's problem list: a way to set the same exercise over and over and check the answers without redoing the work. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [080 school-interpretation-3] Read the quotation against the p. 118 page image. Its wording is accurate and supports the proposed teacher's repeated exercise list; the page-number error has its own line.
- CONFIRMED | Plimpton 322 is also a repetition of the same mathematical set-up fifteen times, each with a different group of well-behaved regular numbers. It would have enabled a teacher to set his students repeated exercises on the same mathematical problem, and to check their intermediate and final answers without repeating the calculations himself. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [081 teacher-quotation] Read the quotation against the p. 118 page image. Its wording is accurate and supports the proposed teacher's repeated exercise list; the page-number error has its own line.
- CONFIRMED | The tablet does not say what it is for. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [083 no-stated-purpose] Read the quotation against the p. 118 page image. Its wording is accurate and supports the proposed teacher's repeated exercise list; the page-number error has its own line.
- CONFIRMED | Robson writes that of the fifteen reciprocal pairs, only five pairs occur in the standard list of reciprocals a scribe memorised. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [085 five-pairs-quotation] p. 113 says five standard pairs and at most four sexagesimal places. The source's count is quoted accurately; the separately recorded four-pair recount is preserved as an observation. Numerical ordering and bounds also reproduce.
- CONFIRMED | It does not touch her argument either way, since her point is that most of the pairs are not in the standard list but were easy to derive. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [089 most-pairs-derivable] p. 113 says five standard pairs and at most four sexagesimal places. The source's count is quoted accurately; the separately recorded four-pair recount is preserved as an observation. Numerical ordering and bounds also reproduce.
- CONFIRMED | Robson and Abdulaziz count seven; Mansfield and Wildberger, and Phillips, count six, the difference being row 13 column I, where a missing empty place may not be an error at all given that scribes of this period did not consistently mark one. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [103 seven-error-count] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Robson and Abdulaziz count seven; Mansfield and Wildberger, and Phillips, count six, the difference being row 13 column I, where a missing empty place may not be an error at all given that scribes of this period did not consistently mark one. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [104 abdulaziz-error-count] Read Abdulaziz Tables 10 and 11, pp. 23-24: four typographical and three computational entries. It additionally calls the row 13 empty-place issue "hardly an error", consistent with reporting its disputed status.
- CONFIRMED | Robson and Abdulaziz count seven; Mansfield and Wildberger, and Phillips, count six, the difference being row 13 column I, where a missing empty place may not be an error at all given that scribes of this period did not consistently mark one. | https://jmhammond.github.io/history/Mansfield2017.pdf | Checker re-read retained from N2: [105 mansfield-error-count] Read 2017 Table 1: six bracketed errors; row 13 column I is not one of them. Raw values in the other cells match Robson and Abdulaziz.
- CONFIRMED | Robson and Abdulaziz count seven; Mansfield and Wildberger, and Phillips, count six, the difference being row 13 column I, where a missing empty place may not be an error at all given that scribes of this period did not consistently mark one. | https://arxiv.org/pdf/1109.3814 | Checker re-read retained from N2: [106 phillips-error-count] Phillips p. 2 states "six generally accepted corrections". The same paragraph says initial 1s are generally accepted, evidence the record did not apply to its missing-digit claims.
- CONFIRMED | How much of column I is restoration is itself not agreed. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [109 restoration-disagreement] Abdulaziz p. 2 describes the broken edge and the older leading-1 uncertainty. Conditional ratio arithmetic checks out; later physical evidence is treated in the specific WRONG lines rather than inferred from this older excerpt.
- CONFIRMED | The greying on rows 1 to 3 follows Robson 2002 Figure 3, which brackets [(1) 59], [(1) 56 56] and [(1) 55 07]. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [110 robson-brackets] Visually checked Figure 3 on p. 107: [(1) 59], [(1) 56 56], [(1) 55 07]. Rendered greying covers the corresponding fractional places; the optional leading 1 is handled by the ratio control.
- CONFIRMED | Abdulaziz brackets row 1 as [1 59 00] instead, so whether the 00 survives on the clay is a live difference between editors. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [112 abdulaziz-brackets] Read Abdulaziz p. 3 Table 1: [1 59 00] 15. Its bracket differs from Robson's [leading material] 00 15; both give the same numerical value.
- CONFIRMED | The values are not in dispute, only how much of them is reading and how much is reconstruction. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [113 values-versus-restoration] Abdulaziz p. 2 describes the broken edge and the older leading-1 uncertainty. Conditional ratio arithmetic checks out; later physical evidence is treated in the specific WRONG lines rather than inferred from this older excerpt.
- CONFIRMED | The photograph is the property of Columbia University's Rare Book and Manuscript Library. | https://cdli.earth/artifacts/254790 | Checker re-read retained from N2: [115 photo-owner] CDLI photograph credit names the Rare Book and Manuscript Library, Columbia University. No tablet photograph is embedded in the page; the displayed wedges are generated paths.
- CONFIRMED | Eleanor Robson, “Neither Sherlock Holmes nor Babylon: A Reassessment of Plimpton 322”, Historia Mathematica 28 (2001), 167–206. doi:10.1006/hmat.2001.2317 | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [119 reference-33] The Oxford PDF opens without authentication and its title page identifies Eleanor Robson, the title, Historia Mathematica 28 (2001), 167-206 and DOI 10.1006/hmat.2001.2317. Crossref MATCH.
- CONFIRMED | Eleanor Robson, “Words and Pictures: New Light on Plimpton 322”, American Mathematical Monthly 109:2 (2002), 105–120. JSTOR 2695324. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [120 reference-34] The PDF cover gives Robson, American Mathematical Monthly 109:2 (February 2002), 105-120, and JSTOR 2695324.
- CONFIRMED | Abdulrahman A. Abdulaziz, “The Plimpton 322 Tablet and the Babylonian Method of Generating Pythagorean Triples”, arXiv:1004.0025 (2010). | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [121 reference-35] The arXiv PDF title page gives Abdulrahman A. Abdulaziz and arXiv:1004.0025v1, 31 March 2010.
- CONFIRMED | D. F. Mansfield and N. J. Wildberger, “Plimpton 322 is Babylonian exact sexagesimal trigonometry”, Historia Mathematica 44 (2017), 395–419. doi:10.1016/j.hm.2017.08.001, open access. | https://jmhammond.github.io/history/Mansfield2017.pdf | Checker re-read retained from N2: [122 reference-36] The PDF title page gives Daniel F. Mansfield and N.J. Wildberger, Historia Mathematica 44 (2017), 395-419, with the cited title. Crossref MATCH.
- CONFIRMED | Anthony Phillips, “The numbers behind Plimpton 322”, arXiv:1109.3814 (2015). | https://arxiv.org/pdf/1109.3814 | Checker re-read retained from N2: [123 reference-37] The arXiv version stamp is 1109.3814v2, 2 October 2015. The author-date line separately says November 3, 2018; citing the 2015 arXiv version is supported, not a bibliographic error.
- CONFIRMED | Open access via Oxford ORA. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [124 oxford-open-access] The Oxford PDF opens without authentication and its title page identifies Eleanor Robson, the title, Historia Mathematica 28 (2001), 167-206 and DOI 10.1006/hmat.2001.2317. Crossref MATCH.
- CONFIRMED | Robson collated the tablet at Columbia. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [126 robson-collation] p. 203 thanks Columbia's Jane Siegel for permission to collate and copy the original tablet.
- CONFIRMED | Tables 10 and 11 tabulate the raw readings independently. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [128 independent-tabulations] Read Abdulaziz Tables 10 and 11, pp. 23-24: four typographical and three computational entries. It additionally calls the row 13 empty-place issue "hardly an error", consistent with reporting its disputed status.
- CONFIRMED | doi:10.1016/j.hm.2017.08.001, open access. | https://jmhammond.github.io/history/Mansfield2017.pdf | Checker re-read retained from N2: [129 2017-open-access] p. 395 states that the article is open access under CC BY-NC-ND.
- CONFIRMED | The tablet itself: CDLI P254790, Columbia University Rare Book and Manuscript Library, CULC 460, provenance Larsa (modern Tell as-Senkereh). | https://cdli.earth/artifacts/254790 | Checker re-read retained from N2: [130 tablet-reference] CDLI P254790 identifies the tablet, Larsa/Tell as-Senkereh and Columbia's collection. Archaeological provenance is attributed, not a controlled excavation proved by this catalogue.
- CONFIRMED | CULC 460, provenance Larsa (modern Tell as-Senkereh). | https://cdli.earth/artifacts/254790 | Checker re-read retained from N2: [131 museum-number] Read CDLI's museum-number field: "CULC 460". The artifact description independently identifies Larsa/Tell as-Senkereh.
- CONFIRMED | Britton, Proust and Shnider, “Plimpton 322: a review and a different perspective”, Archive for History of Exact Sciences 65 (2011), 519–566 | https://link.springer.com/article/10.1007/s00407-011-0083-4 | Checker re-read retained from N2: [132 britton-reference] The Springer metadata gives Britton, Proust and Shnider, Archive for History of Exact Sciences 65, 519-566 (2011), DOI 10.1007/s00407-011-0083-4. Crossref MATCH.
- CONFIRMED | is the most detailed modern treatment of these readings | https://arxiv.org/pdf/1109.3814 | Checker re-read retained from N2: [133 most-detailed-treatment] Phillips p. 1 calls the 2011 study a comprehensive review and a state-of-the-art assyriological document. This supports the page's characterization of its importance; no objective ranking is inferred.
- CONFIRMED | Neugebauer and Sachs, Mathematical Cuneiform Texts (1945) | https://cdli.earth/publications/1785950 | Checker re-read retained from N2: [137 mct-reference] CDLI publication 1785950 identifies Neugebauer and Sachs, Mathematical Cuneiform Texts, American Oriental Series 29, New Haven, 1945.
- CONFIRMED | the first publication of the tablet and the ultimate source of everyone's readings | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [138 first-publication] Robson p. 174 and Abdulaziz's abstract identify Neugebauer and Sachs 1945 as the tablet's first mathematical publication. Later scholars also collated the original, as the page explicitly notes.
- CONFIRMED | Simple copying error: two places conflated | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [149 row-comment-1] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Simple arithmetical error: two places added together | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [150 row-comment-2] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Simple copying error: the empty sexagesimal place should be marked with a blank space | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [151 row-comment-3] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Simple copying error: 9 looks very similar to 8 in cuneiform | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [152 row-comment-4] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Calculation error: square of the correct value | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [153 row-comment-5] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Calculation error: twice or half the correct value | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [154 row-comment-6] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Calculation error: no obvious numerical relationship | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [155 row-comment-7] Read Robson 2001 Table 2, Comments column, p. 175. This quotation matches, including its error classification. Numeric and alternative-reading details were also checked.
- CONFIRMED | Column I: the tablet has 56 56 58 14 56 15 where the arithmetic requires 56 56 58 14 50 06 15. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [160 raw-row-02-I] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column III: the tablet has 3 12 01 where the arithmetic requires 1 20 25. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [161 raw-row-02-III] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column I: the tablet has 41 33 59 03 45 where the arithmetic requires 41 33 45 14 03 45. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [174 raw-row-08-I] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column II: the tablet has 9 01 where the arithmetic requires 8 01. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [177 raw-row-09-II] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column I: the tablet has 27 03 45 where the arithmetic requires 27 00 03 45. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [187 raw-row-13-I] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column II: the tablet has 7 12 01 where the arithmetic requires 2 41. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [188 raw-row-13-II] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Column II: the tablet has 56 where the arithmetic requires 28. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [193 raw-row-15-II] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Counted as an error by Robson and Abdulaziz, not by Mansfield and Wildberger. | https://jmhammond.github.io/history/Mansfield2017.pdf | Checker re-read retained from N2: [195 disputed-error-caption] Read 2017 Table 1: six bracketed errors; row 13 column I is not one of them. Raw values in the other cells match Robson and Abdulaziz.
- CONFIRMED | Which of the two cells is the wrong one is not settled. The alternative correction is the diagonal, 53 for 1 46. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [196 fork-caption] Table 2 footnote a gives 56, 1 30, 1 46 and 28, 45, 53, attributes the alternatives to Neugebauer/Sachs and Friberg, and states Robson's preference. In the table's error-for-correction notation, "53 for 1 46" means raw 53 should be 106, not the reverse.
- CONFIRMED | Plimpton 322 is a clay tablet from southern Iraq, written around 1800 BCE, holding fifteen rows of numbers. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [202 metadata-plain-1] p. 111 dates the comparable Larsa tables to 1822-1784 BCE and places the author near Larsa around 1800 BCE; pp. 106-107 confirm four columns and fifteen rows.
- CONFIRMED | Each row is a Pythagorean triple: two sides are written down and the third comes out a whole number every time, which you can check here. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [203 metadata-plain-2] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | The scribe made seven mistakes, and they show how the work was done, including one row carrying 25921 in place of 161, which is 161 squared. | https://ora.ox.ac.uk/objects/uuid%3Ae3d8eedb-e745-45b3-8612-71f8951599aa/files/m9ffcce389fd7629667ec59059ffe79c1 | Checker re-read retained from N2: [204 metadata-plain-3] Read Table 2 on p. 175 and its footnote. All seven listed raw/corrected cell pairs and the quoted comments match. The disputed row 13 zero and row 15 alternative are explicitly retained.
- CONFIRMED | Around 1800 BCE someone near Larsa ruled a tablet into four columns and filled it with fifteen rows of numbers, and the argument about what it is for has run for eighty years. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [205 metadata-dek-1] p. 111 dates the comparable Larsa tables to 1822-1784 BCE and places the author near Larsa around 1800 BCE; pp. 106-107 confirm four columns and fifteen rows.
- CONFIRMED | Columns II and III give two sides of a right triangle; subtract the squares, take the root, and a whole number falls out fifteen times in a row, which is a list of Pythagorean triples a thousand years before Pythagoras. | https://jmhammond.github.io/history/Robson2002.pdf | Checker re-read retained from N2: [207 metadata-dek-3] pp. 106-107 describe a four-column, fifteen-row tablet and identify its short-side and diagonal columns. These are the conventional integer-scaled, corrected readings. All fifteen corrected triples were independently recomputed; neither the raw erroneous cells nor absolute ancient scales are being certified.
- CONFIRMED | Every generating number turns out regular, built from 2, 3 and 5 alone, so its reciprocal terminates and a scribe could actually write it. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [208 metadata-dek-4] p. 3 defines regular numbers by primes 2, 3 and 5, allowing negative exponents. Recreated all fifteen row generators to verify the individual application, not just the general theorem.
- CONFIRMED | Here, a mathematical tablet whose numbers were never in much doubt and whose *purpose* has been argued for eighty years instead. | https://arxiv.org/pdf/1004.0025 | Checker re-read retained from N2: [231 related-3-3] The abstract dates publication by Neugebauer and Sachs to 1945. About eighty years to the page's August 2026 date is supported.
- CONFIRMED | (added) 169² − 119² = 14400 = 120² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [245 added-row-01-equation] The explicit squared difference printed in row 1 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 4825² − 3367² = 11943936 = 3456² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [246 added-row-02-equation] The explicit squared difference printed in row 2 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 6649² − 4601² = 23040000 = 4800² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [247 added-row-03-equation] The explicit squared difference printed in row 3 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 18541² − 12709² = 182250000 = 13500² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [248 added-row-04-equation] The explicit squared difference printed in row 4 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 97² − 65² = 5184 = 72² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [249 added-row-05-equation] The explicit squared difference printed in row 5 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 481² − 319² = 129600 = 360² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [250 added-row-06-equation] The explicit squared difference printed in row 6 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 3541² − 2291² = 7290000 = 2700² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [251 added-row-07-equation] The explicit squared difference printed in row 7 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 1249² − 799² = 921600 = 960² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [252 added-row-08-equation] The explicit squared difference printed in row 8 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 769² − 481² = 360000 = 600² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [253 added-row-09-equation] The explicit squared difference printed in row 9 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 8161² − 4961² = 41990400 = 6480² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [254 added-row-10-equation] The explicit squared difference printed in row 10 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 75² − 45² = 3600 = 60² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [255 added-row-11-equation] The explicit squared difference printed in row 11 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 2929² − 1679² = 5760000 = 2400² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [256 added-row-12-equation] The explicit squared difference printed in row 12 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 289² − 161² = 57600 = 240² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [257 added-row-13-equation] The explicit squared difference printed in row 13 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 3229² − 1771² = 7290000 = 2700² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [258 added-row-14-equation] The explicit squared difference printed in row 14 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) 53² − 28² = 2025 = 45² | git show f5e5f84f34:research/plimpton-322/reconstruct.mjs ; git show f5e5f84f34:research/plimpton-322/page.json | Checker re-read retained from N2: [259 added-row-15-equation] The explicit squared difference printed in row 15 is omitted from the record's claims, although the operands and triple are included. Independently recomputed with integer arithmetic; identity holds. Positive control: the same search finds 25921 in source entries 52 and 204.
- CONFIRMED | (added) The trigonometric reading in its recent form. | https://jmhammond.github.io/history/Mansfield2017.pdf | Checker re-read retained from N2: [260 added-recent-trigonometry] The omitted source-list characterization is supported by the title and abstract of the cited 2017 paper, which proposes an exact ratio-based trigonometric table. Positive control: record entry 122 contains this paper's bibliographic citation, but not this sentence.

## Findings

- [fixed] X1 (MINOR): Added "one of" and pluralised "object" in the HTML dek. MINOR; no page correction line.
- [fixed] X2 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X3 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X4 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X5 (WRONG): Removed the unsupported count from the prose. The Corrected line reports the data-position count with its restoration and error qualifications, without substituting it for a count of correct surviving digits.
- [fixed] X6 (MINOR): Distinguished copying from arithmetic in the body, markdown dek and lab README; preserved every quoted Table 2 comment. MINOR; no page correction line.
- [fixed] X7 (WRONG): Removed the exclusivity from the HTML body, markdown dek and related note, and lab README. Kept the historical values. Added and mirrored the notation Corrected line.
- [fixed] X8 (WRONG): Removed the exclusivity from the HTML body, markdown dek and related note, and lab README. Kept the historical values. Added and mirrored the notation Corrected line.
- [fixed] X9 (MINOR): Restored "probably" in the page paraphrase. MINOR; no page correction line.
- [fixed] X10 (WRONG): Scoped exactness to integer and rational tablet calculations in the body, markdown dek, README and verifier comments. Labelled angles approximate. Added and mirrored the arithmetic Corrected line.
- [fixed] X11 (MINOR): Changed attribution p. 117 to p. 118 and preserved the quotation. MINOR; no page correction line.
- [fixed] X12 (MINOR): Limited the two checks to columns II and III and named the separate column-I ratio check in the HTML and README. MINOR; no page correction line.
- [fixed] X13 (WRONG): Scoped exactness to integer and rational tablet calculations in the body, markdown dek, README and verifier comments. Labelled angles approximate. Added and mirrored the arithmetic Corrected line.
- [fixed] X14 (WRONG): Scoped exactness to integer and rational tablet calculations in the body, markdown dek, README and verifier comments. Labelled angles approximate. Added and mirrored the arithmetic Corrected line.
- [fixed] X15 (WRONG): Changed 175 offline to 213 offline. Two new browser assertions bring the current full total to 221; updated the layer count copies. Added and mirrored the offline-count Corrected line.
- [fixed] X16 (MINOR): Narrowed the verification promise to the implemented comparisons in the HTML and README, and corrected verifier/build comments. MINOR; no page correction line.
- [fixed] X17 (WRONG): Linked the freely readable author copy and the paywalled publisher copy. Preserved that the original author did not consult it, and distinguished this pass reading p. 524. Updated HTML and README; added and mirrored the access Corrected line. Repair pass narrowed the access history to current availability and the recorded upload date in the HTML source entry, dated line, markdown, generator and lab README.
- [fixed] X18 (MINOR): Kept full computed angle precision in build-page.mjs and regenerated page.json and inline data. Display rounds once; all 15 rounded values now match Figure 4. Added a browser regression assertion. MINOR; no page correction line.
- [fixed] X19 (MINOR): Kept full computed angle precision in build-page.mjs and regenerated page.json and inline data. Display rounds once; all 15 rounded values now match Figure 4. Added a browser regression assertion. MINOR; no page correction line.
- [fixed] X20 (WRONG): Removed the exclusivity from the HTML body, markdown dek and related note, and lab README. Kept the historical values. Added and mirrored the notation Corrected line.
- [fixed] X21 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X22 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X23 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X24 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X25 (WRONG): Corrected the lost-evidence and purpose-settling claims in the HTML dek, body and descriptions, markdown dek, and layer research comments. Preserved the arithmetic limit and editorial restoration flags. Added and mirrored the leading-one Corrected line.
- [fixed] X28 (WRONG): Changed the related note to a constraint whose selection remains unexplained; scoped rationality to sides and ratios. Added and mirrored the selection Corrected line.
- [fixed] X29 (MINOR): Already fixed by a concurrent worker before this pass: retained the narrower mean-ratio and separate-input wording at src/content/strata/the-one-that-broke-off.md:14. No additional edit and no new page correction line for this MINOR item; preserved the pre-existing gear-inventory correction, which addresses a different error.
- [fixed] X30 (WRONG): Removed the exclusivity from the HTML body, markdown dek and related note, and lab README. Kept the historical values. Added and mirrored the notation Corrected line.
- [fixed] X31 (WRONG): Labelled the column "corrected-row check" and explained that the check and detail use corrected readings in both modes. Labelled the detail as corrected, corrected row accessibility labels to the shown sexagesimal operands, and tested raw mode in the browser. Added and mirrored the raw-table Corrected line.