# --------------------------------------------------------------------------- # SEARCH HEADLINE (rules as in apollo-landings/claims.yaml). DRAFT: the second half of the headline rests on a # contested claim (lamp-swan-before-edison / lamp-first-practical-priority), not on the headline claim, so it stays draft. # --------------------------------------------------------------------------- headline: "Edison did not invent the bulb alone: carbon lamps were patented by 1845; Edison v Swan is open" headline_claim: lamp-edison-sole-inventor headline_status: draft dig_status: open search_summary: > Who invented the light bulb? Not Edison alone: a British patent of 1841 covers an electric light fed with powdered charcoal in an exhausted globe, a British patent of 1845 covers a heated carbon conductor in a vacuum, Swan reported a glowing carbon rod in 1878, and US courts upheld the second claim of Edison's 1880 patent, a carbon filament in an all-glass exhausted bulb. Whether Edison or Swan first made a practical lamp is unsettled. assessment: question: "Who invented the light bulb?" answer: partly headline: "Partly Edison, but not him alone: patents for an electric light go back to 1841, and whether Edison or Swan first made a practical lamp is unsettled." text: > "The light bulb" covers at least four different firsts: a patent for an electric light or lamp, a lamp with a heated carbon conductor in a vacuum, a lamp practical to make and use, and a lamp run from a central supply. On the first two, the documents read here put others ahead of Edison: de Moleyns's British patent of 1841 (an electric light fed with powdered charcoal in an exhausted globe, which the Supreme Court calls incandescent but whose abridgment does not), King's of 1845 (heated carbon conductors, carbon in a vacuum; Starr is named by later sources), and, as a reported experiment, Swan's glowing carbon rod of December 1878. On the third, the courts found Edison's lamp "practically operative and successful" and upheld the second claim of his patent (a carbon filament in an all-glass exhausted bulb; Wallace J. credits him with high resistance, which Lacombe J. refused to make a limit of that claim); Swan, by his own account, showed a very thin carbon lamp in October 1880 and Swan lamps lit Cragside (1880, per an archive blog, a proposed claim only) and the Savoy stage (28 December 1881, per a transcription of The Times), both before Edison's Pearl Street service began on 4 September 1882, but no document read shows who first made or sold a practical lamp. On the fourth, Edison's company is the one documented, and nothing earlier was searched for. Two national stories (Lodygin's, and Woodward and Evans's) each hold a real record that is stretched beyond what it shows (Lodygin's "first" is refuted, the Canadian sale is unsupported on the documents read), and a Latimer misattribution is refuted on his patents; the Goebel 1854 lamps are contested, resting on affidavit testimony (about forty affidavits per Colt J.) that US courts weighed differently on preliminary motions, with no dated document from before 1882 either way; de Changy's 1858 lamps are known only through Fontaine and his claim to be first is not dug. One stated standard (S1 to S4, in the description) is applied to every claimant, but the depth differs: some anchors are patent abridgments, a blog or a transcription, and the exceptions are named: the Cragside date rests on a blog about a letter nobody read, so it is filed as its own proposed claim (lamp-swan-cragside), not as established; the Savoy report and the 22 Oct 1879 notebook entry are established on transcriptions (see the description for why). We therefore say "partly", not "Edison" and not "Swan". key_points: - "An 1841 patent (de Moleyns) covers an electric light in an exhausted globe and an 1845 one (King) a heated carbon conductor: Edison was first to patent neither." - "US courts upheld claim 2 of Edison's 1880 patent (a carbon filament in an all-glass exhausted bulb); his first Pearl Street district was lit on 4 Sep 1882." - "Swan lamps lit the Savoy (transcript) and Cragside (blog) before Pearl Street; Edison lamps were seen by visitors at Menlo Park on 30-31 Dec 1879; no document shows who sold first." basis: [lamp-edison-sole-inventor, lamp-de-moleyns-patent, lamp-starr-king-patent, lamp-earliest-patent, lamp-edison-patent-upheld, lamp-first-practical-priority, lamp-swan-before-edison, lamp-swan-in-use, lamp-swan-cragside, lamp-menlo-park-dec-1879, lamp-pearl-street, lamp-goebel-claim] would_settle: "A dated sale, order, price list or company record showing which of Swan's or Edison's lamps was first made and sold for ordinary use, and a dated pre-1882 document for or against Goebel's lamps." subject: incandescent-lamp question: "Who invented the light bulb?" title: "Who invented the light bulb? (the invention era, 1802-1895)" description: > "Who invented the light bulb?" has a famous answer (Edison) and several national counter-answers (Swan for Britain, Woodward and Evans for Canada, Göbel for Germany, Lodygin for Russia, de Changy for Belgium). This dig treats every one of those answers the same way: each is a claim that must be anchored to a dated document from the time. The founding core runs from Davy's battery lights to the 1895 Supreme Court ruling that closed the patent fight. Later metal filaments (tungsten, 1900s) are out of scope for now except where a claimant's own patent reaches into them. THE STANDARD, stated so that it can be checked (and applied to Edison's side as well): (S1) a claim that someone did something by a date is "established" only if a dated document of the time that records it was read directly; a document known only as quoted in one later book, or a later self-account, is a single channel (filed as proposed, contested or a searched gap, never established). (S2) A claim that someone was "first" is refuted only by a dated document, read directly, that shows an earlier person did the same thing; the absence of an earlier document never refutes it (that is a searched gap, capped at provisional). (S3) A patent proves what was claimed and described on its date, not that a lamp was built or worked; "built" needs a report of the lamp. (S4) A court's finding on affidavits is primary text for what the court found, not a document of the period it describes. Under S1 to S4 the claims for de Changy and Lodygin are single-channel (Fontaine), Cragside is single-channel (an archive blog about a letter not read), Goebel's 1854 lamps are contested, and Swan's 1850s work is a gap. Why a transcription is treated differently from Fontaine's quotations: a transcription (the Times page as the Gilbert and Sullivan Archive prints it; Batchelor's entry in the Edison Papers edition) reproduces the dated document itself, whose text can be checked against the original and whose authenticity record is its own entry, whereas Fontaine (1878) reports and selects from a report whose original was not seen, so errors of quotation or selection cannot be told apart. It is a judgment, and both transcription-based claims stay at secondary and moderate, not primary. Two established claims are still anchored below primary (the Savoy report, read in a transcription; the 22 Oct 1879 notebook entry, read in the Edison Papers' transcription) and are shown as limits. # --------------------------------------------------------------------------- # VERIFICATION LEVELS (as in apollo-landings): # V0 single source | V1 2+ independent secondary sources agree | V2 primary document opened and read # What was opened at V2 in the first session (2026-10-02): # US patents 223,898 (Edison), 181,613 (Woodward), 252,386 (Latimer), 47,632 and 266,358 (Göbel), # 575,002 (Lodygin): USPTO image server and Google Patents # Court opinions: 40 F. 21 (1889), 47 F. 454 (1891), 52 F. 300 (1892), 54 F. 678, 56 F. 496, # 57 F. 616 (all 1893), 61 F. 834 (1894): Caselaw Access Project / law.resource.org scans # Supreme Court, The Incandescent Lamp Patent, 159 U.S. 465 (decided 11 Nov 1895): LoC scan # British patent abridgments, Electricity and Magnetism (1859): de Moleyns 1841, King 1845 # Chemical News 39 (1879) and 42 (1880): Swan's demonstrations and his own 1880 lecture # Scientific American 12 Sep 1874: Canadian patent 3,738 listing # NY Herald 21 and 31 Dec 1879; NY Tribune 1 Jan 1880 and 5 Sep 1882 (Chronicling America) # Davy, Elements of Chemical Philosophy (1812; the Bradford and Inskeep edition on the Internet Archive); # Fontaine, Electric Lighting (1878 trans.) # Dyer and Martin, Edison: His Life and Inventions (1910), vol. II pp. 615-616 # Electrical World and Engineer, 14 Apr 1900 # Re-read in the revision of 2026-10-02: all the court opinions, the Supreme Court scan, the abridgments (pp. 62-63, 97-98), # Chemical News 39 (pp. 133, 168 and Proctor's note) and 42 (pp. 227-230), Fontaine pp. 169-173 and 186-191, the Tribune of # 5 Sep 1882 and 1 Jan 1880, the Herald of 21 and 31 Dec 1879, both Kommersant pages, the Edison Papers essay, the # Edison Papers edition of Batchelor's notebook entry of 22 Oct 1879 (Doc. 1831), the Edison Papers pages on Latimer and on # Edison and Swan United, the Times transcription, the IET blog, and US 247,097 (new). Davy's 1812 pp. 85-86 and Edison's # patent text were re-read from the first session's OCR files. # Blocked or unreachable in the revision: the Edison Papers notebook viewer (browser check; the notebook images were not opened), # The Canadian Encyclopedia (HTTP 403), Google Patents (intermittent), Espacenet, Library and Archives Canada. # --------------------------------------------------------------------------- divergence_note: > Adoption says "Edison" almost everywhere, and in each claimant's country it says "ours". The evidence says the question is badly formed: "invented" covers at least four different firsts (a patent, a lamp with a heated carbon conductor in a vacuum, a lamp practical to make and use, a complete wired system), and they have different answers. Several national stories hold one true fact (a real patent, a real report, a real demonstration) that is then stretched into "first": that is shown for Lodygin and for Woodward and Evans; Göbel's is contested and de Changy's first-claim is not dug. The two loudest priority claims against Edison, Göbel's 1854 lamps and the Canadian sale, differ in kind: for Göbel there are about forty affidavits but no dated document from before 1882 either way (the courts split on preliminary motions); for Woodward and Evans there is a real patent and no document of a sale. Two gaps are left open and are not closed by either side: Swan's early work survives only in his own 1880 account, and nothing read shows who first sold a practical lamp, Swan or Edison. claims: # =========================== PRECURSORS =========================== - id: lamp-davy-arc state: established statement: > Humphry Davy published an account of a brilliant arc of light between two charcoal points, powered by the Royal Institution's 2,000-plate battery; the charcoal itself was "ignited to whiteness". This is an arc light, not an incandescent lamp: the light comes mainly from a discharge across a gap, not from a solid glowing inside a bulb. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 3 adoption_note: > "Davy invented the first electric light" is common in school material, often run together with the bulb. would_change_if: > an earlier published description of a sustained carbon arc is found (claims of earlier arcs by others exist and are not yet dug) anchor: type: primary_text description: "Davy, Elements of Chemical Philosophy (1812; the Bradford and Inskeep edition scanned on the Internet Archive), pp. 85-86, read in OCR." sources: [src-davy-1812] nodes: [{node: lamp-1812-davy-publishes-carbon-arc, verb: supports}] basis: > # V: V2 for the text. The book gives no date for the demonstration; 1808, 1809 and 1810 # all trace to later histories (the Supreme Court in 1895 says 1810). Date filed as a gap (lamp-davy-dates). Proves electricity could make useful light by 1812. Does not bear on the incandescent lamp. - id: lamp-davy-dates state: searched_gap statement: > The date of Davy's first arc demonstration (given variously as 1808, 1809 or 1810; the Supreme Court in 1895 says "as early as 1810") and whether his 1802 papers describe wires or charcoal glowing as a light source are not yet read in the original. statement_kind: search_result evidence_class: primary_text confidence: low evidential_weight: 1 anchor_checked: no adoption_weight: 3 adoption_note: "Popular accounts give a single confident year; they do not agree on which." would_change_if: "the Journals of the Royal Institution (1802) and the Royal Institution's lecture records are read" next_step: > Read Davy's 1802 papers in the Journals of the Royal Institution, vol. 1 (the University of Pennsylvania transcription returned 403 from here; try the Royal Institution archive or a Google Books scan). anchor: type: search-record description: "Searched: Davy's 1812 book (no date for the demonstration) and the Supreme Court's passage (a court's statement, 1810). Not read: Journals of the Royal Institution, vol. 1 (1802)." sources: [src-davy-1812, src-scotus-1895] basis: "A search snippet quotes iron wire 'rendered white-hot'. Not read; V0." - id: lamp-de-la-rue-1840 state: searched_gap statement: > The often-repeated claim that Warren de la Rue made a platinum-coil lamp in an evacuated tube in 1840 has no primary record found. The trail leads to a 1927 history that dates a de la Rue lamp to 1820 and says its vacuum is unexplained, while a 1923 Smithsonian history credits the 1840 platinum-coil lamp to William Grove instead. statement_kind: search_result evidence_class: synthetic confidence: low evidential_weight: 1 anchor_checked: secondary adoption_weight: 4 adoption_note: > Very widely repeated online and in school material as the "first light bulb", usually with the 1840 date and no citation. would_change_if: > a paper, notebook or lecture report by de la Rue from 1820-1850 describing the lamp is found, or Grove's own 1840 description is read and shows the attribution swapped next_step: > Search de la Rue's and Grove's papers in Philosophical Magazine and Proceedings of the Royal Society, 1838-1845, for a platinum lamp; then check which history first wrote "de la Rue, 1840". anchor: type: search-record description: "Howell and Schroeder, History of the Incandescent Lamp (1927): 'De la Rue ... 1820'; Schroeder, History of Electric Light (Smithsonian, 1923): Grove, 1840. Searched and not found: Fontaine (1878), the 1859 British abridgments." sources: [src-howell-schroeder-1927, src-schroeder-1923] basis: > # V: V2 for what the two histories say; V0 or less for the lamp itself. Fontaine (1878) and the 1859 British patent abridgments do not mention a de la Rue lamp. Absence is not refutation (S2), so this stays a gap, not refuted. The likeliest reading is a merge of the 1820 de la Rue story with Grove's 1840 lamp, which is itself unread. - id: lamp-de-moleyns-patent state: established statement: > Frederick de Moleyns received British patent No. 9053 on 21 August 1841. One of its heads is "a means of generating the electric light" by supplying pulverized charcoal to the junction of two platinum-wire coil electrodes, the whole enclosed in an exhausted glass globe. The Patent Office's abridgment does not use the words "glow" or "incandescent"; the Supreme Court (1895) describes the patent as one for an incandescent electric lamp using finely pulverized boxwood charcoal or plumbago. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 2 adoption_note: "Known to historians of the lamp; almost absent from popular accounts." would_change_if: "the full specification shows a different design from the abridgment" anchor: type: primary_text description: "Patents for Inventions: Abridgments of Specifications relating to Electricity and Magnetism (1859), pp. 62-63 (No. 9053, A.D. 1841, August 21): second head quoted; 159 U.S. 465 (1895), the Court's description of the patent." sources: [src-abridgments-1859, src-scotus-1895] nodes: [{node: lamp-1841-08-21-de-moleyns-patent-9053, verb: supports}] basis: > # V: V2 for the abridgment (the Patent Office's own published summary). The full specification # was not read. The abridgment describes a flow of powdered charcoal at the electrodes, which reads like an arc fed with charcoal; # whether the lamp glows as an incandescent body is the Supreme Court's description, not the abridgment's. Whether this is the first such patent is a separate, absence-based claim: lamp-earliest-patent. - id: lamp-earliest-patent state: established statement: > Two absences, not one. (1) No patent earlier than de Moleyns's of 1841 for an electric light in an enclosed globe (his: powdered charcoal fed to platinum-coil electrodes in an exhausted glass globe) has been found in the 1859 British abridgments. (2) No document earlier than King's 1845 patent for a heated conductor in a vacuum has been found: the abridgment says "intensely heated" conductors and carbon "enclosed in a Torricellian vacuum", and Fontaine's 1878 search found nothing before it ("The first document on the question that we have found"; he had missed de Moleyns, which he did not find). Absence (2) could fail at 1841: the Supreme Court (1895) calls the de Moleyns patent one "for an incandescent electric lamp" with finely pulverized charcoal, and the full specification of No. 9053 was not read; if it shows a heated charcoal conductor in an exhausted globe, absence (2) moves to 1841. statement_kind: search_result evidence_class: primary_text confidence: provisional # absence anchor: capped at provisional evidential_weight: 3 anchor_checked: primary absence_anchor: true adoption_weight: 2 adoption_note: "Histories that mention de Moleyns usually call his the first patent." would_change_if: > an earlier patent (in Britain, France, the US or elsewhere) for an electric light in an enclosed globe, or for an electrically heated conductor in a vacuum, is found; or the full specification of de Moleyns's No. 9053 shows a heated charcoal conductor in an exhausted globe (which would move absence 2 to 1841) anchor: type: primary_text description: "Abridgments ... Electricity and Magnetism (1859): no earlier lighting entry read; Fontaine (1878), p. 169: 'The first document on the question that we have found, is an English patent of the 4th November, 1845.'" sources: [src-abridgments-1859, src-fontaine-1878] basis: > Two searched absences in two countries' records (British abridgments; a French engineer's survey). French and US patent indexes before 1841 were not searched. - id: lamp-starr-king-patent state: established statement: > British patent No. 10,919 of 4 November 1845, issued to Edward Augustin King "(a communication)", covers "the application of continuous metallic and carbon conductors, intensely heated by the passage of a current of electricity, to the purposes of illumination": thin platinum foil, or carbon enclosed in a Torricellian vacuum (the empty space above a column of mercury). Judge Colt (1893) says Starr "suggested in the King patent a lamp composed of a thin pencil of carbon, inclosed in a Torricellian vacuum"; Fontaine (1878) draws the lamp as a carbon rod in a glass bell over mercury. The abridgment does not name Starr; later sources identify the inventor behind the communication as John Wellington Starr of Cincinnati. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 2 adoption_note: "Known to specialists; Starr is sometimes promoted as the 'American before Edison'." would_change_if: > the full specification shows something other than a carbon conductor in a vacuum, or the Starr identification is contradicted by a document from the 1840s anchor: type: primary_text description: "Abridgments ... Electricity and Magnetism (1859), pp. 97-98 (No. 10,919); 40 F. 21 (1889), quoting the specification; 54 F. 678 (1893): the court's description of Starr's lamp as prior art; Fontaine (1878), pp. 169-170." sources: [src-abridgments-1859, src-40f21, src-54f678, src-fontaine-1878] nodes: [{node: lamp-1845-11-04-king-starr-patent-10919, verb: supports}] basis: > # V: V2 for the patent abridgment and for the courts' and Fontaine's descriptions. The Starr identity is V1 (Judge Colt; # Schroeder 1923; Howell and Schroeder 1927, whose authors overlap, so they are one channel). The full specification was not read. Carbon in a vacuum, the core idea of the later lamp, is patented here 34 years before Edison's US patent. A patent shows the design described on its date, not that a lamp was built (S3). - id: lamp-de-changy-belgium state: proposed statement: > Fontaine (1878) prints a communication he says "M. Jobart, of Brussels" made to the Academy of Sciences on 27 February 1858 about the lamps of de Changy: small lamps "enclosed in hermetically sealed glass tubes", lit by "the incandescence of platinum", intended for mines and street lighting, one of which "was lit in the hollow of my hand". Fontaine places de Changy "in Belgium" and dates his substitution of platinum for carbon to 1857; Howell and Schroeder (1927) call him a French civil engineer who held a Belgian patent of 1856. His nationality is therefore not settled here. This is one witness (Jobard) relayed through one book. statement_kind: reported evidence_class: primary_text confidence: low evidential_weight: 2 anchor_checked: secondary adoption_weight: 2 adoption_note: "Promoted in Belgium (not sourced in this dig); Howell and Schroeder (1927) call de Changy French." would_change_if: "Jobard's report in the Comptes rendus (1858) is read and says otherwise, or confirms it in the original" anchor: type: primary_text description: "Jobard's 1858 communication as quoted in Fontaine, Electric Lighting (1878), pp. 186-187; Howell and Schroeder (1927), p. 33 (nationality and patent)." sources: [src-fontaine-1878, src-howell-schroeder-1927] nodes: [{node: lamp-1858-02-27-jobard-reports-de-changy-lamps, verb: supports}] basis: > # V: V2 for Fontaine's quotation; the Comptes rendus original was not opened, so anchor_checked is secondary. # Lowered from established (first version of this dig): under S1 a document known only through one later quotation is a single # channel. A Belgian patent of 1856 is given by Howell and Schroeder only; Hammer (1900), cited in the first version, was not read. SINGLE-CHANNEL FLAG: this claim and lamp-lodygin-1872-74 both rest on one book, Fontaine (1878), which is the only channel to the Comptes rendus 1858 report and to the Academy's 1874 report. Re-anchor each to its original before raising either. Whether the tubes were exhausted is not stated in the passage read. A separate claim that Jobard made a carbon-in-vacuum lamp in 1838 was not found in Fontaine, Hammer, Schroeder or the abridgments, and is not filed as a claim here. # =========================== RUSSIA: LODYGIN =========================== - id: lamp-lodygin-1872-74 state: proposed statement: > Fontaine (1878) says that in 1873 a Russian physicist, Lodygin, invented a new lamp in which small carbon bars are heated in a hermetically sealed glass receiver from which the oxygen is expelled, and that in 1874 the St Petersburg Academy of Sciences awarded him "a high prize" (the Lomonosov prize) on a report by Wild; Fontaine adds "Nothing is less practical nor less studied than the apparatus of this inventor." Kommersant (2024) gives 1872 for Lodygin's first demonstration (the sentence that gives 1872 names no place; elsewhere the page speaks of a demonstration "in Petersburg") and says the home patent office issued him privilege No. 1619 on 11 July 1874 with its priority dated "retroactively, from 2 October 1872" (one newspaper history, no scan; the same page says the Russian patent came "only a year" after the 1872 petition, which does not fit July 1874, so the number and date are unverified, low confidence). statement_kind: reported evidence_class: primary_text confidence: low evidential_weight: 2 anchor_checked: secondary adoption_weight: 4 adoption_note: "Standard in Russian textbooks and media; little known elsewhere." would_change_if: "the Academy's 1874 prize report or the privilege itself is read and disagrees or confirms" anchor: type: primary_text + secondary description: "Academy report by Wild, quoted in Fontaine (1878), pp. 171-173; Kommersant (St Petersburg, doc 6822124): 1872 first demonstration and petition, privilege No. 1619 of 11 July 1874 with priority from 2 Oct 1872." sources: [src-fontaine-1878, src-kommersant-lodygin] nodes: - {node: lamp-1873-lodygin-carbon-rod-lamps, verb: supports} - {node: lamp-1874-lomonosov-prize-lodygin, verb: supports} - {node: lamp-1874-07-11-lodygin-russian-privilege, verb: supports} basis: > # V: V2 for Fontaine's quotation, V0 for the newspaper. Lowered from established and re-scoped (first version of this dig): under S1 the # Academy report is known only through one later book. Dates disagree: Fontaine says 1873, Kommersant 1872 (shown as a ghost marker). # The privilege number 1619 and the backdated priority date 2 Oct 1872 ARE in Kommersant doc 6822124 ("privilege No. 1619 ... the priority was # dated retroactively, from 2 October 1872"); the first version of this revision wrongly said they were not (corrected in log L-19). They rest on one later # newspaper history with no scan of the privilege. The often-quoted 50,000-rouble prize is not supported by either source: Fontaine says only # "a high prize"; the figure is printed in Howell and Schroeder (1927). PRIVILEGE DETAILS, LOW CONFIDENCE (owner decision, L-24): the number 1619 and the backdated priority date 2 Oct 1872 come from one newspaper history (Kommersant 6822124) with no scan of the privilege, and the same page says the Russian patent came "only a year" after the 1872 petition ("только через год"), which does not fit a July 1874 grant; the number and date are unverified. They are shown, not relied on. SINGLE-CHANNEL FLAG: the contemporary anchor is Fontaine alone (shared with lamp-de-changy-belgium); Kommersant is a later newspaper history and gives dates and a privilege number without a scan. Next step: read the Academy's 1874 prize report. Rods in a sealed vessel are real prior art; Fontaine says they did not last. Fontaine's own phrase "capital errors" refers to the report. - id: lamp-lodygin-first-claim state: refuted refutes_target: "Lodygin was the first inventor of the incandescent lamp (the first to describe or patent it)" refutation_class: narrative_drift statement: > Lodygin was not the first to patent an electric light or lamp in an enclosed globe: a British patent of 1841 (de Moleyns) covers an electric light fed with pulverized charcoal in an exhausted glass globe (the Supreme Court calls it incandescent; the abridgment does not), and one of 1845 (King, for Starr) covers carbon and metal conductors "intensely heated" by a current, the carbon "enclosed in a Torricellian vacuum". The 1845 patent is the sound anchor for a heated carbon conductor in a vacuum. Those patents show earlier published designs, not that a lamp was built or worked (S3), so this does not show that anyone built a carbon lamp before Lodygin; the claim "first to build a working lamp" is not tested here. statement_kind: judgment evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 4 adoption_note: > Widely taught in Russia. A Russian newspaper history (Kommersant, doc 5338615, read) places the "Lodygin before Edison" push in the 1947-48 campaign against "kowtowing to the West": it says that within a year of May 1947 Soviet newspapers reported that Lodygin made the lamp before Edison. That placement is V0, rests on one newspaper history, and is not used as evidence here. would_change_if: "the 1845 patent is shown not to describe a heated carbon conductor in a vacuum (or the 1841 patent not to describe a light in an exhausted globe), or a document earlier than Lodygin's records a working carbon lamp" anchor: type: primary_text description: "British patents 9053 (1841) and 10,919 (1845), as abridged by the Patent Office (1859), pp. 62-63 and 97-98." sources: [src-abridgments-1859] nodes: - {node: lamp-1841-08-21-de-moleyns-patent-9053, verb: supports} - {node: lamp-1845-11-04-king-starr-patent-10919, verb: supports} basis: > The true part ("before Edison") is kept in lamp-lodygin-1872-74. The drift is from "before Edison" to "first". Class narrative_drift, not institutional_propaganda: the Soviet campaign is reported by one newspaper history and is not anchored here. - id: lamp-lodygin-refractory-metals state: established statement: > Lodygin's US patent 575,002 (filed 4 January 1893, granted 12 January 1897; Lodygin "a citizen of Russia, residing in the city of Pittsburg") covers lamp filaments of refractory metals including molybdenum and tungsten. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 2 adoption_note: "Cited in Russian accounts as 'Lodygin invented the tungsten filament'." would_change_if: "the patent text is shown to cover something other than metal filaments" anchor: type: primary_text description: "US 575,002, Illuminant for incandescent lamps (Google Patents rendering of the USPTO record)." sources: [src-us575002] nodes: [{node: lamp-1897-01-12-lodygin-patent-575002, verb: supports}] basis: > # V: V2. Outside this dig's founding core (metal filaments are the 1900s); filed because it is # Lodygin's own record. The story that he sold tungsten rights to General Electric in 1906 is # V1 with no document and no price; filed as a gap. - id: lamp-lodygin-ge-sale state: searched_gap statement: "Whether, when and for what Lodygin sold patent rights to General Electric (often given as 1906)." statement_kind: search_result evidence_class: primary_text confidence: low evidential_weight: 1 anchor_checked: no adoption_weight: 3 adoption_note: "Repeated in Russian popular histories; no price or document cited." would_change_if: "a USPTO assignment record or GE document is found" next_step: "Search USPTO assignment records for patent 575,002 and the GE archives (Schenectady) for Lodyguine." anchor: type: search-record description: "Seen in search results in the first session: skillbox.ru and minervaclassics.com (no documents cited). Neither is in the sources manifest and neither was re-read in the revision; the Kommersant page read (doc 6822124) does not mention General Electric." basis: "V1 only as repetition." # =========================== CANADA: WOODWARD AND EVANS =========================== - id: lamp-woodward-evans-patent state: established statement: > Henry Woodward and M. Evans (Matthew, per Electrical World, 1900) of Toronto received Canadian patent No. 3,738 on 3 August 1874 for "Woodward and Evans' Electric Light". Woodward alone received US patent 181,613 (filed 4 January 1875, dated 29 August 1876). In it a piece of carbon between two electrodes is put in a globe that is emptied of air and then filled with a rarefied gas "that will not unite chemically with the carbon when hot". The patent describes no thin filament and no high vacuum. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 4 adoption_note: "Widely taught in Canada, on CBC, Canada's History and The Canadian Encyclopedia." would_change_if: "the Canadian specification itself (not read) describes a different lamp from the US patent" anchor: type: primary_text description: 'US 181,613 (1876), full specification read; Scientific American 31:11 (12 Sep 1874), Canadian patent list: "3,738 ... H. Woodward and M. Evans, Toronto, Ont. ... Aug. 3, 1874" (OCR damaged, read by hand).' sources: [src-us181613, src-sciam-1874] nodes: - {node: lamp-1874-08-03-woodward-evans-canadian-patent, verb: supports} - {node: lamp-1876-08-29-woodward-us-patent-181613, verb: supports} basis: > # V: V2 for the US patent and for the 1874 listing. The Library and Archives Canada record # returned 403. The 24 July 1874 filing date is V0 (Canadian Encyclopedia only, which could not be re-opened on 2026-10-02). The US patent names later assignees in Toronto and Lindsay, Canada, not Edison. The first name Matthew is from Electrical World (1900), not the patent list. - id: lamp-woodward-evans-performance state: proposed statement: > A trade-press piece of 14 April 1900 (Electrical World and Engineer), based on two Toronto electrical men, says the first lamp was built at Morrison's brass foundry in Toronto and was "a very crude affair", that Woodward filled the globe with nitrogen after exhausting the air, that investors declined to put up more money, and that Woodward left for Europe. Fontaine, who toured the US the year before his 1877 French edition, mentions the 1876 patent only in passing. No contemporary report of the lamp working in use, or being made for sale, was found. statement_kind: reported evidence_class: primary_text confidence: low evidential_weight: 2 anchor_checked: secondary absence_anchor: true adoption_weight: 2 adoption_note: "Popular Canadian accounts say it was 'more effective than any' earlier light; no source given." would_change_if: "a contemporary 1874-79 report of the lamp working in use is found" anchor: type: primary_text description: "Electrical World and Engineer 35:15 (14 Apr 1900): the Toronto informants' account; Fontaine, Electric Lighting (1878), pp. 190-191: 'a patent taken out by Mr. Henry Woodward, in 1876'." sources: [src-elecworld-1900, src-fontaine-1878] basis: > # V: V2 for both texts; the 1900 piece is second-hand testimony 25 years on. Lowered from established and marked as resting partly on # absence: under S1 a later second-hand account is a single channel. No record of commercial production found. - id: lamp-edison-bought-woodward state: searched_gap statement: > Whether Edison bought the Woodward and Evans patent rights, and when and for how much (popular accounts: 1879, $5,000). No document of a purchase was found. statement_kind: search_result evidence_class: primary_text confidence: low evidential_weight: 1 anchor_checked: no adoption_weight: 4 adoption_note: > Stated as fact by CBC (2017), Canada's History and The Canadian Encyclopedia (the last could not be re-opened on 2026-10-02); none cites a document. One site gives both 1879 and 1885. would_change_if: "an assignment, contract, payment record or Edison letter about Woodward is found" next_step: > Search the Edison Papers (Rutgers) for "Woodward" (the notebook viewer is blocked here by a browser check), the USPTO assignment digests for patent 181,613 after 1876, and Library and Archives Canada assignment records for Canadian patent 3,738. anchor: type: search-record description: "Electrical World 1900: the purchase is reported second-hand by two Toronto electrical men, 'purchased by Mr. Edison at the time when he was making his original investigations', with no date or price. US 181,613 shows no assignment to Edison." sources: [src-elecworld-1900, src-us181613] basis: > The earliest trace found is 1900, second-hand, from two Toronto electrical men. The 1876 US patent shows no assignment to Edison. A purchase is plausible (Edison bought other patents to clear the field) but unproven. - id: lamp-canada-invented-claim state: refuted refutes_target: "Woodward and Evans invented the light bulb, and Edison's lamp was built on their patent" refutation_class: narrative_drift statement: > Woodward and Evans were not the first to patent an electric lamp with carbon in an enclosed globe (carbon in a vacuum was patented in Britain in 1845), and their patented lamp is a different design from the one Edison patented: a piece of carbon in a rarefied gas, against Edison's thin high-resistance carbon filament in an all-glass globe exhausted to about one-millionth of an atmosphere. No document read shows that Edison used their design. statement_kind: judgment evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 4 adoption_note: "See lamp-woodward-evans-patent; the claim is common in Canadian media and heritage material." would_change_if: > a document shows Edison used the Woodward patent's design, or the Canadian specification describes a filament in high vacuum anchor: type: primary_text description: "US 181,613 (1876) against US 223,898 (1880), both read in full; British patent 10,919 (1845), abridgment p. 97." sources: [src-us181613, src-us223898, src-abridgments-1859] nodes: - {node: lamp-1876-08-29-woodward-us-patent-181613, verb: supports} - {node: lamp-1880-01-27-edison-patent-223898-granted, verb: supports} reception_records: # what the popular claim says; carries no weight - src-cbc-2017 - src-canadas-history - src-canadian-encyclopedia basis: > The true core (a real 1874 Canadian patent for an electric lamp) is kept. The drift is twofold: "a lamp" becomes "the lamp", and a possible patent purchase becomes "Edison's breakthrough". The Canadian Encyclopedia's "stronger vacuum than Swan's" (as recorded in the first session; the page could not be re-opened on 2026-10-02) is contradicted by the patent, which refills the globe with gas. # =========================== BRITAIN: SWAN =========================== - id: lamp-swan-1850s state: searched_gap statement: > Joseph Swan's experiment with a carbonised-paper arch in a vacuum "about twenty years ago" (that is, about 1860) is known only from his own lecture of 20 October 1880. No record from the 1850s or early 1860s was found. statement_kind: search_result evidence_class: primary_text confidence: low evidential_weight: 2 anchor_checked: primary adoption_weight: 4 adoption_note: "British accounts commonly state 1850 or 1860 as fact." would_change_if: "a dated notebook, letter or society report from 1850-1865 is found" next_step: > Search Swan's papers (Tyne and Wear Archives) and the 1929 memoir by his children (not on archive.org) for contemporaneous records. anchor: type: search-record description: "Swan, lecture of 20 Oct 1880, Chemical News 42 (5 Nov 1880), pp. 227-230, read: 'an experiment which I tried about twenty years ago'; the carbon arch 'became red-hot' and bent until it broke. Nothing contemporary found." sources: [src-chemnews-1880] nodes: [{node: lamp-1880-10-20-swan-lecture-lit-and-phil, verb: supports}] basis: > # V: V2 for the lecture. The id says 1850s because the first version of this dig did; Swan says "about twenty years ago", i.e. about 1860. # This is a self-account given while he was disputing priority with Edison, so it is a primary text for what Swan claimed, not for # what happened around 1860 (S4; the same standard as for the other claimants). - id: lamp-swan-1878 state: established statement: > In December 1878, at the Newcastle Chemical Society meeting "immediately preceding Christmas", Swan (in the chair) described an experiment in which a Siemens dynamo current was passed through a slender rod of carbon in an exhausted globe. The rod glowed with great splendour, but the glass became coated inside with a sooty deposit and the rod curved into a bow "as if from softening". statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 3 adoption_note: "Popular accounts say Swan 'demonstrated a working light bulb' in December 1878." would_change_if: "the Society's own minutes show a different lamp or a live, durable demonstration" anchor: type: primary_text description: "Chemical News 39 (28 Mar 1879) p. 133, the meeting 'immediately preceding Christmas' with Swan in the chair, and (18 Apr 1879) p. 168, the chairman's experiment; B. S. Proctor's note in the same volume: 'At our meeting in December, 1878, Mr. J. W. Swan exhibited an electric lamp'." sources: [src-chemnews-1879] nodes: [{node: lamp-1878-12-swan-rod-lamp-newcastle-chemical-society, verb: supports}] basis: > # V: V2. The report prints no day (the first version of this dig gave 18 December from secondary summaries it did not keep). # The report says the chairman "described an experiment"; Proctor says Swan "exhibited an electric lamp ... which had broken down". # Neither says the lamp was lit before the members. A working principle with a known failure (blackening), not a lasting lamp. - id: lamp-swan-early-1879 state: established statement: > In his lecture of 20 October 1880 to the Newcastle Literary and Philosophical Society, Swan said of "this little lamp, which I showed here two years ago in action" that it was "composed entirely of three substances, namely, glass, platinum, and carbon", and he began by referring to his last lecture on electric lighting "almost two years ago". No 1879 report of the showing was found. Secondary accounts date it 3 February 1879 (or 17 January). statement_kind: reported evidence_class: primary_text confidence: moderate evidential_weight: 3 anchor_checked: primary adoption_weight: 4 adoption_note: "British accounts treat 3 February 1879 as the date the light bulb was first shown." would_change_if: "Society records show no such demonstration in 1879, or a contemporary report describes it" anchor: type: primary_text description: "Swan, lecture of 20 Oct 1880, Chemical News 42 (5 Nov 1880), pp. 227 and 230, read: 'this little lamp, which I showed here two years ago in action'." sources: [src-chemnews-1880] nodes: [{node: lamp-1879-02-03-swan-shows-lamp-lit-and-phil, verb: supports}] basis: > # V: V2 that Swan said so in 1880 (a self-account, S4: primary for what he said, not for what happened); the date is not in it. The # 3 Feb 1879 date (Newcastle University and the Linda Hall Library) and 17 Jan 1879 (Histelec) are from secondary summaries seen only as # search results. No 1879 report found in Nature (Dec 1878 to Mar 1879 searched) or Chemical News. The first version of this # claim said "to a large audience"; Swan says only that "many here" will remember it. See lamp-swan-litphil-date. - id: lamp-swan-litphil-date state: searched_gap statement: "The exact date and content of Swan's early-1879 Literary and Philosophical Society demonstration." statement_kind: search_result evidence_class: primary_text confidence: low evidential_weight: 1 anchor_checked: no adoption_weight: 3 adoption_note: "3 February 1879 is widely given as fact." would_change_if: "the Society's 1879 minutes or a Newcastle newspaper report is read" next_step: "Read the Lit and Phil's 1879 minute book or annual report, and the Newcastle Daily Chronicle for Jan-Feb 1879." anchor: type: search-record description: "Newcastle University Special Collections blog; Linda Hall Library; Histelec (dates disagree), seen only as search summaries; Swan's 1880 lecture gives only 'two years ago'." sources: [src-linda-hall-swan, src-chemnews-1880] basis: "Two secondary dates, no primary." - id: lamp-swan-before-edison state: contested statement: > Whether Swan, not Edison, should be credited with the incandescent lamp. Swan reported a glowing carbon rod in an exhausted globe in December 1878 (a rod that blackened the glass and bent) and said in 1880 that he had shown a glass, platinum and carbon lamp "two years ago"; Edison's laboratory kept a thread lamp lit 13 1/2 hours on 22 October 1879 and applied for its patent on 4 November 1879; in his October 1880 lecture Swan described his own lamp's carbon as "extremely thin — a mere hair" and said (of his own experiments) that blackening of the globe stopped once the vacuum and the contacts were good. Swan also says that in October 1877 he sent Mr. Stearn carbons of carbonised cardboard to be mounted in glass globes and exhausted. No Swan patent on the basic lamp dated before Edison's was found, but Swan's patents were not searched. statement_kind: judgment evidence_class: primary_text confidence: moderate evidential_weight: 4 anchor_checked: primary adoption_weight: 3 adoption_note: "Split by country: 'Swan' in Britain, 'Edison' in the US." would_change_if: "Swan's 1879 lamp is shown to have lasted in use, or an earlier Swan patent on the basic lamp is found" positions: - {label: "Swan first", holder: "Swan himself (Chemical News, Nov 1880); British popular accounts", ground: "a glowing carbon lamp shown in 1878-79, before Edison's October 1879 test"} - {label: "Edison first to a practical lamp", holder: "US courts (47 F. 454; 52 F. 300)", ground: "Edison first to make a high-resistance carbon burner and combine it into 'a lamp which was practically operative and successful'"} - {label: "Neither first; both improvers", holder: "this dig's reading", ground: "carbon in a vacuum was patented in 1845; each solved different problems; the facts of each position are anchored"} disputed_by: - {who: "J. W. Swan", kind: person, position: "Edison was not the first to use carbonised paper in an incandescent lamp, and Swan's glass, platinum and carbon lamp was shown earlier", source_read: "Chemical News 42 (5 Nov 1880), pp. 227-230", via: "his own lecture, given while disputing priority with Edison"} - {who: "Wallace J. (47 F. 454) and Lacombe J. (52 F. 300)", kind: institution, position: "Edison's patent covers the first lamp that was practically operative and successful", source_read: "47 F. 454; 52 F. 300", via: "a patent suit between the Edison company and a rival, not a priority contest with Swan"} anchor: type: primary_text description: "Chemical News 39 (1879) and 42 (1880), read; Edison Papers, Doc. 1831 (notebook transcription), read; 47 F. 454 (1891); US 223,898." sources: [src-chemnews-1879, src-chemnews-1880, src-edison-papers-doc1831, src-47f454, src-us223898] nodes: - {node: lamp-1878-12-swan-rod-lamp-newcastle-chemical-society, verb: supports} - {node: lamp-1879-02-03-swan-shows-lamp-lit-and-phil, verb: qualifies} - {node: lamp-1879-10-22-batchelor-cotton-thread-lamp, verb: qualifies} basis: > Contested on what "credit" means, not on the facts. The facts in each position are anchored. Swan's own account (Chemical News 42, p. 229) dates the start of his carbon-in-globe work to "October, 1877, I sent to Mr. Stearn a number of carbons, made from carbonised cardboard, with the request that he would get them mounted for me in glass globes by a glassblower"; like the "two years ago" and "twenty years ago" passages this is Swan's claim, a primary text for what he said and not a dated record of what happened (S4). Swan's own patents begin with British No. 4933 of 27 Nov 1880 (V1, not read); Grace's Guide mentions 1860 and 1878 patents with no numbers (V0, doubtful). Edison's British patent of 10 Nov 1879 is listed on the USPTO certificate printed with US 223,898 (the number is illegible in the scan). - id: lamp-swan-in-use state: established statement: > The Times of 29 December 1881 (read in a transcription) reports that Swan incandescent lamps lit the stage of the Savoy Theatre, London, for the first time at a performance of Patience on the afternoon of 28 December 1881; the auditorium had had electric light since the theatre opened. (The Cragside half of the earlier version of this claim is now lamp-swan-cragside.) statement_kind: reported evidence_class: primary_text confidence: moderate evidential_weight: 3 anchor_checked: secondary adoption_weight: 3 adoption_note: "Underhill, Swan's own house, is also called 'the first house lit by electric light'; no primary date found." would_change_if: "the original Times page gives different dates or wording" anchor: type: primary_text description: "The Times, 29 Dec 1881, as transcribed by the Gilbert and Sullivan Archive (the original page was not compared): 'the stage was for the first time lit up by the electric light ... The Swan incandescent lamps were used.'" sources: [src-times-savoy] nodes: - {node: lamp-1881-12-28-savoy-stage-lit-by-swan-lamps, verb: supports} basis: > # V: Times report read in transcription, not the original page, so anchor_checked is secondary (a limit shown on the page). The first version said Swan lamps were in use "from # 28 December 1881" on the Savoy stage; The Times describes one experimental performance. Histelec gives 29 Dec and 1,200 lamps (V0). # Narrowed in the revision of round 3: the Cragside half moved to lamp-swan-cragside (L-21). - id: lamp-swan-cragside state: proposed statement: > An IET Archives blog (2023) says that Armstrong replaced arc lamps at Cragside, his house at Rothbury, Northumberland, in 1880 with 45 incandescent lamps, and, in a separate passage, that Armstrong wrote to the editor of The Engineer on 17 January 1881 "shortly after he utilised Joseph Swan's newly invented incandescent light bulbs". That the 1880 lamps were Swan's, and the date they were first lit, are therefore joined by the blog from two passages; the letter itself was not read. statement_kind: reported evidence_class: primary_text confidence: low evidential_weight: 2 anchor_checked: secondary adoption_weight: 3 adoption_note: "Cragside is widely called the first house lit by hydro-electricity; the blog says so and that is not used here." would_change_if: "Armstrong's letter of 17 January 1881 (IET Archives NAEST 046/24, The Engineer) is read and gives different lamps or dates" anchor: type: secondary description: "IET Archives blog 'Cragside: harnessing the power of water' (9 Mar 2023): 'replaced by Armstrong in 1880 with 45 incandescent lamps'; the letter of 17 January 1881 'shortly after he utilised Joseph Swan's newly invented incandescent light bulbs'. The letter itself was not read." sources: [src-iet-cragside] nodes: - {node: lamp-1880-cragside-lit-with-swan-lamps, verb: supports} basis: > # V: single channel (S1): a later blog about a letter nobody here read. Split from lamp-swan-in-use in the round-3 revision (L-21). The blog's "first house in the world to be lit by hydro-electricity" is not used. - id: lamp-first-practical-priority state: contested statement: > Who first made, and who first sold, a lamp practical for ordinary use is not settled by anything read here. Edison's laboratory had by 16 November 1879 a lamp it called "the first lamp that answers the purpose we have wished" (carbonised card), and Wallace J. (1891) found Edison "the first to make a carbon of materials and by a process which was especially designed to impart high specific resistance to it" and the first to combine such a burner with the necessary adjuncts, so that "he made a lamp which was practically operative and successful". Lacombe J. (52 F. 300, 1892) said "the knowledge that practical stability could be given to a carbon filament was not gained until October, 1879", a court's view from the patent suit and not a dated record. Edison's company turned on its first Pearl Street district on 4 September 1882. Swan said in October 1880 that he now used a material better than carbonised cardboard; Swan lamps lit Cragside in 1880 (an IET blog; the letter of 17 January 1881 is not read) and the Savoy stage on 28 December 1881. Both Swan dates come before Pearl Street, and no document read shows who sold the first lamp. statement_kind: judgment evidence_class: primary_text confidence: low evidential_weight: 3 anchor_checked: secondary adoption_weight: 4 adoption_note: "Popular accounts name Edison as the maker of the first practical bulb; British accounts name Swan." would_change_if: > a dated sale, order, price list or company record for lamps by Swan or Edison before the other's is found, or Armstrong's letter and the original Times page are read and give different dates positions: - {label: "Edison first practical", holder: "US courts (47 F. 454; 52 F. 300); popular accounts in the US", ground: "Wallace J.: a high-resistance carbon burner in an all-glass vacuum, found 'practically operative and successful' (claim 2 upheld on appeal without a high-resistance limit); patent 223,898 filed 4 Nov 1879; Edison lamps seen by visitors at Menlo Park on 30-31 Dec 1879, a laboratory and contested as an exhibition (lamp-menlo-park-dec-1879)"} - {label: "Swan's lamps in use first", holder: "Swan (1880 lecture); British accounts", ground: "Swan lamps at Cragside (1880) and the Savoy (28 Dec 1881) before Pearl Street (4 Sep 1882); sale not documented"} - {label: "Not decided by the documents read", holder: "this dig's reading", ground: "the courts compared Edison's lamp with prior art in the patent suits, not with Swan's sales; the Swan dates rest on a blog and a transcription"} disputed_by: - {who: "J. W. Swan", kind: person, position: "His glass, platinum and carbon lamp was shown before Edison's, and Edison was not first to use carbonised paper", source_read: "Chemical News 42 (1880), pp. 227-230", via: "his own lecture"} anchor: type: primary_text description: "47 F. 454 (1891), Wallace J., quoted; Edison Papers essay and Doc. 1831 (22 Oct and 16 Nov 1879); NY Tribune 5 Sep 1882, p. 8 (Pearl Street); Chemical News 42 (1880); IET Archives blog (Cragside); The Times 29 Dec 1881 via transcription (Savoy)." sources: [src-47f454, src-52f300, src-edison-papers-essay, src-edison-papers-doc1831, src-tribune-1882-09-05, src-chemnews-1880, src-iet-cragside, src-times-savoy] nodes: - {node: lamp-1891-47f454-wallace-upholds-edison-lamp-patent, verb: supports} - {node: lamp-1879-11-16-first-lamp-that-answers-the-purpose, verb: supports} - {node: lamp-1882-09-04-pearl-street-district-service-begins, verb: supports} - {node: lamp-1881-12-28-savoy-stage-lit-by-swan-lamps, verb: qualifies} - {node: lamp-1880-cragside-lit-with-swan-lamps, verb: qualifies} basis: > New in the revision of 2026-10-02: the first version put "first practical to make and sell" into Edison's column although the Swan dates precede Pearl Street and nothing read shows a first sale. Kept contested at low confidence, with evidential weight 3, because two of the three Swan anchors are a blog and a transcription (S1) and the court finding compares Edison with the prior art in the case, not with Swan's commercial lamps. Contested on dates and on what "practical" and "first" mean. # =========================== UNITED STATES: EDISON =========================== - id: lamp-edison-oct-1879-test state: established statement: > The Edison Papers' edition of Charles Batchelor's laboratory notebook entry for 22 October 1879 records a lamp of carbonised cotton thread ("6 cord thread No 24") that was "on from 1.30 AM till 3 pm 13 1/2 hour and was then raised to 3 gas jets for 1 hour then cracked glass & busted". The Papers' essay says that by 16 November 1879 the laboratory had "the first lamp that answers the purpose we have wished", a carbonised card horseshoe. statement_kind: reported evidence_class: primary_text confidence: moderate evidential_weight: 4 anchor_checked: secondary adoption_weight: 4 adoption_note: "Popular accounts give 13.5 or 40 hours, and 21 or 22 October. 40 hours was not found in any source opened." would_change_if: "the notebook image (N-79-07-31, p. 105; Edison Papers digital edition TAED N052:52) is read and gives a different figure" anchor: type: primary_text description: "Papers of Thomas A. Edison, Doc. 1831 (a transcription of Batchelor's notebook entry, '[Menlo Park,] Oct 22nd 1879 9 A.M.'); Edison Papers essay 'The Carbon-Filament Lamp' (22 Oct and 16 Nov 1879)." sources: [src-edison-papers-doc1831, src-edison-papers-essay] nodes: - {node: lamp-1879-10-22-batchelor-cotton-thread-lamp, verb: supports} - {node: lamp-1879-11-16-first-lamp-that-answers-the-purpose, verb: supports} basis: > # V: V1 to V2. The Edison Papers is the documentary edition, and its transcription (Doc. 1831) was read directly in the revision of # 2026-10-02, but the notebook image behind the TAED viewer was behind a browser check and was not opened, so anchor_checked stays # secondary. The first version cited the notebook as N052105 (wrong: the edition gives N-79-07-31:105, TAED N052:52) and # quoted 52 F. 300 for "not gained until October, 1879". The first revision wrongly said those words were not in that opinion; they are # (Lacombe J.: "The knowledge that practical stability could be given to a carbon filament was not gained until October, 1879."), a court's # view from the patent suit, not a dated record (corrected in log L-19; the sentence is cited in lamp-first-practical-priority). The essay says "By November 16", not that the lamp is dated to that day. - id: lamp-edison-patent state: established statement: > Edison's US patent 223,898, "Electric-Lamp", was applied for on 4 November 1879 (signed 1 November) and granted on 27 January 1880. Its first two claims (of four) are: a carbon filament of high resistance; and carbon filaments in a receiver made entirely of glass, exhausted of air, with wires sealed through the glass. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 5 adoption_note: "The patent is the anchor of the popular story; few readers know what it actually claims." would_change_if: "the USPTO record is shown to be misread" anchor: type: primary_text description: "US 223,898, full specification and claims (signed 1 Nov 1879; patented 27 Jan 1880); 40 F. 21 (1889): 'applied for November 4, 1879, and issued January 27, 1880'; 52 F. 300 (1892): 'under an application filed November 4, 1879, letters patent No. 223,898 were issued'." sources: [src-us223898, src-40f21, src-52f300] nodes: - {node: lamp-1879-11-04-edison-files-patent-223898, verb: supports} - {node: lamp-1880-01-27-edison-patent-223898-granted, verb: supports} basis: > # V: V2. The specification names a "mercury-pump" and a vacuum of "one-millionth of an # atmosphere"; it does not name Sprengel (see lamp-sprengel-link). The first version cited 52 F. 300 for the filing date; the # first revision wrongly said the quoted words were not in that opinion (they are: "under an application filed November 4, 1879"; log L-19), so # both 40 F. 21 and 52 F. 300 are cited. The application file wrapper was not read. - id: lamp-edison-high-resistance state: established statement: > What set Edison's lamp apart, as his patent and Wallace J. (47 F. 454, 1891) described it, was high resistance (100 to 500 ohms in the patent's example) in a thin carbon filament in an all-glass high-vacuum bulb, which let many lamps run side by side ("in multiple arc") from one supply without enormous copper mains. High resistance was NOT made a limit of what was upheld: the circuit court held claim 1 (the high-resistance filament "made as described") limited to leading wires secured by cement carbonised in situ and not infringed, which the complainant did not appeal; on appeal Lacombe J. (52 F. 300, 1892) upheld claim 2 and refused to read into it "that the filament must be one of high specific resistance, or of at least 100 ohms resistance", the draughtsman having omitted "of high resistance" and "made as described". The high-resistance idea itself was not unique to him: the 1891 court noted Lane-Fox's 1878-79 British patents and a letter of his to the Times of December 1878, and the 1892 appeals court noted Edison's own French patent of May 1879 (a platinum burner) as already stating it. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 2 adoption_note: "Popular accounts credit Edison with the bulb, rarely with the electrical reason it worked as a system." would_change_if: "the 1891-92 opinions are shown to rest the patent on something else, or to make high resistance a limit of the upheld claim" anchor: type: primary_text description: "US 223,898 specification; 47 F. 454 (1891), Wallace J. (Lane-Fox patents and letter); 52 F. 300 (1892), Lacombe J. (Edison's French patent No. 130,910 of 28 May 1879)." sources: [src-us223898, src-47f454, src-52f300] basis: > # V: V2. In the courts' words (not this dig's): Wallace J. held Edison "the first to make a carbon of materials and by a process which was # especially designed to impart high specific resistance to it", the first to make it in a form with "high total resistance", and # the first to combine such a burner with "the necessary adjuncts of lamp construction"; Colt J. (54 F. 678) records the earlier ruling as # holding that claim 2 "covered a broad and fundamental invention, namely, an incandescent lamp, composed of a carbon filament, # hermetically sealed in an all glass chamber exhausted to a practically perfect vacuum, and having leading-in wires of platinum". # The first version said the courts upheld the patent "on that specific combination, not on the idea of the lamp"; that paraphrase # is replaced by these quotations. - id: lamp-menlo-park-dec-1879 state: contested statement: > Whether Edison gave a public demonstration on New Year's Eve 1879. The New York Herald of 21 December 1879 reported that "the first public exhibition" of his light was "announced to take place on New Year's Eve at Menlo Park"; the Herald of 31 December (dateline Menlo Park, 30 December) said "the public exhibition may be said to be actually going on now"; the New-York Tribune of 1 January 1880 reported crowds at Menlo Park the evening before under a headline that no public exhibition was "promised or given", called the announcement "unauthorized" and quoted Edison as disclaiming "any formal exhibition". statement_kind: judgment evidence_class: primary_text confidence: moderate evidential_weight: 4 anchor_checked: primary adoption_weight: 4 adoption_note: "Popular accounts name 31 December 1879 as the public unveiling." would_change_if: "an Edison laboratory record or third newspaper settles the event's nature" positions: - {label: "Public exhibition on 31 Dec", holder: "NY Herald, 21 Dec 1879; later popular accounts", ground: "the Herald's announcement"} - {label: "Open-house crowds, no formal exhibition", holder: "NY Tribune, 1 Jan 1880; NY Herald, 31 Dec 1879 ('thronged with visitors' from 30 Dec; 'fixed no day for a general inspection')", ground: "same-week reports; Edison's own disclaimer as quoted by the Tribune"} disputed_by: - {who: "New-York Tribune (1 Jan 1880)", kind: work, position: "No public exhibition was promised or given; the announcement was unauthorised", source_read: "NY Tribune, 1 Jan 1880, p. 8", via: "reporter's account quoting Edison"} - {who: "New York Herald (21 Dec 1879)", kind: work, position: "A first public exhibition was announced for New Year's Eve", source_read: "NY Herald, 21 Dec 1879, p. 5", via: "unsigned article"} anchor: type: primary_text description: "NY Herald, 21 Dec 1879, p. 5; NY Herald, 31 Dec 1879, p. 3 (dateline 30 Dec); NY Tribune, 1 Jan 1880, p. 8; all read in OCR of the Library of Congress scans (ALTO text)." sources: [src-herald-1879-12-21, src-herald-1879-12-31, src-tribune-1880-01-01] nodes: - {node: lamp-1879-12-21-herald-announces-new-years-eve-exhibition, verb: supports} - {node: lamp-1879-12-31-menlo-park-visitors-no-formal-exhibition, verb: qualifies} basis: > # V: V2 (OCR of Chronicling America scans). Not disputed: lamps were lit and seen by many visitors # at Menlo Park in the last days of 1879 (sixty lights per the Herald; fifty to sixty per the Tribune). The Herald article is unsigned in the scan; the usual # attribution to Marshall Fox is not confirmed. Its headings include "A scrap of paper" and "Success in a cotton thread". # The Tribune headline's OCR is damaged and is read as "no public exhibition promised or given". - id: lamp-edison-bamboo state: established statement: > In the Supreme Court's 1895 account, which rests on trial testimony, Edison, after trying "thirty or forty different woods" and many other fibres, found his best burner in a strip of bamboo (one that "formed the edge of a palm-leaf fan", in the OCR's 'palmieaf'), "despatched a man to Japan" to arrange supplies, and the messenger he sent "to different parts of Japan and China sent him about forty different kinds of bamboo". The Court describes the commercial Edison lamp of the case as having a burner of carbonised bamboo. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 4 anchor_checked: primary adoption_weight: 3 adoption_note: "The popular figure of 1,200 hours of lamp life was not found in any source opened." would_change_if: "Edison laboratory or factory records give a different material or date" anchor: type: primary_text description: "159 U.S. 465 (1895), Brown J., the Court's account of Edison's search, read in the Library of Congress scan." sources: [src-scotus-1895] basis: > # V: V2 for the Court's account, which rests on trial testimony. The first version said the commercial lamps "of 1880 onward" used # bamboo and that an agent "sent back about 40 kinds"; the opinion does not date the lamps to 1880 and says the messenger sent # them from Japan and China. The date is dropped. A court-built lamp to the 1880 # design lasted 600 hours (52 F. 300, expert testimony, V2). Lamp-life claims are filed only at # that strength. - id: lamp-pearl-street state: established statement: > On 4 September 1882, at 3 p.m., the Edison Electric Illuminating Company turned on the electric light in part of the first of its districts in New York, supplied from a central station on Pearl Street just below Fulton Street (New-York Tribune, 5 September 1882). statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 4 adoption_note: "Well known; often presented as the moment 'the light bulb' arrived, which conflates lamp and system." would_change_if: "company records give a different date" anchor: type: primary_text description: "NY Tribune, 5 Sep 1882, p. 8, read in the ALTO OCR text of the Library of Congress scan: 'The company at 3 p.m. turned on the electric light in part of the first of the districts ... The central station ... is at ... Pearl-st., just below Fulton-st.'; 'yesterday' dates it to 4 September." sources: [src-tribune-1882-09-05] nodes: [{node: lamp-1882-09-04-pearl-street-district-service-begins, verb: supports}] basis: "# V: V2. Generation, wiring and metering were the commercial invention; the lamp was one part. The street number is garbled in the OCR and is not used." - id: lamp-sprengel-link state: searched_gap statement: > The common statement that Edison's vacuum came from Hermann Sprengel's mercury pump (1865) is not yet anchored. Edison's patent and the 1879 Herald say only "mercury pump". statement_kind: search_result evidence_class: primary_text confidence: low evidential_weight: 2 anchor_checked: no adoption_weight: 3 adoption_note: "Stated as fact in most technical histories." would_change_if: "an Edison laboratory record or Francis Upton letter names the Sprengel design" next_step: "Search the Edison Papers for 'Sprengel' (1879), and read Sprengel's 1865 paper in J. Chem. Soc." anchor: type: search-record description: "US 223,898: 'mercury-pump', no name; NY Herald 21 Dec 1879: 'mercury pump'. Swan (1880) names the Sprengel pump for his own lamps." sources: [src-us223898, src-herald-1879-12-21, src-chemnews-1880] basis: "Plausible and widely agreed, but nothing read names the source of Edison's pump." # =========================== THE PATENT FIGHT =========================== - id: lamp-sawyer-man-priority state: established statement: > In the US Patent Office, an interference declared in September 1880 between Edison (application of December 1879) and William Sawyer and Albon Man (application of 9 January 1880), over carbon made from paper, was "not finally decided until the beginning of 1885", when the Commissioner of Patents awarded priority of invention to Sawyer and Man (as 40 F. 21 recounts it). Their patent (No. 317,076, granted 12 May 1885) was held invalid in 1889: apart from its third claim (carbonised paper), it amounted to a claim to all fibrous carbons. The Supreme Court affirmed the decree on 11 November 1895, holding that the claims "with the exception of the third, are too indefinite to be the subject of a valid monopoly" and that the question of priority "becomes unnecessary"; it did not itself decide that Sawyer and Man's third claim was invalid, which was admitted not to be infringed by the Edison lamp. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 1 adoption_note: "Almost unknown outside patent history; it cuts against both 'Edison won everything' and 'Edison stole it'." would_change_if: "the Patent Office interference record is read and disagrees with the 1889 opinion's account" anchor: type: primary_text description: "Consolidated Electric Light Co. v. McKeesport Light Co., 40 F. 21 (1889), Bradley J.; The Incandescent Lamp Patent, 159 U.S. 465 (1895), Brown J., headnote and the passages 'we are all agreed that the claims of this patent, with the exception of the third, are too indefinite' and 'becomes unnecessary'." sources: [src-40f21, src-scotus-1895] nodes: - {node: lamp-1885-early-sawyer-man-win-interference, verb: supports} - {node: lamp-1889-10-05-bradley-holds-sawyer-man-claims-invalid, verb: supports} - {node: lamp-1895-11-11-supreme-court-affirms-sawyer-man-decree, verb: supports} basis: > # V: V2. The priority award is recorded in testimony quoted in the 1889 opinion, not in the # Commissioner's decision itself (not read). Some popular accounts date the Patent Office ruling # to 1883; the 1889 opinion says the case was not decided until the beginning of 1885. # Corrected in the revision (see log): the first version, and a first draft of the log, overstated what the Supreme Court held. The scan's # heading misprints the decision date as 1815 and the headnote the patent date as 14 May; the term, the argument dates and 40 F. 21 fix 1895 and 12 May. - id: lamp-edison-patent-upheld state: established statement: > Edison's lamp patent was upheld in court: in Edison Electric Light Co. v. United States Electric Lighting Co., the second claim (carbon filaments in an all-glass exhausted bulb) was held valid and infringed in 1891 (July, per 57 F. 616), and the appeals court affirmed on 4 October 1892. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 2 adoption_note: "Rarely mentioned; the patent's term was tied to foreign patents, so its end is not stated in anything read here." would_change_if: "the opinions are misread" anchor: type: primary_text description: "47 F. 454 (C.C.S.D.N.Y. 1891); 52 F. 300 (2d Cir. 4 Oct 1892); 57 F. 616 (1893) for the July 1891 date of the first judgment." sources: [src-47f454, src-52f300, src-57f616] nodes: - {node: lamp-1891-47f454-wallace-upholds-edison-lamp-patent, verb: supports} - {node: lamp-1892-10-04-appeals-court-affirms-edison-patent, verb: supports} basis: > # V: V2. Claim 1 was held not infringed. Expiry: 52 F. 300 holds the patent unaffected by the lapse of a Swedish patent (directly or through # Canadian patent No. 10,654 of 17 Nov 1879, whose 5-year term was extended by 10 on a fee paid 4 May 1883) and holds the # Commissioner's attempted "certificate of correction" limiting its term "wholly void". No document read states when the US patent # expired; the first version's "17 Nov 1894" is derived and unverified. # =========================== GERMANY: GÖBEL =========================== - id: lamp-goebel-claim state: contested statement: > Whether Heinrich (Henry) Göbel built practical carbon-filament lamps in New York in 1854, before Edison. No dated document from before 1882 that records such lamps has been found, on either side. On motions for preliminary injunctions, the Boston court (54 F. 678, Colt J.) found it "extremely improbable that Henry Goebel constructed a practical incandescent lamp in 1854" while saying the weight of evidence "is in the direction that he made" the three early "fiddle bow" lamps or similar ones, but "assuming they were made, they do not anticipate the invention of Edison. At most they were experimental toys"; and found it "simply impossible, under the circumstances, to believe that a lamp so constructed [lamp No. 4] could have been made by Goebel before 1872" (a finding on credibility). The Wisconsin court (57 F. 616, Seaman J.), and the Seventh Circuit on appeal in the same case (61 F. 834), held the new defence not proved on the motion (the Seventh Circuit saying it was "without designing to suggest what result should be reached upon final hearing"); the St. Louis court (56 F. 496, Hallett J.) found "a fair preponderance of testimony in support of the Goebel claim" and refused an injunction on condition of a $20,000 bond. That is two cases against one, on a preliminary record and under different burdens of proof. statement_kind: judgment evidence_class: primary_text confidence: low evidential_weight: 3 anchor_checked: primary adoption_weight: 3 adoption_note: > Taught in Germany for over a century (streets and schools named for Göbel; promoted under the Nazi state, per V0 sources). His hometown, Springe, is reported in the first version of this dig to call the question "debatable" (springe.de); that wording could not be found again on 2026-10-02 and is unverified. would_change_if: > a dated document from before 1879 (newspaper, advertisement, letter, patent caveat) describing Göbel's lamp is found, or a full trial record tests the affidavits positions: - {label: "Improbable or not proved", holder: "Colt J. (D. Mass.), Seaman J. (E.D. Wis.), Seventh Circuit", ground: "the state of the art, the lamps' defective vacuum, lamp No. 4's impossible date, and the rule that a new defence against an upheld patent needs convincing proof"} - {label: "A fair preponderance of testimony for Göbel", holder: "Hallett J. (E.D. Mo.)", ground: "affidavits of witnesses 'of good repute', recent test lamps 'reasonably effective', and a defence that 'puts the case in doubt' sufficing to refuse an injunction"} - {label: "No dated document before 1882 either way", holder: "this dig's reading", ground: "the earliest dated traces are Curtis (early 1882), Eaton (May 1882), Dreyer's option (1882) and the 28 Nov 1882 offer; Göbel's only lamp patent (1882) is a socket detail"} disputed_by: - {who: "Colt J. (54 F. 678)", kind: person, position: "Extremely improbable that Göbel built a practical lamp in 1854; lamps 1 to 3, if made, were experimental toys that do not anticipate Edison", source_read: "54 F. 678 (D. Mass., 18 Feb 1893)", via: "preliminary injunction on affidavits"} - {who: "Seaman J. (57 F. 616) and the Seventh Circuit (61 F. 834)", kind: institution, position: "The Göbel defence is not proved by convincing evidence; the story is surrounded by 'an atmosphere of improbability'", source_read: "57 F. 616 and 61 F. 834", via: "one case, preliminary injunction and appeal"} anchor: type: primary_text description: "54 F. 678 (D. Mass., 18 Feb 1893), Colt J.; 56 F. 496 (E.D. Mo., 21 Apr 1893), Hallett J.; 57 F. 616 (E.D. Wis., 20 Jul 1893), Seaman J.; 61 F. 834 (7th Cir., 1 May 1894); US 266,358 (1882), Göbel's lamp patent, a socket detail; US 47,632 (1865), Göbel's sewing-machine hemmer." sources: [src-54f678, src-56f496, src-57f616, src-61f834, src-us266358, src-us47632] nodes: - {node: lamp-1893-02-18-colt-rules-on-goebel-defence-boston, verb: disputes} - {node: lamp-1893-04-21-hallett-refuses-injunction-goebel-defence-st-louis, verb: supports} - {node: lamp-1893-07-20-seaman-grants-injunction-goebel-defence-wisconsin, verb: disputes} - {node: lamp-1894-05-01-seventh-circuit-affirms-wisconsin-goebel-ruling, verb: disputes} - {node: lamp-1882-early-curtis-hears-goebel-story, verb: qualifies} - {node: lamp-1882-05-eaton-told-of-goebel-claim, verb: qualifies} - {node: lamp-1882-dreyer-option-on-goebel-inventions, verb: qualifies} - {node: lamp-1882-10-24-goebel-patent-266358, verb: qualifies} - {node: lamp-1882-11-28-goebel-offer-to-edison-company, verb: qualifies} - {node: lamp-1865-05-09-goebel-sewing-machine-hemmer-patent, verb: qualifies} basis: > # V: V2 for all opinions and both patents. Changed from "refuted" in the revision of 2026-10-02, under the dig's single standard # (S2, S4): a refutation needs a dated document read directly that shows the claim false; the record has the court rulings and # no dated document either way before 1882. The rulings are findings on affidavits, preliminary and not a trial; the absence of a # pre-1882 document does not refute (the same reason the dig keeps de la Rue, Davy and Swan's 1850s as gaps). The first version also # counted "four courts" and "three of four"; the Seventh Circuit appeal is the same case as the Wisconsin ruling, so it is # two against one. About forty affidavits, five of them by Göbel's children and one by a niece, put his lamps between 1850 and # 1882 (Colt J.); the Edison side's affidavits deny lamp displays about 1860 and say lamps 1-3 were never practically operative. # Göbel's 1882 patent on a small lamp part sits badly with having invented the whole lamp 28 years earlier, but that is an inference. # Rohde (2007) reportedly found the false details came largely from the lawyers' presentation (V0; book not read). # The earliest dated traces are not the 28 Nov 1882 offer alone: Curtis first heard of Göbel "in the early part of 1882" (he made further # inquiries in 1892; the defence was put forward in the 1893 suits), Eaton was told in May 1882, and Dreyer took an option in 1882. - id: lamp-goebel-rohde state: searched_gap statement: "What Hans-Christian Rohde's 2007 study of the court record found about how the Göbel story was built." statement_kind: search_result evidence_class: synthetic confidence: low evidential_weight: 2 anchor_checked: no adoption_weight: 2 adoption_note: "Accepted in German history-of-technology writing (per V0 summaries); Springe has not retracted (not verified)." would_change_if: "the book is read" next_step: > Read Rohde, Die Göbel-Legende (zu Klampen, 2007), and Dittmann's review in Technikgeschichte 74(2), 2007, pp. 149-160 (paywalled). anchor: type: search-record description: "Rohde 2007, known only through summaries on lerncafe.de, goebel-projekt and radiomuseum.org, which say the court record does not support Göbel as inventor; the book was not read." sources: [src-rohde-2007] basis: "Synthetic class: a modern study of the record. Cannot serve as a primary refutation (Rule 9) and is not used as support for any claim here." # =========================== ATTRIBUTION MYTHS =========================== - id: lamp-latimer-patent state: established statement: > Lewis Latimer's US patent 252,386, "Process of Manufacturing Carbons" (filed 19 February 1881, granted 17 January 1882), improves how carbon filaments are made: blanks of paper or wood are carbonised inside cardboard envelopes so that they are protected from sticking, breaking and warping. It was assigned to the United States Electric Lighting Company of New York. With J. V. Nichols he also held US patent 247,097, "Electric Lamp" (filed 18 April 1881, dated 13 September 1881), for mounting the carbon strip on its wires. statement_kind: reported evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 3 adoption_note: "Often retold as 'Latimer invented the carbon filament' or 'improved Edison's bulb'." would_change_if: "the patent record is misread" anchor: type: primary_text description: "US 252,386 and US 247,097, both read in OCR of the USPTO scans (claims and assignment lines)." sources: [src-us252386, src-us247097] nodes: - {node: lamp-1882-01-17-latimer-patent-252386, verb: supports} - {node: lamp-1881-09-13-nichols-latimer-patent-247097, verb: supports} basis: > # V: V2. The first version called the company "Hiram Maxim's firm and a rival of Edison's" and anchored that to the patent; the # patent names only the United States Electric Lighting Company. That Maxim was its chief engineer, that it was one of Edison's # competitors and that Latimer worked for it from March 1880 (in Bridgeport, Connecticut, before the company moved to Brooklyn) # come from the Queens Library essay hosted by the Edison Papers (secondary, read in the revision; node # lamp-1880-03-latimer-joins-us-electric-lighting-bridgeport). A company of the same name was the defendant in the Edison company's suit # upheld in 47 F. 454 and 52 F. 300 (identity assumed from the name). His later work for Edison's companies is in # lamp-latimer-edison-role. - id: lamp-latimer-edison-role state: searched_gap statement: "Latimer's later role with the Edison companies (engineering staff, patent expert witness, Edison Pioneers)." statement_kind: search_result evidence_class: synthetic confidence: low evidential_weight: 2 anchor_checked: secondary adoption_weight: 3 adoption_note: "Widely stated; well supported in secondary accounts." would_change_if: "Edison company records or court testimony naming him are read" next_step: "Read the Latimer papers (Queens Borough Public Library) and the Edison Papers for Latimer, 1884-1912." anchor: type: search-record description: "The Queens Borough Public Library essay 'Inventing a Better Life: Latimer's Technical Career', hosted by the Edison Papers (secondary, read 2026-10-02): from about 1885 Latimer was in the Engineering Department of the Edison Electric Light Company, from about 1889 in its Legal Department, testified in a number of cases, and was a founding Edison Pioneer in 1918. The essay cites records that were not read." sources: [src-edison-papers-latimer] basis: > Gap, not doubt: the secondary record is consistent. The first version cited the Edison Papers essay 'The Carbon-Filament Lamp' (src-edison-papers-essay) for this, which does not mention Latimer; the source is now the Queens Library essay, and the claim stays at anchor_checked secondary because the underlying records (Latimer's logbook, company records, court testimony) were not read. - id: lamp-latimer-invented-claim state: refuted refutes_target: "Lewis Latimer invented the light bulb or the carbon filament" refutation_class: misattribution statement: > Latimer did not invent the light bulb or the carbon filament. His 1882 patent is a manufacturing improvement filed in 1881, after carbon filaments had been patented by Edison (applied for November 1879, granted 1880) and carbon in a vacuum by King (1845); his other 1881 patent with Nichols is for mounting the carbon on the wires. His real contribution was making carbons more reliably, first for a rival company. statement_kind: judgment evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 3 adoption_note: "A modern corrective story that overshoots; it grows out of real neglect of Latimer's work." would_change_if: "an earlier Latimer document claiming the filament or lamp itself is found, or US 247,097 or 252,386 is read as claiming the lamp" anchor: type: primary_text description: "US 252,386 (1882) and US 247,097 (1881) against US 223,898 (1880) and the 1845 King patent (abridgment, p. 97)." sources: [src-us252386, src-us247097, src-us223898, src-abridgments-1859] nodes: - {node: lamp-1882-01-17-latimer-patent-252386, verb: supports} - {node: lamp-1881-09-13-nichols-latimer-patent-247097, verb: supports} - {node: lamp-1880-01-27-edison-patent-223898-granted, verb: supports} basis: > The attribution moves from a process patent to the whole invention. Recognition of his actual work stands. "A rival company" rests on the Queens Library essay (secondary) and on the company's name in the Edison suits. - id: lamp-10000-ways-quote state: refuted refutes_target: "Edison said 'I have not failed, I've just found 10,000 ways that won't work' about the light bulb" refutation_class: narrative_drift statement: > The earliest version found is in Dyer and Martin's 1910 biography, where Walter Mallory recalls Edison saying "I know several thousand things that won't work" about the nickel-iron storage battery after "over nine thousand experiments", not about the lamp; Mallory adds that the experiments later "had run up to over ten thousand". The "10,000 ways" wording has not been found in anything earlier than the modern forms traced by Quote Investigator (1946, 1982). statement_kind: search_result evidence_class: primary_text confidence: provisional confidence_reasons: start: high start_because: "The 1910 text was read directly in a scan of the book (anchor checked: primary)." steps: - {domain: unreported_negative_results, effect: down1, reason: "The claim that the 10,000-ways wording about the lamp is not documented earlier rests on not finding it. The dig has not searched Edison's own papers, newspapers or letters of his lifetime, so absence cannot be shown."} - {domain: indirectness, effect: down1, reason: "The 1910 text records Mallory's recollection, written down by Edison's authorised biographers; it is not Edison's own record and it says nothing about what he said about the lamp."} - {domain: risk_of_bias, effect: down1, reason: "One recollection in an authorised biography is the only anchor; Quote Investigator, a secondary tracing, is kept out of the anchors."} reviewed: false evidential_weight: 3 anchor_checked: primary absence_anchor: true adoption_weight: 5 adoption_note: "One of the most repeated quotations in English; nearly always attached to the light bulb." would_change_if: "an earlier printed or manuscript version about the lamp is found, or Edison's own record of the remark is found" anchor: type: primary_text description: 'Dyer and Martin, Edison: His Life and Inventions (Harper, 1910), vol. II, pp. 615-616, read in OCR of the Internet Archive scan.' sources: [src-dyer-martin-1910] basis: > # V: V2 for the 1910 text, the only anchor. Quote Investigator (secondary) traces the later "999" (1946) and "10,000" (1982) # forms; that history is in transmission.yaml (tx-10000-ways), not here, and is not used for the refutation. # Quote Investigator (secondary, read) also reports a 1921 American Magazine interview in which Edison recounts the incident without a number # and an 1882 Christian Union article by Edison ("the enormous number of things that won't work"); neither was read directly; both are in the # transmission chain, and the claim says only "earliest version found", not that Edison never spoke of it. # Reworded in the revision: the first version said the wording was "not documented in Edison's lifetime". The 1910 book was itself # published in Edison's lifetime (he died in 1931), and one book cannot show what was or was not said or printed about the lamp in the # other twenty-odd years or earlier. The claim now says only "earliest version found" and is held at provisional confidence as an # absence claim. # =========================== THE HEADLINE =========================== - id: lamp-edison-sole-inventor state: refuted refutes_target: "Thomas Edison invented the light bulb (as its first or only inventor)" refutation_class: narrative_drift statement: > Edison did not invent the incandescent lamp from nothing and was not its first or only inventor. British patents of 1841 (de Moleyns) and 1845 (King, for Starr) already described, respectively, an electric light fed with powdered charcoal in an exhausted globe and a heated carbon conductor in a vacuum; Woodward patented a carbon lamp in rarefied gas in 1874-76; and Swan reported a glowing carbon rod in an exhausted globe in December 1878, a rod that blackened its globe. US courts upheld the second claim of Edison's own patent (1880), a carbon filament in an all-glass exhausted receiver (claim 1, the high-resistance filament, was held limited and not infringed); in Wallace J.'s words Edison made "a lamp which was practically operative and successful" (47 F. 454, 1891), and Colt J. records that the earlier ruling held claim 2 to cover "a broad and fundamental invention, namely, an incandescent lamp, composed of a carbon filament, hermetically sealed in an all glass chamber exhausted to a practically perfect vacuum, and having leading-in wires of platinum" (54 F. 678). Whether Edison or Swan was first to a practical lamp, and a complete wired system (Pearl Street, 1882), are separate claims (lamp-first-practical-priority, lamp-pearl-street). statement_kind: judgment evidence_class: primary_text confidence: high evidential_weight: 5 anchor_checked: primary adoption_weight: 5 adoption_note: "The default answer in the US and much of the world; reinforced by schoolbooks and the quote myth." would_change_if: > the 1845 patent is shown not to describe a heated carbon conductor in a vacuum (or the 1841 patent not to describe a light in an exhausted globe), or the courts are shown to have upheld Edison's patent as the first lamp of any kind anchor: type: primary_text description: 'British patents 9053 (1841) and 10,919 (1845), Patent Office abridgments; Chemical News 39 (1879): Swan''s December 1878 rod lamp; US 181,613 (Woodward); 47 F. 454 (1891) and 52 F. 300 (1892); 54 F. 678 (1893), Colt J.; US 223,898.' sources: [src-abridgments-1859, src-chemnews-1879, src-us181613, src-47f454, src-52f300, src-54f678, src-us223898] nodes: - {node: lamp-1841-08-21-de-moleyns-patent-9053, verb: supports} - {node: lamp-1845-11-04-king-starr-patent-10919, verb: supports} - {node: lamp-1878-12-swan-rod-lamp-newcastle-chemical-society, verb: supports} - {node: lamp-1891-47f454-wallace-upholds-edison-lamp-patent, verb: qualifies} basis: > The true core is large and is kept in lamp-edison-patent, lamp-edison-high-resistance, lamp-edison-patent-upheld and lamp-pearl-street. The drift is from "made a lamp the courts called practical, and a system to run it" to "invented it". Under the dig's standard (S2, S3) the refutation rests on dated documents (patents, a society report, a court opinion) that show earlier descriptions and an earlier reported experiment, not on any claim that a lamp was built before Edison's beyond Swan's reported rod. The first version also said lamps were "built" by Lodygin and by Woodward and Evans before Edison, and that Edison's team "made first a lamp practical to manufacture and sell ... by 1882"; neither was anchored to this claim and both are removed (Lodygin: lamp-lodygin-1872-74; Woodward: lamp-woodward-evans-patent, lamp-woodward-evans-performance; the practical lamp and sale: lamp-first-practical-priority; Pearl Street: lamp-pearl-street).