{"id":"5f80e77f-9a10-4098-b988-17f670702122","arxiv_id":"2501.11943","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Most 20th-century physics textbooks wrongly credit Cavendish with measuring G; Persson traces the error to C.V. Boys' 1889 presentation.","lead":"A survey of 84 physics textbooks finds that most repeat the myth that Cavendish measured the gravitational constant G, when his actual goal was to weigh the Earth. The paper traces the false story to a 1889 Royal Society talk by C.V. Boys and suggests a better way to teach the experiment.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Boys-origin claim is undermined by the very quotation used to support it: 'so well known' implies the G-attribution predates Boys, so Section 7's causal story needs a systematic pre-1889 search.","rationale":"The paper contains two intertwined claims: a prevalence claim supported by 63/84 textbooks coded as attributing G to Cavendish, and a provenance claim identifying Boys 1889 as the likely origin. The prevalence claim is reasonably robust despite the convenience sample, because multiple independent titles across the century exhibit the pattern and the coding rule is transparent. The provenance claim is more load-bearing: it is the paper's distinctive explanatory contribution, and its evidence is explicitly the earliest occurrence the author found rather than a systematic search. More importantly, the Boys quotation contains an internal signal against the causal reading: 'so well known' presupposes prior familiarity, whereas the author treats Boys as inadvertently creating a new narrative. This is not a question of disagreeing with the historical consensus; it is a question of whether the cited evidence can bear the causal weight put on it. A targeted pre-1889 search is feasible and would settle the issue. The reader's conditional verdict already treats the provenance as an argument from silence; my analysis gives a sharper reason for that caution. Therefore no change in verdict is needed, but the condition should be explicit: the paper should either verify the absence of earlier attributions or reframe Section 7 as identifying an early influential occurrence rather than the origin.","tokens_in":12557,"tokens_out":4800,"duration_ms":55385,"concrete_test":"Run a systematic keyword search of pre-1889 digitized sources (Philosophical Transactions, Nature, Comptes Rendus, textbooks, encyclopedias) for phrases pairing 'Cavendish' with 'constant of gravitation' or 'gravitation constant'. If any such attribution predates June 1889, Section 7's origin claim fails; if none exists, the claim is supported but the 'so well known' wording still needs explanation as rhetorical, not evidential.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 5 identifies C.V. Boys's 1889 Royal Society paper as the earliest explicit attribution of G to Cavendish, and Section 7 calls it the likely origin of the textbook myth. The supporting quotation, however, reads: 'the Cavendish experiment for determining the constant gravitation ... is so well known that there is no occasion to describe it.' If Boys were introducing a new attribution, calling the experiment 'so well known' for that purpose is anomalous; the phrase suggests the identification of the Cavendish experiment with G was already familiar to his audience. The provenance argument is therefore based on the earliest occurrence the author found, not on a systematic pre-1889 search, as the wording 'the earliest ... I found' concedes. An earlier occurrence in a treatise, encyclopedia, or periodical would make Boys a transmitter rather than the originator, leaving the textbook-survey result intact but falsifying the headline conclusion about origins.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper investigates how 20th-century physics textbooks present the Cavendish experiment, contrasting the historical record (Cavendish measured the Earth's mean density) with the common textbook claim that Cavendish measured the gravitational constant G. Using a five-criterion checklist applied to 84 textbooks, the author reports that 63 of 84 books directly or indirectly credit Cavendish with determining G, while only a handful clearly state the Earth-density goal. The paper further proposes that the misattribution originated with C.V. Boys's 1889 Royal Society paper and argues that textbooks' anachronistic framing reflects contemporary scientific interests. The paper closes with a didactic proposal to reframe the experiment in connection with the development of absolute units.","tokens_in":12743,"tokens_out":3579,"duration_ms":40151,"significance":"If the empirical claims were fully supported, this would be a valuable contribution to physics education research and the historiography of textbook errors: it offers a systematic, quantified survey of an anachronism across a substantial corpus, a transparent coding table, and direct quotations from primary sources. The use of a published checklist, the conservative coding rule, and the inclusion of ambiguous cases as a separate category are strengths. However, the paper's two headline claims are currently unevenly supported: the prevalence of the G-myth in the sampled corpus is credible, but the corpus's representativeness is unclear, and the specific origin claim about Boys is contradicted in part by the very quotation used to support it. These issues are load-bearing for the conclusions as stated.","major_comments":[{"comment":"The claim that Boys's 1889 paper is the earliest explicit attribution of G to Cavendish, and therefore the likely origin of the textbook myth, is not supported by the evidence presented. The supporting quotation itself—\"the Cavendish experiment for determining the constant gravitation ... is so well known that there is no occasion to describe it\"—indicates that the identification was already familiar to Boys's audience, not that he introduced it. The manuscript even concedes the search was not systematic: \"The earliest explicit attribution ... I found\" is an occurrence-based statement, not a provenance proof. An earlier occurrence in a pre-1889 treatise or encyclopedia would reduce Boys to a transmitter, leaving the textbook-survey result intact but falsifying the Section 7 conclusion. I recommend either conducting a systematic search of pre-1889 sources (e.g., Maxwell's Treatise, Everett, encyclopedias, and periodical literature) or reframing the conclusion as \"one influential early articulation\" rather than \"the likely origin.\"","section":"§5 and §7"},{"comment":"There is an internal inconsistency in the reporting of Criterion 2. Table 1 reports 2 textbooks with a clear yes and 3 with an inferred yes, yet §4.2 states that \"Only five out of 84 textbooks state in a clear and definite way that Cavendish's primary goal was to measure the density of the Earth.\" Inspection of Table 2 shows the five entries include three coded Y* (Edser, Watson, Anderson, PSSC 1960, Lerner), so only two are clear yeses by the paper's own coding scheme. This discrepancy matters because it blurs the boundary between clear and inferred statements and affects the interpretation of the prevalence of accurate presentations. The text and table should be reconciled, and the definition of the Y* code should be applied consistently throughout.","section":"Table 1 and §4.2"},{"comment":"The 84-textbook corpus is assembled from personal, institutional, and university libraries plus Internet Archive and Google Books without a defined sampling frame. The paper repeatedly generalizes to \"physics textbooks\" as a whole (\"prevalence of the myth in most textbooks throughout the 20th century\"), but without a documented selection procedure it is unclear whether the sample is representative of the population of textbooks, as opposed to a convenience sample that may overrepresent certain publishers, countries, or educational levels. At minimum, the paper should add a limitations paragraph acknowledging this, and ideally it should report the distribution of the corpus by decade, publisher, and intended audience so that readers can judge the generalizability. The central prevalence claim would be strengthened by a sensitivity check restricted to textbooks from major commercial publishers or by framing all prevalence statements explicitly as true only of the analyzed corpus.","section":"§3 and Table 1"}],"minor_comments":[{"comment":"The final sentence of the abstract is incomplete: \"An explanation of the prevalence of the myth and a different approach\" is a fragment, not a sentence.","section":"Abstract"},{"comment":"In the quotation from Boys, \"the constant gravitation\" likely should be \"the constant of gravitation\"; please verify against the original source.","section":"§5"},{"comment":"The conclusion refers to \"C.V. Boy's presentation\" but the correct name elsewhere is \"Boys\"; correct the typo.","section":"§7"},{"comment":"For Criterion 4, Table 1 gives 12 yes and 12 inferred (total 24), and §4.4 says \"I found this in 24 out of 84 textbooks.\" This is internally consistent only if the inferred cases are counted, but the phrasing \"I found this\" might mislead readers into thinking all 24 are explicit; rephrase to indicate that the total includes inferred cases.","section":"Table 2 / §4.4"},{"comment":"The criteria are numbered 1–5, but Table 1 uses \"Criteria 1\" while the text uses \"Criterion 1\"; standardize the terminology.","section":"§3"}],"recommendation":"major_revision","confidential_remarks":"The paper's scope fits the journal, and the textbook-survey contribution is potentially useful. The main risk is overclaiming the provenance result; the internal counting inconsistency is easily fixable but currently undermines the precision of the central finding. The author might also consider engaging more explicitly with the historiography of physics textbooks and the wider literature on textbook anachronism, beyond the single prior study by Slisko and Hadzibegovic."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the useful part: Persson coded 84 textbooks over a century using Leite's checklist, and the full coding table is included. The core finding—that most textbooks credit Cavendish with measuring G rather than the Earth's density—is robust, and the examples show the error persisting in recent editions. The survey itself is a real contribution, and the inclusion of Table 2 makes it auditable: you can disagree with individual codings, but the work is transparent. That's better than most textbook-history papers.\n\nThe provenance story is where I part ways. The claim that Boys' 1889 paper is the earliest explicit attribution of G to Cavendish, and the likely origin of the myth, is undermined by the very quotation used to support it: \"the Cavendish experiment for determining the constant gravitation ... is so well known that there is no occasion to describe it.\" If Boys thought it so well known, the identification predates him. The author also concedes he found the earliest occurrence rather than searching systematically. So the stress-test note is right: Boys is better described as a transmitter or early popularizer than the originator. The textbook survey doesn't depend on this, so the main conclusion survives, but Section 7 overreaches.\n\nTwo smaller issues. Table 1 says two textbooks clearly state the Earth-density goal, while Section 4.2 says five; the author is likely counting Yes plus \"Inferring\" rows, but the phrase \"clear and definite way\" makes it look sloppy. And the 84-book corpus is a convenience sample, so the exact prevalence numbers should be presented as indicative, not population estimates. That caution is already implicit but deserves explicit placement in the abstract or conclusion.\n\nOverall, this is a solid, honest contribution to the history of physics education. It deserves a serious referee, not a desk reject. I would not accept the origin claim as proven, but that's fixable with a systematic pre-1889 search. The survey alone is worth publishing.","headline":"Useful, auditable textbook survey, but the Boys-1889 origin claim overreaches and needs a systematic pre-1889 search.","tokens_in":13251,"tokens_out":2893,"would_cite":true,"duration_ms":31369,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["01.40.-d","01.65.+g"],"model":"deepseek-v4-flash","headline":"A century of physics textbooks has mislabelled Cavendish's experiment as a measurement of the gravitational constant G, although his 1798 aim was the density of the Earth.","keywords":["history of physics","Cavendish experiment","physics textbooks","gravitational constant","Earth's density","science education","textbook analysis","anachronism"],"falsifier":"Finding a published source earlier than 1889 that explicitly credits Cavendish with determining the gravitational constant would overturn the origin claim. Failing that, a randomly drawn, regionally and education-level balanced sample of 20th-century textbooks that showed substantially lower G-attribution rates would undercut the prevalence claim.","tokens_in":12352,"feed_emoji":"⚖️","tokens_out":8421,"duration_ms":88047,"temperature":0.7,"pith_summary":"The paper claims that the standard textbook story of the Cavendish experiment is historically wrong: Cavendish's 1798 torsion-balance experiment was designed to measure the density of the Earth, not the gravitational constant G, and most 20th-century physics textbooks nonetheless credit him with measuring G. The author applies a five-criterion rubric to 84 textbooks published between 1911 and 2015 and finds that about three-quarters of them give students the G version, while only a handful state the density aim. The paper also proposes an origin for the error: an 1889 Royal Society presentation that reframed the experiment as a measurement of G, picked up by other scientists and the popular press. The point matters because textbook histories shape how students understand scientific change, and the paper suggests a different, unit-system framing that would make the experiment accurate and still relevant today.","feed_headline":"Textbooks have told a false story about Cavendish for a century","feed_subtitle":"A survey of 84 physics textbooks shows most credit Cavendish with measuring G, though he aimed to weigh the Earth.","key_machinery":"The analysis runs on two instruments. First, a five-criterion rubric for historical textbook content—whether the setup is described, whether Cavendish is said to measure Earth's density, whether he is said to determine G, whether he is said to weigh the Earth, and whether the original designer of the torsion balance is acknowledged—applied with a student-like 'if it can be misread as yes, it counts as yes' convention. Second, a provenance hypothesis centred on an 1889 Royal Society paper on the Cavendish experiment whose opening line casually treats the experiment as one 'for determining the constant gravitation'; this is the earliest explicit G attribution the author found, and the paper argues that this framing then spread through scientific and popular reports.","core_discovery":"The paper's central discovery is that the received textbook account of the Cavendish experiment is a durable anachronism. In Cavendish's own 1798 paper, the experiment was reported as 'Experiments to determine the density of the earth'; the gravitational constant G did not become a standard quantity until later in the nineteenth century. Yet of the 84 textbooks surveyed, 59 unambiguously present Cavendish as the experimenter who determined G and 12 more do so by implication; only five clearly present the density aim, and only 13 mention the original designer of the torsion balance. The paper argues that the G framing entered the literature through an 1889 Royal Society presentation in which the Cavendish experiment was assumed to be about the gravitational constant, and that this framing then reproduced itself through the century because textbook authors followed earlier textbooks rather than primary sources.","pith_inferences":["A parallel corpus study of other classic experiments whose modern meaning changed would test whether the same author-copies-textbook mechanism generalizes; I would expect comparable anachronism rates.","The intermediate 'weighed the Earth' framing, found in 12 of the 84 textbooks, suggests the density-to-G shift happened in stages; a first-edition-versus-later-edition comparison could map how each edition mutated.","The proposed unit-system reframing has a testable pedagogical consequence: students taught the dimensional-analysis route to G should be better able to explain why Cavendish's raw result needed re-expression in a standardized unit system.","A systematic search of pre-1889 sources could turn up an even earlier explicit G attribution than the one the paper identifies; if none exists, the provenance hypothesis becomes much stronger."],"forward_implications":["Textbook authors and editors who accept this finding should stop writing that Cavendish measured G; at minimum they should say Cavendish measured the Earth's density and that G was derived from his data later in the century.","The five-criterion rubric can be reused to measure how other canonical experiments are anachronised in textbooks.","A stated connection between G, the Earth's density, and the nineteenth-century standardization of units (CGS, dimensional equations) could replace the false history with an accurate and currently relevant narrative.","Because the false framing appears as early as 1911 and persists across publishers, correcting it requires active checking against primary sources rather than relying on textbook precedent.","Encyclopaedias already gave a more accurate account in 1911, showing that an accurate version was available alongside the textbook story."],"supporting_citations":[{"why":"Supplies the original 1798 report whose title states the density objective that later textbooks misreport.","marker":"[8]"},{"why":"Provides the published checklist on which the textbook assessment is based.","marker":"[9]"},{"why":"An earlier limited study of the same textbook error, whose criteria are extended here.","marker":"[12]"},{"why":"The 1889 Royal Society paper identified as the earliest explicit G attribution and the likely origin of the myth.","marker":"[34]"},{"why":"Historical analysis of the Michelson-Cavendish apparatus and the origin of the anachronism.","marker":"[6]"},{"why":"Detailed historical account of the experiment's actual aim and the later shift to G.","marker":"[7]"},{"why":"Example of a textbook that correctly separates Cavendish's density goal from the later calculation of G.","marker":"[19]"},{"why":"A contemporary reference on the mean density of the Earth, showing the density framing was still standard when the G attribution appeared.","marker":"[35]"}],"fun_headline_variants":["Cavendish myth: 59 of 84 textbooks misstate his experiment","Study: 84 textbooks, only 5 got Cavendish's aim right","A century of textbooks miscast Cavendish's experiment","Cavendish measured Earth's density, not G—textbooks still err"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The percentages in Table 1 are treated as representative of 20th-century physics textbooks, but the 84 books were gathered from personal, institutional, and digital-library collections without a defined sampling frame; the provenance claim also rests on the earliest G attribution the author happened to find, not on an exhaustive pre-1889 archive search.","fun_headline_variants_meta":{"raw":{"variants":["Cavendish myth: 59 of 84 textbooks misstate his experiment","Study: 84 textbooks, only 5 got Cavendish's aim right","A century of textbooks miscast Cavendish's experiment","Cavendish measured Earth's density, not G—textbooks still err"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000903,"raw_usage":{"total_tokens":3823,"prompt_tokens":818,"completion_tokens":3005,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":434,"completion_tokens_details":{"reasoning_tokens":2923}},"tokens_in":434,"tokens_out":3005,"duration_ms":21956,"temperature":1.0,"reasoning_tokens":2923,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T17:40:35.488643+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Finding a published source earlier than 1889 that explicitly credits Cavendish with determining the gravitational constant would overturn the origin claim. Failing that, a randomly drawn, regionally and education-level balanced sample of 20th-century textbooks that showed substantially lower G-attribution rates would undercut the prevalence claim.","supporting_citations":[{"cited_title":"XXI. Experiments to determine the density of the earth","cited_arxiv_id":null,"evidence_quote":"Supplies the original 1798 report whose title states the density objective that later textbooks misreport."},{"cited_title":"History of science in science education: Development and validation of a checklist for analysing the historical content of science textbooks","cited_arxiv_id":null,"evidence_quote":"Provides the published checklist on which the textbook assessment is based."},{"cited_title":"Cavendish Experiment in Physics Textbooks: Why Do Authors Continue to Repeat a Denounced Error?","cited_arxiv_id":null,"evidence_quote":"An earlier limited study of the same textbook error, whose criteria are extended here."},{"cited_title":"I. On the cavendish experiment","cited_arxiv_id":null,"evidence_quote":"The 1889 Royal Society paper identified as the earliest explicit G attribution and the likely origin of the myth."},{"cited_title":"``Weighing''the Earth: a Newtonian Test and the Origin of an Anachronism","cited_arxiv_id":null,"evidence_quote":"Historical analysis of the Michelson-Cavendish apparatus and the origin of the anachronism."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Detailed historical account of the experiment's actual aim and the later shift to G."},{"cited_title":"A., et al","cited_arxiv_id":null,"evidence_quote":"Example of a textbook that correctly separates Cavendish's density goal from the later calculation of G."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"A contemporary reference on the mean density of the Earth, showing the density framing was still standard when the G attribution appeared."}],"review_version":1}