Marconi and the Maxwellians: the Origins of Wireless Telegraphy Revisited Author(S): Sungook Hong Source: Technology and Culture, Vol

Home , Adolf Slaby, John Ambrose Fleming, Samuel Langley

Marconi and the Maxwellians: The Origins of Wireless Telegraphy Revisited Author(s): Sungook Hong Source: Technology and Culture, Vol. 35, No. 4 (Oct., 1994), pp. 717-749 Published by: The Johns Hopkins University Press and the Society for the History of Technology Stable URL: http://www.jstor.org/stable/3106504 . Accessed: 20/02/2015 20:40

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

The Johns Hopkins University Press and Society for the History of Technology are collaborating with JSTOR to digitize, preserve and extend access to Technology and Culture.

http://www.jstor.org

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions Marconiand theMaxwellians: The Originsof Wireless Telegraphy Revisited

SUNGOOK HONG

The point is which of the two was the firstto send a wirelesstelegram? Was it Lodge in 1894 or Marconi in 1896? [SILVANUSP. THOMPSON, London Times, July 15, 1902]

We live in a world where technologicalpriority disputes and patent litigationare so commonplace that only a spectacular case, such as Kodak versusPolaroid over the instantcamera, attracts our attention.In the past two hundred years,such disputes have become increasingly frequent.Notable examples include those over the inventionof spin- ning machines (John Hargreavesvs. RichardArkwright), steelmaking (Henry Bessemer vs. William Kelly), the incandescentlamp (Thomas Edison vs. Joseph Swan), the telephone (Alexander Graham Bell vs. Elisha Gray),the airplane (theWright brothers vs. Samuel Langley),and amplifiersand the heterodyneprinciple in radio (Lee De Forest vs. Edwin Howard Armstrong). Historiansof technology,however, have generallypaid littleattention to the conflictingpriority claims themselves,except when priorityand patent disputescan be used as a windowthrough which the character- istics of the evolution of technologyare analyzed.' There are two well-groundedreasons for this neglect.First, unlike scientificdiscover-

DR. HONG received his Ph.D. from Seoul National Universitywith the dissertation "Forgingthe Scientist-Engineer:A ProfessionalCareer ofJohn Ambrose Fleming" and is workingon the science-technologyrelationship in powerand earlyradio engineering.He thanksJedBuchwald, Bert Hall, Bruce Hunt,JanisLangins, and the Technologyand Culture refereesfor theirvaluable comments.He is indebted to ProfessorThad Trenn of the Universityof Toronto and Roy Rodwell of the Marconi CompanyArchives for their help withthe archivesquoted here, and he thanksYoungran Jo, ShinkyuYang, Jane Jenkins, Andre Leblanc, and Ben Olshin for theirassistance, as well as the Instituteof Electrical and ElectronicsEngineers Fellowship in ElectricalHistory for facilitating the research. 'See, e.g., the important research of David E. Hounshell, "Elisha Gray and the Telephone: On the Disadvantageof Being an Expert," Technologyand Culture16 (1975): 133-61; Robert C. Post, "StraySparks from the Induction Coil: The Volta Prize and the Page Patent,"Proceedings of the Institute ofElectrical and ElectronicsEngineers (IEEE) 64 (1976): 1279-86; JamesE. Brittain,"The Introductionof the Loading Coil: GeorgeA. Campbell ? 1994 by the Societyfor the Historyof Technology.All rightsreserved. 0040-165X/94/3504-0004$01.00 717

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 718 SungookHong ies, prioritydisputes in technologyoften develop into patentlitigation, which ultimatelyinvolves judicial decisions or interferencesby the PatentOffice. These court decisions act like a forcedjudgment on the question of priority."Closure" of the controversy(to use the social constructivist'sterm) is not broughtabout by negotiationamong the engineers involved,but ratherby external,compulsory forces. These courtjudgments, which historianscannot overruleand which funda- mentallydetermine future histories, sometimes differ from those based on detailed historical analysis. Historians thereforetreat historical assessmentsof inventionsas a sphere separate from legal decisions about patentsand avoid enteringinto the latterrealm.' Second, and more important,historians of technology have usually considered inventionas a long-term,social process,which includes not only the creativeactivity of an inventorbut also historicallyaccumulated tradi- tionsin whichthe workof manypeople is merged."The prioritydispute is itselfinterpreted as evidencefor regarding the inventionas something sociallyconditioned. The question,for example, of who firstinvented wireless telegraphyis hardly meaningfulfrom such a perspective, because "wirelesstelegraphy" itself did not burstinto being as a resultof a single genius's efforts,but was graduallyshaped as several different technologicaltraditions converged. Recent historicalstudies on the originof radio reflectsuch a shiftof emphases in the interpretationof technologicalinventions. In a highly influentialmonograph, Syntonyand Spark:The Originsof Radio, Hugh G. J.Aitken argues thatwireless telegraphy cannot be said to have been

and Michael I. Pupin," Technologyand Culture11 (1970): 36-57. Hounshell contraststhe amateurishstyle of invention(Alexander Graham Bell) withthe professionalstyle (Elisha Gray), arguing for the former'sadvantage, whereas Brittaincompares the organized scientificresearch of Campbell withan independentinventor, Pupin. Post examines how the "notorious Page patent" on the induction coil was constructedin the name of "scientificchauvinism" and exploited by the corporateinterest. 'Compare Brittain (n. 1 above) with Joseph Gray Jackson, "Patent Interference Proceedingsand Priorityof Invention,"Technology and Culture11 (1970): 598-600. See also Thomas Hughes's analysisof Lucien Gaulard andJohn D. Gibbs's (who were defeated byS. Z. de Ferrantiin litigation)pioneering works on alternatingcurrent transformers; in Thomas P. Hughes, NetworksofPower: Electrification in Western Society, 1880-1930 (Baltimore, 1983), pp. 86-96. 3LynnWhite, jr., "The Act of Invention:Causes, Contexts,Continuities, and Conse- quences," Technologyand Culture3 (1962): 486-500; Maurice Daumas, introductionto A Historyof Technologyand Invention(New York, 1979), 3:1-15; Hugh G.J. Aitken, The ContinuousWave: Technology and AmericanRadio, 1900-1932 (Princeton,N.J., 1985), pp. 14 ff.;George Basalla, TheEvolution of Technology(Cambridge, 1988); John Law, "Theory and Narrativein the Historyof Technology:Response," Technologyand Culture32 (1991): 377-84.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 719 inventedby Guglielmo Marconi (1874-1937). He argues instead that WilliamCrookes had conceivedof Hertzianwave telegraphyin 1892 and that Oliver Lodge (1851-1940) demonstratedthis before the British Associationat its annual meetingin Oxford in 1894, one or twoyears before Marconi. In a criticalpassage, Aitken remarks,"Did Lodge in 1894 suggestin public thathis equipment could be used forsignalling? Did his lecturerefer to the applicationof Hertzianwaves to telegraphy? Did he demonstratetransmission and reception of Morse Code? The answerwould seem to be affirmativein each case. In thissense Lodge mustbe recognizedas the inventorof radio telegraphy."'This interpretation is quite novel and revisionist,since, beforeAitken, Marconi had usuallybeen regarded as the firstto inventwireless telegraphy.5 In thisarticle, I shall take up the prioritydispute between Marconi and Lodge over the inventionof wirelesstelegraphy. My analysiswill show that any claim for Lodge's priorityis incorrect.But my main purpose is not to argue instead for Marconi's priority.It is ratherto deconstructthe Lodge versusMarconi debate to reveal how twototally differentdiscourses (as noted in this article'sepigraph fromSilvanus Thompson) firstcame into being and how these were then reinforced by the differentinterests involved. After beginning withAitken's evidence, I then turnto whatwas later claimed as Lodge's demonstration

4Hugh G.J. Aitken,Syntony and Spark: The Originsof Radio (New York, 1976; 2d ed. Princeton,N.J., 1985), p. 123. One reviewerof the second edition of Syntonyand Spark noticed thispoint; see A. N. Strangesin AmericanHistorical Review 91 (1986): 1166-67. 5For claims supportiveof Marconi's priority,see Charles Sfisskind,"Popov and the Beginning of Radiotelegraphy,"Proceedings of theInstitute of Radio Engineers50 (1962): 2036-47, "The Early Historyof Electronics.III. Prehistoryof Radiotelegraphy,"IEE Spectrum6 (April1969): 69-74, and "The EarlyHistory of Electronics.IV. FirstRadioteleg- raphyExperiments," IEEE Spectrum6 (August 1969): 66-70. Aitken'sclaim for Lodge's prioritywas not unprecedented.W. P.Jolly, who has writtenbiographies of both Lodge and Marconi,admitted Lodge's wirelesstelegraphy at the BritishAssociation meeting in 1894. CompareW. P.Jolly, Marconi (London, 1972), pp. 41-42, withhis Sir OliverLodge (London, 1974), p. 97. AfterAitken,however, Lodge's prioritywas widely accepted. A recentbiography of Lodge emphasizesLodge's "radio transmission"in 1894, based on Aitken'saccount and Lodge's own; see Peter Rowlands,Oliver Lodge and theLiverpool Physical Society (Liverpool, 1990), pp. 115-23. RowlandF. Pocock, TheEarly British Radio Industry (Manchester, 1988), thoughadmitting Marconi's originality, mentions Lodge's radio transmissionin the Oxford lecturein 1894, on p. 82. G. A. Isted,a formerassistant to Marconi,has latelywritten that Lodge's demonstrationat theBritish Association in Oxford"is theearliest recorded instance of the transmissionand receptionof a signal by Hertzianwaves and it is clearlyof great historicalimportance." See G. A. Isted,"Guglielmo Marconi and the Historyof Radio: Part I," GeneralElectricCompany Review 7, no. 1 (1991): 45-56 (esp. on 46). Aitken'sargument is also picked up by Basalla (n. 3 above), p. 99. I should mentionhere thatmy criticism of Aitkenis restrictedto theorigin of wireless telegraphy with reference to Lodge and Marconi. My work is much indebted to Aitken'svaluable analysison the interactionof scientific, technological,and economic factorsin the earlystage of wirelesstelegraphy.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 720 SungookHong of wirelesstelegraphy in 1894. It willbe shownthat this had nothingto do with telegraphy,nor with alphabetic signals, nor with dots and dashes. I then turnto the impact of Marconi and his Britishpatent on the British Maxwellian physicists--inparticular Lodge, Thompson (1851-1916), George F. FitzGerald (1851-1901), and John Ambrose Fleming (1849-1945).6 The transformationfrom Hertzian laboratory apparatusinto commercialwireless telegraphy was in factaccomplished by Marconi, an Italian "practician."A certain disharmonybetween theory and practice became apparent. Moreover, Marconi's patent appeared so strongthat it threatenedto monopolize Hertzianwaves and the Britishnational interest.Under these circumstances,the image of Lodge as the inventor of wireless telegraphywas deliberatelycon- structedby his friendsand by Lodge himself. This studyclarifies not only the origin of wirelesstelegraphy with special reference to Marconi and Lodge but also the interaction between theoryand practice in early radio history.It shows that two differentdiscourses on the theory and practice in early wireless telegraphy--discourseswhich either emphasized the influence of science on technologyor denied any relationshipbetween them--were constructedby different groups of participants.'My study also illustrates the wayin which the historical"facts" are at timesconstructed, as well as the wayin which these factsare analyzedby carefullycross-checking the sources. My ultimate hope is that this article will contributeto rehabilitatingthe prioritydispute as an object of historicalresearch.8

FIemingsMarconi Memorial Lecture in 1937 Aitkenhas criticallyexamined varioussources concerningthe Lodge versusMarconi priorityissue. Besides Lodge's own recollections,Aitken bases his conclusionson twoother sources. The firstis a shortarticle in

'For the livesand worksof the BritishMaxwellian physicists, see Jed Z. Buchwald,Fmm Maxwellto Microphysics (Chicago, 1985); and BruceJ. Hunt, TheMaxwellians (Ithaca, N. Y., 1991). 7Forthe relationbetween science and technologyin earlywireless telegraphy, see the analysisof Hugh G.J. Aitken,"Science, Technologyand Economics: The Inventionof Radio as a Case Study,"in TheDynamics of Science and Technology,ed. W. Krohn,Edwin T. Layton,Jr., and Peter Weingart(Dordrecht, 1978), pp. 89-111. I have examined the theoryand practiceissue in myforthcoming paper "From Hertz to Marconi'sTelegraphy: The Laboratoryand the Field in EarlyWireless Experiments, 1888-1896." 8Patentrecords and patent interferencesas sources for historicalresearch have been pointed out by N. Reingold, "U.S. Patent OfficeRecords as Sources for the Historyof Inventionand Technological Property,"Technology and Culture1 (1959/60): 156-67; and SeymourL. Chapin, "Patent Interferencesand the Historyof Technology:A High-flying Example," Technologyand Culture12 (1971): 414-46. See also Hounshell (n. 1 above); Post (n. 1 above); and Brittain(n. 1 above).

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 721 the Electricianof 1897, which stated that "both at Oxford [in August 1894] and at the Royal Institution[in June 1894], Dr. Lodge described and exhibited publicly in operation a combination of sending and receivingapparatus constituting a systemof telegraphysubstantially the same as thatnow claimed in the patentwe have referredto [Marconi's patent no. 12,039 of 1896]."' Aitken's second source is Fleming's Marconi memoriallecture in 1937. Flemingsaid thatMarconi was "not the firstperson to transmitalphabetic signals by electromagnetic waves." He instead admittedLodge's priority:

[Lodge] was able to transmita dot or a dash signaland by suitable combinationsto send any letterof the alphabet on the Morse code and consequentlyintelligible messages. He had also on his table a Morse inker(so he tellsme), and could have used it witha sensitive relayto printdown the signals,but as he wishedthe audience to see the actual signalshe preferredto use the mirrorgalvanometer. It is, therefore,unquestionable that on the occasion of his Oxford lecture in September [sic], 1894, Lodge exhibited electric wave telegraphyover a shortdistance."'

Since the testimonywas given by Fleming it seems trulyconclusive. Before Marconi's arrivalin England in 1896, Fleming and Lodge had been close friends,having studied togetherin theiryouth at Edward Frankland'slaboratory in South Kensington.Their relationshipdeterio- rated rapidly,however, after Fleming became scientificadvisor of the Marconi Company (at that time WirelessTelegraph and Signal Com- pany) in 1899. In all the yearsafterward, right up until 1937, Fleming had never admitted Lodge's priority,reiterating that "here [at the Oxfordmeeting] again no mentionof the applicationof thesewaves to telegraphywas made."" Only after Marconi's death, it seems, did Fleming decide to tell the truth.Aitken comments that "Fleming's memoryalso was capable of improvementwith the passage of time,or perhaps as commercialand scientificrivalries receded into the past.""

9"Dr. Oliver Lodge's Apparatusfor Wireless Telegraphy," Electrician 39 (1897): 686-87. Also quoted in Aitken,Syntony and Spark(n. 4 above), p. 122. 'John Ambrose Fleming, "Guglielmo Marconi and the Development of Radio- Communication,"Journal of the Society of Arts 86 (1937): 42-63 (quoted on 46); cited in Aitken,Syntony and Spark,p. 123. Aitkenalso recognized (on p. 174, n. 70) thatthe phrase of "it is, therefore,unquestionable" is changed to read "it is, therefore,questionable" in Degna Marconi,My Father Marconi, 2d ed. (Ottawa, 1982), p. 21. Flemingslipped the date. The BritishAssociation annual meetingwas held at Oxfordin August 1894, and Lodge's experimentswere done on August 14. "JohnAmbrose Fleming, The Principles ofElectric Wave Telegraphy (London, 1906), p. 424; Aitken,Syntony and Spark,p. 120. 2"Fleming,"Guglielmo Marconi" (n. 10 above), p. 42; Aitken,Syntony and Spark,p. 122.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 722 SungookHong But one thingshould be made clear. Flemingwas not presentat the Oxford BritishAssociation meeting in August 1894, the most crucial eventin the discussion.Fleming's source was not his own memory,but Lodge's remark. That is clearly revealed by three letters between Fleming and Lodge in 1937. Before his lecture in November 1937, Flemingwrote to Lodge:

I have been asked bythe Council of the RoyalSociety of Arts to give next November 10th a Memorial Lecture on the "Work of Mar- coni." ... One of the factsI am anxious to learn about is whether in yourlecture to the BritishAssociation at OxfordMeeting in 1894 you used a telegraphicrelay in series with your coherer to print on Morse Inker tape dot and dash signals? In his little book on "Wireless"Dr. Eccles giveson page 54 a diagramof the apparatus he says you employed at Oxford in 1894 [see fig. 1].... I was presentin June 1894 at yourfamous lecture at the RoyalInstitution on "The Workof Hertz"and rememberwell your experiments with yourcoherer. But to the best of myrecollection there was no direct referenceto "telegraphy"in thatlecture. I was not presentat the B.A. meetingat Oxford,but ... it is veryimportant to knowfrom you whetherat Oxford in 1894 you exhibited a Hertz oscillator connectedwith coherer and used a telegraphicrelay in connection withit and morse inkerand showed the transmissionand printing of dotsand dashsignals over any shortdistance.13

Lodge replied that at Oxford he had actually used telegraphic instrumentsand transmittedalphabetic signals, that is, dots and dashes:

You are perfectlyright that in 1894 at the Royal InstitutionI did not refer to telegraphy.But, stimulatedby Muirhead, who had close connection with telegraphyand cables, I did at Oxford demonstrateactual telegraphy.I had a Morse instrumentthere, but itwas not convenientfor the largeaudience in the Museum theatre, and thereforeI used as receiver a Thomson marine signalling device supplied by Muirhead'sfirm for that purpose, thoughI had a Morse instrumenton the table whichI could have used instead. But the deflectionsof the spot of lightwere plainlyvisible to the audience, and gave quick and prolonged response corresponding to the dots and dashes accordingto the manipulationof the keyat the distantend.'4

"JohnAmbrose Fleming to OliverLodge, August24, 1937,Lodge Collection,University College London (hereafterUCL) (emphasis in original). W.H. Eccles's book is titled Wireless(London, 1933). "4Lodgeto Fleming,August 26, 1937 (copy), Lodge Collection,UCL.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 723

BATTERY BATTERY

COHERER RELAY

TREMBLER

FIG.1.-W. H. Eccles's diagramof OliverLodge's receiverin 1894. (W H. Eccles, Wireless [London, 1933], p. 54.)

Fleming'sreply to Lodge, which foretoldthe contentof his lecture, showsthat he entirelyaccepted Lodge's claims:"What you tellme about your Oxford lecturein 1894 is veryvaluable and important.It is quite clear thatin 1894 you could send and receivealphabetic signals in Morse Code byElectric Waves and did send them180 feetor so. Marconi'sidea that he was the firstto do that is invalid..... Marconi was always determinedto claim everythingfor himself.His conduct to me about the firsttransatlantic transmission was veryungenerous.... However, these thingsget knownin time and justice is done."'5 As we can see in thislast letter,Fleming had been hurtby Marconi's attitudetoward his employees.His resignationas scientificadvisor to the Marconi Company in 1931 and Marconi's death in 1937 mighthave influencedFleming to be more sympatheticto Lodge. He mighthave feltas if "thingsget knownin time and justice is done." But thiscould not have improvedhis memoryof somethinghe had neverexperienced. It was only Lodge who informedFleming about the Oxford meeting. Therefore,Fleming's lecture in 1937 cannot be regardedas conclusive. For later analysis,I divide Lodge's claim into two parts.First, Lodge actuallysent telegraphicsignals, that is, dots and dashes, during the Oxford meeting of the BritishAssociation in August 1894. Second, Lodge had a Morse instrumentthere, but, owing to the size of the

'5Flemingto Lodge, August29, 1937, Lodge Collection,UCL (emphasis in original).

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 724 SungookHong audience, he used a mirrorgalvanometer to show the signals.Leaving aside Aitken'sfirst evidence (a shortarticle in theElectrician) until a later sectionof thisarticle on Marconi'spatent, I shall examine Lodge's Royal Institutionand BritishAssociation lectures in 1894. We willsee thatboth assertionsare incorrect.

LodgeSExperiments with Hertzian Waves Oliver Lodge, an ambitiousMaxwellian and professorof physicsat UniversityCollege, Liverpool,worked on the various characteristicsof Hertzianwaves between 1888 and 1894. The linksbetween optics and electromagnetismparticularly attracted him. The subject was faithfully Maxwellian,as it had a root in Maxwell'sdoctrine that lightand electromagneticwaves were the same. It was also trulyLodgian "imperial science,"as itled electricalscience to theconquest of otherfields-in this case, opticsand physiology.The subjectcame to be dividedinto two parts: first,the physicalinvestigation of the quasi-opticalproperty of electromagneticwaves-that is, reflection,refraction, and polarizationof the electromagneticwaves in air,in othermedia, and in some cases along the wires;second, the physiologicalinvestigation of the mechanismof the perceptionof light(color, intensity, and so on) by human eyes.'" Withthese experiments,Lodge made two importantadvances. First, he constructeda radiator that generated waves with wavelengthsof severalinches. Hertz had once used the wavelengthof 66 centimeters, but thatwas stilltoo long formost optical experiments.Because of the difficultyin decreasing the wavelengthwith Hertz's dipole radiator, Lodge turned to a spherical radiator. In 1890, Lodge used three 12-centimeterballs and obtained 17-centimeterwaves, "the shortest yet dealt with."'7Lodge then went furtheralong this line of development and devised twomore sphericalradiators that he exhibitedin his FridayEvening Lecture on the "Workof Hertz" at the Royal Institution in June 1894. Lodge's second line of researchwas on detectors.The Hertzianwave was at firstdetected by a small spark-gapresonator. But thisspark-gap

'"ForLodge's earlyconceptions of electromagneticwaves, see Jed Z. Buchwald,"Wave Guides and Radiators in MaxwellianElectrodynamics," published as app. 1 to his The Creationof Scientific Effects: Heinrich Hertz and ElectricWaves (Chicago, 1994). See also Hunt, TheMaxwellians (n. 6 above), pp. 24-47. Lodge's researchafter 1888 is best describedin Aitken,Syntony and Spark(n. 4 above), pp. 80-102. Lodge's programwith Hertzian waves, as well as his concept of "imperial science," was promulgatedin Oliver Lodge, Modern Viewsof Electricity (London, 1889), pp. 303-7. For the earlyquasi-optical experiments with Hertzian waves, see John F. Ramsay,"Microwave Antenna and Waveguide Technique before 1900, " Proceedingsof the IRE 46 (1958): 405-15. '7OliverLodge, "ElectricRadiation fromConducting Spheres, an ElectricEye, and a Suggestionregarding Vision," Nature 41 (1890): 462-63.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 725 resonator was not suitable for Lodge's physiologicalresearch. For example, nothingin the spark-gapdetector corresponded to different color perceptionsin human eyes.Lodge, therefore,concentrated on the constructionof an "electriceye." In 1890, his assistantEdward Robinson constructeda "gradated receiver,"and Lodge tried "a series of long cylinders"of variousdiameters. The principleof both detectorswas to make each of themrespond to a specificradiation, forming "an electric eye witha definiterange of colour sensation."In 1891 Lodge exhibited an electriceye of Robinson'stype at the PhysicalSociety, London, which had "stripsof tin foil of differentlengths attached to a glass plate, and sparkgaps at each end whichseparate them from other pieces of foil."' Yet, it was not the electric-eyeresonator that was associatedwith the name of Lodge. Rather, it was the coherer. To understand Lodge's coherer, we need to examine its prehistorybriefly. In 1890, while experimentingon lightning rods, Lodge found that two metallic conductors separated by a very tiny air gap were fused when the oscillatorydischarge passed throughthem. At thattime, Lodge accepted David Hughes's explanationthat this was a thermoelectricphenomenon and dropped the subject.In 1890, Edouard Branlyin France found that fine copper filings,capsuled into a glass tube, were conductingonly feebly under ordinary conditions but that their conductivitywas abruptlyincreased when a sparkwas generatednearby. Branly's tube was introduced in Britain when the Electricianfully translated his articles withfigures, but theywere apparentlyoverlooked at thattime. The tube was noticed later,however, by Dawson Turner,who demonstratedthe decrease of the resistanceof copper filingsat the BritishAssociation meeting in Edinburgh in 1892. Turner's demonstrationwas seen by W. B. Croft,who addressed a shortexperiment on the same phenom- enon before the Physical Society,London, in October 1893. There, George M. Minchin,one of those interestedin Hertzianwaves, noticed the similaritybetween Croft's (actually Branly's) tube and his solar cell's response to Hertzianwaves. Minchin immediatelyread a paper on the subject at the PhysicalSociety. While hearing Minchin's paper, Lodge noticed that Branly'sand Minchin's discoverywas very similarto his previous research on the action of lightningdischarge to a very tiny metallicgap. Lodge reasoned thatelectromagnetic radiation made the metallic molecules both in the filingsand in the microscopicair gap actuallycohere withone another.Based on this similarity,Lodge soon devised a single-pointcontact "coherer,"in whicha springwire formed a slight contact with an aluminum plate, and soon found that its

'sIbid.; and Oliver Lodge, "Some Experimentswith Leyden Jars" (abstract),Nature 43 (1891): 238-39.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 726 SungookHong sensitivityas a detector was not only much better than ordinary spark-gapresonators but also betterthan Branly'sfiling tube."9 Lodge thereforehad two new detectors,his coherer and Branly's tube. Initially,Lodge called onlyhis single-pointdetector a coherer,but the name "coherer" soon came to designateboth types.Both Lodge's coherer and Branly'stube were connected in series to a batteryand a galvanometer.Under this condition, they act like an on-offswitch: before a Hertzianwave strikesthem, their resistances are veryhigh, as if the switchwere off,but when a Hertzian wave strikesthem, their resistancesfall off, as ifthe switchwere turned on. This actionmakes the currentflow from a battery,and the currentcan be detected by a galvanometer.The two detectors,however, differed in sensitivity.At Liverpool on April 17, 1894, Lodge found that the filingtube could detect radiation emittedfrom 40 yards away. However, "a sender in ZoologyTheatre affectedthe cohererin PhysicsTheatre perceptibly,"a distance of perhaps 70 yards.20Though more sensitive,Lodge's coherer was less stable than the filingtube. In addition, Branly'stube had a crude metrical character: its decrease in resistance seemed roughly proportionalto the intensityof the Hertzian waves.This resembled a human eye's perception of the light of differentintensity. For physi- ological experiments,therefore, Branly's tube was more suitable than Lodge's single-pointcontact coherer. On January1, 1894, Hertz died at the age of 36, and Lodge delivered a Hertz Memorial Lecture at the Royal InstitutionFriday Evening Lecture on June 1. Here Lodge spoke on the life and work of Hertz, exhibited Hertz's and his own radiators and detectors, and then performedseveral experiments.21 The demonstrationswere dividedinto a physicaland a physiologicalpart. In the firstpart, he demonstrated reflection,refraction, and polarizationof the Hertzianwaves. For this purpose, Lodge used his sphericalradiator enclosed in a metallicbox

'9For Lodge's experimentson the air gap of lightningconductors, see OliverLodge, "On Lightning-Guardsfor Telegraphic Purposes and on the Protectionof Cables fromLight- ning,"Journal of the Institution of Electrical Engineers 19 (1890): 346-79, on 352-53. For the storyof Branly'stube in Britain,see OliverLodge, "The Historyof the CohererPrinciple," Electrician40 (1897): 87-91. Referalso to E. Branly'spapers under the title"Variations of Conductivityunder ElectricalInfluence," Electrician 27 (1891): 221-22, 448-49. Also useful is VivianJ. Phillips,Early Radio WaveDetectors (London, 1980), pp. 18-37. 'Rowlands (n. 5 above), pp. 116-17. 21Thelecture, "The Work of Hertz," was published in Nature,the Electrician(with illustrations),and later in the Proceedingsof the Royal Institution. The referencehere is to Oliver Lodge, "The Work of Hertz," Nature50 (1894): 133-39. The lecture, with appendixes,was published in 1894 as a book, The Workof Hertz and Someof His Successors (London, 1894). From the thirdedition (1900), its titlewas changed to Signallingthrough Spacewithout Wires.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 727

paraffin prism O Q a 6" spherical

galvanometer metallicradiator boxin a

polarization grating

a Bfanly tube in a copper hat

FIG. 2.--Oliver Lodge's quasi-optical experimentwith Hertzian waves at the Royal Institutionin June 1894. A sphericalradiator is in a metallicbox, and a Branlytube is in a copper hat. Notice the mirrorgalvanometer. (Electrician 33 [1894]: 205.) and a Branlytube in a copper hat as a detector,and a mirrorgalvanometer as a signal indicator(fig. 2). In the second part,he explained the functionof human eyesby means of the analogyof the coherer."When lightfalls upon the retina,"Lodge said, "thesegaps become more or less conducting,and thenerves are stimulated."''Lodge also triedan outdoor experiment,in whichthe receiverwas in the theaterand the transmitter was in the libraryof the Royal Institution,separated across 40 yardsby three rooms and stairs.I shall returnto thisoutdoor experimentafter examiningLodge's otherdemonstrations. The coherer,in particularthe Branlytube, had a characterthat was absent in the spark-gap resonator.After detecting electromagnetic waves,the coherer needed to be mechanicallyvibrated or "tapped" to make it ready for the next wave trains.This featureraised a question withrelation to physiologicalconcerns. To what,in human eyes,did this tappingcorrespond? Lodge assumed that,in the eye, "the tappingback is done automaticallyby the tissues,so thatit is alwaysready for a new impression."How to demonstratethis automatic tapping in human eyes?Lodge prepared an electricbell, whichwas mounted on the same board as the filingtube. By constantlyvibrating itself, and thus by constantlyshaking the table and the cohereron it,the bell alwaysmade the coherer readyto detect new waves.23 Was Lodge's lecturesuccessful? It is true thatthe published abstracts in Natureand the Electricianwere read worldwide.Nevertheless, the

nLodge, "Workof Hertz," p. 137. "Ibid. It is noteworthythat the bell was neitherconnected to the coherer circuitnor tapped the coherer directly.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 728 SungookHong demonstrationswere rather unsuccessful. The Electriciannoted that"the experimentswere performedunder very unfavourableconditions.""24 Moreover,the "lack of enthusiasm"in Lodge's lecture was contrasted withthe successof Nikola Tesla's lecturea yearearlier, where "the weird waving of glowing tubes in the suitablydarkened room" impressed everyone.What Lodge lackedwas a "theatricaleffect" or "scenicsetting." Neither the sound of the spark nor the "moderate galvanometer" connectedto the cohererwas theatrical.In particular,the galvanometer wasvery tricky. It provedto be a "verylively kind of galvanometer" for the coherer circuit.The swingof the needle was not stable,not even when therewere no waves.For success subsequently,the Electriciansuggested usinga more effectivegalvanometer such as a deadbeat galvanometer. No detaileddescriptions about the galvanometerused byLodge survive. From the abstractand the figure,we see thatLodge used a mirrorgalvanometer.25From the commentin the Electrician,we understandthat it was notof a deadbeattype. From other pieces of evidence, we knowthat Lodge had notpaid muchattention to the galvanometer, in contrastto the situation severalyears earlier. Before the coherer,for example, FitzGerald, Lodge's closestfriend and professorof naturaland experimentalphilosophy at TrinityCollege, Dublin, had constructedan extremelysensitive galvanom- eterto showto an audience the detectionof waves. This instrumentwould have needed to detectthe disturbanceof electricequilibrium caused bya tinyspark.26 The coherer,an on-offswitch, made such a sensitivegalvanom- eterunnecessary, because the galvanometerhad to detectonly a relatively largecurrent from a battery,triggered by the action of the coherer. As Lodge noted,"a roughgalvanometer" was thereforesufficient27 But whywas the galvanometertroublesome at the crucial moment? Lodge suspectedthat the source of the troublewas the electricbell used forthe automatictapping. There is a "jerk current"in the electricbell, whichwould certainlyinfluence the adjacent coherer electrically.The jerky current "produces one effect,and a mechanical vibration... produces an opposite effect;hence the spot of light can hardlykeep still."He knewthe wayto eliminatethis: a "clockwork"that did not use an electriccurrent "might do better."2As we shall see, Lodge actually employed the clockwork in his Oxford lecture two months later.

24"HertzianWaves at the RoyalInstitution" (lead article),Electrician 33 (1894): 156-57. "Lodge mentioned"the spot of light"in a mirrorgalvanometer. See Lodge, "Workof Hertz,"p. 137. "George F. FitzGerald,"Electro-Magnetic Radiation" (FridayEvening Lecture at the Royal Institutionon March 21, 1890), Nature42 (1890): 172-75. "See also Lodge's exhibitionof the portable detectorof his assistant'sdesign at the RoyalSociety soiree a fewdays after his FridayLecture, in "The RoyalSociety Conversazi- one," Nature50 (1894): 182-83. "Lodge,"Workof Hertz" (n. 21 above), p. 137.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 729 Now recall Lodge's statementconcerning the "Muirheadconnection" in his letterto Fleming. Here Lodge emphasized that in his Oxford demonstrationhe used a deadbeat Thomson (Kelvin) marinegalvanometer he had borrowedfrom Muirhead's firm. In 1900 Lodge stated that "Dr. Alexander Muirhead foresawthe telegraphicimportance of this method of signallingimmediately after hearing the author'slecture on June 1st, 1894, and arrangeda siphon recorderfor the purpose."29In Lodge's much-quotedletter to one of his friendsin 1914, he emphasized that"it was at the firstof theselectures [Royal Institution Friday Lecture] that my friendAlexander Muirhead conceived the telegraphicapplica- tions which ultimatelyled to the foundationof the Lodge-Muirhead Syndicate."'I Elsewhere,Lodge recalled thatthe galvanometerat Oxford "responded to signalssharply, in a dead-beatmanner, without confusing oscillations.""3This Muirheadconnection makes Lodge's telegraphictrial at Oxford,performed only two months after his obviouslynontelegraphic experimentat the RoyalInstitution, both feasibleand timely. Some parts of this Muirhead connection are undoubtedly true. Muirhead constructeda delicate siphon recorderfor a wirelessdetector duringthe late 1890s and early1900s; Lodge and Muirhead,who began to file for patents on wirelesstelegraphy in 1897, formedthe Lodge- Muirhead Syndicate in 1901. Nevertheless,the central point in the Muirhead connection (thatMuirhead lent Lodge a deadbeat Thomson marine galvanometerafter/because he was inspired by Lodge's June lecture) is verydoubtful. According to the recollectionof Muirhead's wife,it was Lodge's Oxford lecture,not the Royal Institutionlecture, that inspired Muirhead to think about wirelesstelegraphy." One of Lodge's biographersdoubts that he actuallyused a Thomson marine deadbeat galvanometerborrowed from Muirhead at Oxford." But that

"Lodge, Signallingthrough Space without Wires (n. 21 above), p. 45. Referalso to Oliver Lodge, "Alexander Muirhead" (obituary), Proceedingsof the Royal Society100, pt. A (1921-22): viii-ix. SOliverLodge toJ.Arthur Hill, December 11, 1914, inJ.Arthur Hill, ed., Lettersfrom Sir OliverLodge (London, 1932), p. 47. 3"OliverLodge, "Reminiscences of the Last BritishAssociation Meeting in Oxford, 1894," Discovery7 (August 1926): 263-66 (quoted on 265-66). See also the same recollectionin OliverLodge, AdvancingScience (London, 1931), p. 164, and Past Years:An Autobiography(New York, 1932), p. 231. 3"Muirheadwas excited afterLodge's Oxford lecture, and "the next day he went to Lodge with the suggestionthat messages could be sent by use of these waves to feed cables." See M. E. Muirhead,Alexander Muirhead (Oxford, 1926), p. 39, quoted in Pocock (n. 5 above), p. 83. "Rowlands (n. 5 above), p. 148, n. 30. Thomson's marine galvanometerwas a very sensitivecurrent-measuring device speciallydesigned so thatthe swingof a ship could not change the readings.In principle,it utilizedrotation of a smallmagnet fixed in the middle of the coils bysilk fiber. When magneticfields were created around the coils bythe action

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 730 SungookHong is, I think,highly plausible, not because Muirheadhad been inspiredby Lodge's June lecture,but because Lodge had borrowedfrom Muirhead such a device at various timessince the late 1880s.' In addition,as we have seen, Lodge had an urgentreason to use a deadbeat galvanometer. He had experiencedserious trouble in his "livelykind" galvanometer in theJune lecture,and the Electricianhad recommendedthe employment of a deadbeat galvanometerfor future success. These factorsmight be the real motivationsfor Lodge's use of a Thomson marinegalvanometer at Oxford,if it was actuallyused there. Let us returnto Lodge's outdoor experimentat the RoyalInstitution. Why did Lodge performthis experiment?Evidently, it was not to determinethe maximumtransmitting distance, nor to show the wave's penetrabilityof walls. Its real purpose layin physiologicalconcerns. With a metricalBranly tube and an electricbell, Lodge wanted to show that the coherer could discernHertzian waves of variousintensities Oust as the human eye could). The easiestway to varythe intensityof waveswas to adjust the distancebetween transmitter and receiver.Lodge placed a 6-inchsphere radiatorin the libraryof the RoyalInstitution, which had two advantages. First,owing to the theoryof Horace Lamb and J.J. Thomson, it was easyto estimatethe wavelength:with a 6-inchradiator, the wavelengthwas about 8 or 9 inches. Second, owing to FitzGerald's theory,it was known that the energyof radiationat a distance, other thingsbeing equal, is inverselyproportional to the fourthpower of the wavelength.That is, the shorterthe wavelength,the more the energyof radiation,and thus the higher the possibilityof being detected at a distance. The belief thatshort waves had more power to travelfarther than long waveswas stronglyinscribed in Lodge's mind.35But even with the shortwave, Lodge estimatedthat "something more like halfa mile of current,the small magnet was forced to rotate,and this effectwas magnifiedby the reflectionof a ray of lightfrom a small mirrorfastened to the magnet. For a detailed descriptionof the device, see George B. Prescott,Electricity and theElectric Telegraph (New York,1888), pp. 154-57. "See, e.g. Lodge, ModernViews of Electricity (n. 16 above), p. 300, whereLodge used the Thomson marine galvanometer lent by Muirhead for his experiments on electric momentum.Notice also thattheir business relation started around the same timewith the constructionof Lodge's lightningguard by the Muirhead Company.For this,see Oliver Lodge, LightningConductors and LightningGuards (London, 1892), pp. 419-26. I thankIdo Yavetzfor thisreference. 35HoraceLamb, "On ElectricalMotion in a Spherical Conductor,"Philosophical Trans- actionsof the Royal Society174, pt. 2 (1883): 519-49; J.J. Thomson, "On Electrical Oscillationsand the EffectsProduced by the Motion of an ElectrifiedSphere," Proceedings of theLondon Mathematical Society 15 (1883/84): 197-219. For FitzGerald'stheory, see George F. FitzGerald,"On the Quantityof EnergyTransferred to the Etherby a Variable Current,"Transactions of the Royal Dublin Society (1883), in TheScientific Writings of the Late

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 731 was nearer the limitof sensitiveness,"even though he appended that "this is a rash statementnot at presentverified."" What was the resultof thisfirst outdoor experiment?Was it success- ful?Lodge and hisfriends later repeated thatthe experimentwas a great success.The answerwas, however, both yes and no: no, because he failed to detect the wave witha metricalfiling tube; yes,because he detected it with his sensitivecoherer. Lodge's manuscriptconfirms this: "The sphericalradiator ... thoughit could excitethe filingstube ... when 60 yards away in the open air, yet could not excite it perceptiblywhen screened offby so manywalls and metal surfacesas exist between the Libraryand Theatre of the Royal Institution.It could, however,still easilyexcite the coherer,which is immenselymore sensitive,and also more troublesomeand occasionallycapricious than is a tube of iron filings.""7With this experiment he was thusunable to show the metrical response of the Branlyfiling tube to radiationsof various intensities. Two months later, on August 14, 1894, at the joint session of the Physicsand Physiologysections of the BritishAssociation, Lodge delivered twolectures and demonstrationson Hertzianwaves at the theater in the museum of Oxford University.The firstlecture was on "Experi- mentsIllustrating Clerk Maxwell'sTheory of Light"; the second was on "An ElectricalTheory ofVision." In a sense, he splithis previousFriday Lecture into two. In the firstlecture, Lodge used a spherical radiator and a copper hat to concentratethe radiation.As before,Branly's tube and Lodge's coherer were used as detectors,with most experiments done withBranly's device. Refractionsand reflectionsof Hertzianwaves were demonstratedwith lenses, gold papers, the human body,paraffin prisms,and a slab of wood. Polarizationwas shownwith a copper wire polarizer;splitting of the polarized rayinto the twoelliptically polarized rayswas also demonstrated.These experimentswere "verybeautifully, verycarefully and veryconvincingly demonstrated," and "the audience ... repeatedlyshowed its warm appreciation." Lodge's employmentof a deadbeat galvanometermight have been a reason forthe success.'

GeogeFrancis FitzGerald, ed. Joseph Lamor (London, 1902), pp. 122-26. For Lodge, see Lodge, AdvancingScience (n. 31 above), p. 165. 6Lodge,"Work of Hertz" (n. 21 above), pp. 135-37. 37OliverLodge, "Notes on the Historyof the Coherer Method of DetectingHertzian Wavesand otherSimilar Matters" (n.d.), Lodge Collection,UCL. In the publishedarticle, a similarparagraph read, "Almostany filingtube could detect signalsfrom a distance of 60 yards,with a mere six-inchsphere as emitterand withoutthe slightesttrouble, but the single-pointcoherer was usuallymuch more sensitivethan any filingtube." See Lodge, "Historyof the Coherer Principle" (n. 19 above), p. 90. 'Since the lectureswere not published,I relyon the briefreports of the meetingsof the BritishAssociation published in Nature,Electrician, Engineering, and London Times,all of

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 732 SungookHong These were really "the prologue" of Lodge's second lecture and demonstrations.In thislecture, Lodge proposed his hypothesisconcern- ing the theoryof vision that the coherer circuit"may be taken as an analogous, and may,ex hypothesi,be an enlarged model of the mecha- nism of vision." According to this hypothesis,"the retinal elements constitutean imperfectconductor, and ... the lightwaves would cause a sudden diminutionin the resistanceof the elements."Yet,once struck by the wave, the coherer "has a tendencyto persistin its lessened resistance"and thereforerequires tapping "to jerk the coherercontact back to itsnormal state of badness." For thistapping, he used "a sortof clockworkapparatus which automatically produces the tap everytenth of a second." With this device, Lodge showed that "for a continuous radiationthe coherer showed continuousindications, which died away when the radiationceased."" Where was the transmitterin the physiologicalexperiments? The issue has never been examined critically.Four years later, in 1898, Thompson-Lodge's close friendand professorof applied physicsand electricalengineering of the FinsburyTechnical College-commented thatthe radiatorwas in the Clarendon Laboratory,the buildingadjacent to the museum,at a distance of 200 yards."I believe thatThompson's statementmay be erroneous,because Lodge, in his various recollections, never mentioned the Clarendon Laboratory.He would only remark that "in both cases, signallingwas easily carried on from a distance throughwalls and other obstacles, an emitterbeing outside and a galvanometerdetector inside the room," or that "this [sending] apparatuswas in anotherroom."41 Contrary to Lodge's and Thompson's remarks,the four sources on which I have relied for my account say nothingabout the outdoor trialat all. Consideringthis evidence, as well as Lodge's previoustrouble with the outdoor experimentat the Royal Institution,it maybe said thatthe distance traversedby Hertzianwaves in the Oxfordlecture was fairlymodest. The lecturewas followedby heated discussionsby such physicistsas Lord Rayleigh,Henry E. Armstrong,and FitzGerald,and the physiolo- whichsent their reporters to the BritishAssociation. See "Physicsat the BritishAssociation," Nature50 (1894): 408; "The BritishAssociation at Oxford:Tuesday, August 14th," Electrician 33 (1894): 458-59; "The BritishAssociation, Section A: Electric Theory of Vision," Engineering58 (1894): 382-83; "The BritishAssociation," London Times, August 15, 1894. 3g"The BritishAssociation at Oxford: Tuesday, August 14," p. 458; "The British Association,Section A: ElectricTheory of Vision"; LondonTimes, August 15, 1894. "Silvanus P. Thompson, "TelegraphyAcross Space" (lecture givenat the Royal Society of Artson March 30, 1898), Journalof the Society of Arts 40 (1897/98): 453-60, esp. 458. 4Lodge, "Historyof the Coherer Principle" (n. 19 above), p. 90, and Past Years(n. 31 above), p. 231.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 733 gistsBurdon Sandersonand EdwardA. Sharpey-Schifer,marking a great success. But thatwas all. There is not the slightesthint of telegraphic signals, nor "dots and dashes." With his improvedautomatic tapper, Lodge showed the persistencyof vision and mere sensation of light, which corresponded to the continuous and short indication of the galvanometer.But that was far from dots and dashes for alphabetic signals.From beginningto end, the lecturewas entirely"Lodgian." His purpose was to investigatethe relationbetween optics and electromag- netism,between light and electromagneticwaves, and between optical receptors and electromagneticones. Afterthe lecture, despite Muir- head's and Rayleigh'ssuggestions, Lodge did not pursue this subject further.He soon busied himselfwith ether experiments,X-rays, and psychicresearches. The preceding examinationhas shown that Lodge's firstargument, namely,that he actually transmitteddots and dashes for alphabetic signalsin the Oxfordmeeting, is doubtful.Let us nextexamine Lodge's second argumentabout "a Morse instrument,"mentioned in his letter to Fleming.Fleming thought that this instrument must be a Morse inker. But it was not. Ironically,Aitken's first source revealsits nature. It listed fiveinstruments used in Lodge's Oxford demonstrations,and one of themis "Morseinstrument toshake the filings"42 (emphasis added). Lodge's Morse instrumentwas nothingbut a clockworkor an automatictapper thathe used fortapping the coherer.To be sure,the Morse instrument thatLodge used forthe clockworkwas a telegraphicdevice, but he used this telegraphicdevice for nontelegraphicpurposes, as confirmedby himself in his description of the automatic tapper in the Oxford meeting: "The tapping back was at firstperformed by hand ... but automatictappers were verysoon arranged; ... an electricbell was not found verysatisfactory, however, because of the disturbancescaused by the little at its contactbreaker ... so a clockwork spark tapper,consisting of a rotatingspoke wheel driven by the clockwork ofa Morseinstrument, and giving to the filingstube or to a coherer a series ofjerks occurringat regular intervals... was also employed."" The "Morse instrument"was neither a Morse inkernor a substitutefor a galvanometer.To understandhow a clockworkwas transformedinto a Morse detector,we now examine the impact of Marconi's wirelesstelegraphy on the BritishMaxwellians.

Marconi,Preece, the Maxwellians, and "Practiceversus Theory" Since 1886, Lodge and his Maxwellian friends,Oliver Heaviside (1850-1925) in particular,had been involvedin a bittercontroversy-

4"Dr. Oliver Lodge's Apparatusfor Wireless Telegraphy" (n. 9 above), p. 686. eLodge, "Historyof the Coherer Principle,"p. 90 (emphasis added).

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 734 SungookHong the so-called Practicevs. Theory controversy--withWilliam H. Preece (1834-1913), an eminentpractical telegrapher of the Post Office.The issues of the controversyinvolved the role of the self-inductionof lines and its implicationin long-distancetelephony and lightningconductors. Heaviside's counterintuitive,theoretical claim for the beneficial effectof self-inductionfor long-distance telephony was severelyrebuked by Preece, who based his argumenton his practiceand experience in the field.The newsthat Hertz had discoveredMaxwell's electromagnetic waves was known to them in 1888, when Lodge was attemptingto generate and detect electromagneticwaves on wires with Leyden-jar discharge.Even thoughHertz deprived Lodge of creditfor the discovery of electromagneticwaves, and even though the electromagneticwave was not directlyrelated to the controversies,Hertz's discoverycertainly had a favorableimpact for the Maxwellians,allowing them to defeat Preece. The mostimportant part of Maxwell'stheory was proved,and it was followedby Sir WilliamThomson's warmrecognition of Heaviside's mathematicalwork in 1889, markingthe victoryof the theoreticalmen over practicians.Hertz's discoveryof Maxwell's electromagneticwaves was timelyand was good for theoreticians." In 1896, Marconi came to England withhis "secretbox" (see fig.3). In July1896, Preece, then chief engineer of the Post Office,became Marconi'sfirst, and mostpotent, patron. As Preece had had interestsin inductiontelegraphy for severalyears, he mighthave realized a possibilityof commercialwireless telegraphy in Marconi's demonstration. But in Marconi's apparatusPreece saw more than commercialpossibil- ity;it was a good means of revengeagainst the theoreticalcamp of the Maxwellians.Like Preece himself,Marconi was "whatMr. Oliver Heavi- side calls a 'practician,'" who knewnothing about Maxwell'smathemati- cal theoryand perhaps littleabout Hertz's physicalexperiments. But Marconi had developed the Hertzianwave telegraphy,which Lodge had failed to do. To Preece, Marconi's success was a marvelousexample of the superiorityof practice over theory.The Hertzian wave that had defeated Preece in 1888 now became his weapon.45

"For thiscontroversy, see BruceJ. Hunt, " 'Practicevs. Theory': The BritishElectrical Debate, 1888-1891," Isis 74 (1983): 341-55; D.W. Jordan, "The Adoption of Self- Induction by Telephony,1886-1889," Annals of Science39 (1982): 433-61; Ido Yavetz, "Oliver Heaviside and the Significanceof the BritishElectrical Debate," Annalsof Science 50 (1993): 135-73. 'For the descriptionof Marconi as "practician,"see "Notes,"Electrician 39 (1897): 207. Differentopinions have existedabout the relationbetween Preece and Marconi.Aitken, in Syntonyand Spark(n. 4 above), pp. 210-16, suggeststhat Preece's interestcame fromthe "bureaucraticresponsibility" of Preece and the Post Officeto overseethe developmentof all formsof electric communicationin Britain. Based on the manuscriptsof the Post Office,Pocock showsthat Preece was rathercool towardthe Marconisystem's commercial

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 735

-L

FIG. 3.-Marconi in 1896 with his "secret box" closed. (Courtesyof the Marconi CompanyArchives, Chelmsford.)

At the meetingof the BritishAssociation in Liverpool in September 1896, Preece prepared twocounterattacks. First, based on the observa- tionsof various submarine cables, he attackedHeaviside's mathematical theoryof distortionlesscables and advocated his own empiricallaw." Then, in his discussionof J. Chunder Bose's paper, Preece stated that "an Italian had come with a box givinga quite new systemof space telegraphy,"advertising Marconi's splendid success in transmitting across 1? mileson SalisburyPlain.47 Preece's announcementastonished most Maxwellians,as shown in the followingquote froma letterfrom possibilityand thenargues that Preece in factfollowed the policyof the Post Officeto new inventions--"neitherto accept the invention,nor to investsubstantial sums" without entirelyignoring it altogether.See Pocock (n. 5 above), pp. 114-17. But Pocock seems to feel the difficultyin explainingwhy Preece ardentlyadvertised Marconi in the British Associationand in his public lectures.The difficultydisappears if the personalfactors are counted in. Amongsecondary materials, Paul Nahin, OliverHeaviside: Sage in Solitude(New York, 1988), p. 281, mentionsthis possibility. "WilliamH. Preece, "On Disturbancein SubmarineCables," AnnualReport of the British Associationfor theAdvancement of Science(1896): 732 (title only), and "ElectricalDistur- bances in SubmarineCables," Electrician37 (1896): 689-91. 47Lodge, AdvancingScience (n. 31 above), p. 168. See also "Physicsat the British Association,"Nature 54 (1896): 567; "The BritishAssociation," London Times, September 23, 1896; "Notes," Electrician37 (1896): 685. Preece also mentionedMarconi's parabolic antenna in the transmitterand a relayand a Morse inkerin the receiver.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 736 SungookHong FitzGeraldto Heaviside: "On the lastday but one Preece surprisedus all by sayingthat he had taken up an Italian adventurerwho had done no more thanLodge & othershad done in observingHertzian radiations at a distance.Many of us werevery indignant at thisover-looking of British workfor an Italian manufacturer.Science 'made in Germany'we are accustomed to but 'made in an unknownfirm was too Italy'by bad."'4 According to the later recollection,Lodge did not get up to refute Preece,who was "farmore ignorantthan he oughtto have been of what had been alreadydone," but "retiredto [his] laboratoryand riggedup an arrangementwhich I showed to Lord Kelvinand a fewothers, saying 'This is what Preece was talkingabout.' "49 In December 1896, Preece again publicized Marconi's feat in his public lectureat ToynbeeHall and promisedthere to spare no expense for Marconi's research.This promise especiallyupset the Maxwellians, because theywere then engaged in difficultnegotiations with the British governmentto secure financialsupport (?35,000) forthe establishment of the National PhysicalLaboratory (NPL). Lodge had initiatedthe movementin 1891 at the BritishAssociation's annual meeting.When it was revived in 1895 by Douglas Galton, Lodge was appointed as secretaryof the BritishAssociation Committee on the Establishmentof an NPL. FitzGeraldhad also emphasizedthe role of science in industrial development."The Maxwellianswere at firstnervous about Preece,who continuouslypublicized Marconi as the "inventorof wirelesstelegraphy,"and who, as an influentialperson at the Post Office,ignored the role of scientificresearch. Yet, theirattitude to Marconi was not very hostile initially.In March 1897, in a letter to Thompson, Lodge expressed his hope that "M[arconi] is improvingthings all around & going to bring it in commercially."It was certainlybecause Lodge thoughtthat "there will be manyimprovements in detailswanted before that can be done.""' But thingswere moving rapidly.Somebody had coined and publicized the term"Marconi waves";Marconi approvedof it. In an interviewwith McClureS Magazine, Marconi remarkedthat his wave from the vertical antenna was not same as Hertz's wave. He

"George F FitzGeraldto Oliver Heaviside, September28, 1896, Heaviside Collection, Institutionof ElectricalEngineers, London. "4Lodgeto Fleming,August 26, 1937 (n. 14 above). Lodge's remarkon Preece is in Lodge to Hill, December 11, 1914 (n. 30 above). Association 5OliverLodge, "PresidentialAddress in SectionA," AnnualReport of the British forthe Advancement ofScience (1891): 550-51; George F. FitzGerald,"Science and Industry" (lectureto the IrishIndustrial League on May 7, 1896), in ScientificWritings (n. 35 above), p. 383. For the NPL, see also E. Pyatt,The National Physical Laboratory: A History(Bristol, 1983), pp. 12-35. 51OliverLodge to SilvanusP. Thompson, March 16, 1897, Lodge Collection,UCL.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 737 emphasized that his wave could penetrate almost everything.5'This strangecomment was accompanied by his splendid practicalsuccesses. In March 1897 Marconi succeeded in transmittingover 4 miles; he conquered 8 miles of the Bristol Channel in May. Popular reports poured forth,and public interestin wirelesstelegraphy ran high. Withhis secretbox and verticalantenna, Marconi pulled the Hertzian waves out of the scientificlaboratories. At first,as is often the case, scientistswere not very effectiveoutside their laboratories.Nobody could exactly guess what constitutedMarconi's secret box. Nobody could explain whythe Marconi wave could communicateacross build- ings and even high hills. Most important,it was not certain whyonly Marconi could send the messagesover several miles when all othershad failed.53The viewsof scientificauthorities on Hertzianwaves no longer held. Instead,a practicalsuccess, along withpublic recognition,became the new authority.As the editorialof theElectrician remarked, "Professor Marconi," along withTesla and Edison, had become an authorityon electricalscience to the Britishpublic, instead of Lord Kelvin,George G. Stokes,and H. von Helmholtz.' An invisiblebattle between theory and practicewas under way. On June 4, 1897, Preece had planned a lecture on the "Signalling through Space withoutWires." This was the firstFriday Lecture on wirelesstelegraphy. Having heard thisnews, Lodge sentPreece a copyof his FridayLecture in 1894, "to remindhim" of whatLodge had already done.55In the lecture, Preece compared Marconi to Columbus and applauded Marconi'sfeat as "a new systemof telegraphy."From Lodge's 1894 lecture, Preece quoted Lodge's comment that "half a mile was nearer the limitof sensibility"and then proudlydeclared that "half a

5"Whenasked about the difference,Marconi answered, "I don't know. I am not a scientist,but I doubt ifany scientistcan tell you." See H. J.W. Dam, "Telegraphywithout Wires:A Possibilityof ElectricalScience. II. The New Telegraphy-Interviewwith Signor Marconi," McClure'Magazine 8 (March 1897): 389-92. On an episode of how much the "Marconi wave" upset Silvanus Thompson, see Jane Smeal Thompson and Helen G. Thompson, SilvanusPhillips Thompson: His Lifeand Letters(New York, 1920), p. 81. 53ConcerningMarconi's secret box, therewas an interestingstory. When FrederickT. Trouton, an assistantof FitzGerald,found an ordinaryglass-tube coherer in Marconi's secretbox, Marconi slammed it down again, saying,"you would steal myinvention." On this,see Jolly,Sir OliverLodge (n. 5 above), p. 148. FitzGeraldseems to have firstsolved the puzzle of the Marconisystem. He analyzedthat "what Marconi is doing withhis kites,poles &c &c, is to manufacturean enormousradiator and it is not the shortwaves of his double ball arrangementthat he is emittingand receivingbut the verymuch longerwaves of his whole system.By connecting to earth he uses the earth as the second plate of his transmitter....Anyway a bigopen systemis the thing."See George E FitzGeraldto Oliver Lodge, October 30, 1897, Lodge Collection,UCL. ""The Man in the Streetof Science" (lead article),Electrician 39 (1897): 546-47. 55OliverLodge to SilvanusP. Thompson,June 1, 1897, Lodge Collection,UCL.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 738 SungookHong mile was the wildestdream." By doing so, Preece successfullyderided a theoretician'srash predictionand "scored an effectivehit.",% The lecture was a blow not only to Lodge but also to most British Maxwellianswho had engaged in controversywith Preece severalyears before. "Preece is," FitzGeraldwrote to Lodge indignantly,"distinctly and intentionallyscoffing at scientificmen and deserves severe re- buke."57Lodge was concerned about his creditsas a mediatorbetween pure scientificresearch and commercial wireless telegraphy.In his immediateresponse as a letterto the London Times,Lodge explained thatthe predictionof a halfmile was "a scientificone, concerningthe smalland earlyapparatus." He emphasizedthat he himselfshowed "the same plan of signallingin 1894." Lodge also emphasizedthat Marconi's coherer had been used by Rayleighand Lodge himself.8 Lodge here triedtwo different, but related,strategies. The firstwas to stress the essential similarityof his 1894 experimentsto Marconi's telegraphy.As Lodge reminded Thompson, "we had the automatic tappingback in '94 at Oxford;... we have reallyhad the tapperworked as a relaytoo & collectorsto the coherer;in fact,the whole thingexcept the best conductingvacuum coherer."59Lodge's second strategywas to findthe connectionbetween the effortsof the Britishscientific men like Lodge and Marconi's wireless telegraphy.But neither of these two strategieswas easy. Lodge's 1894 lectures were not of a telegraphic nature at all, and the connectionsof the Britishscientists with Marconi were too indirect.Frederick T. Trouton,an assistantof FitzGerald,had advised Marconi in 1893 or 1894 via one of Marconi's friends.But Trouton's advice proved neitherscientific nor of the technical kind." Such efforts,however, became meaninglessafter Marconi's patentwas accepted. The impact of Marconi's patent was much more profound than his practicalsuccesses.

Marconi Patent"for Everything" On June 16, 1897, about two weeks afterPreece's Royal Institution lecture,and twoweeks before the finalacceptance of Marconi'spatent, an interestingdemonstration was held at the RoyalSociety soiree. In the entrancehall, Preece and Marconi demonstratedwireless telegraphy in theirreceptive method of "Signallingthrough Space withoutWires"; on

."Notes" (n. 45 above). For Preece's FridayLecture, see WilliamH. Preece, "Signalling through Space withoutWires," Electrician39 (1897): 216-18. The lecture was later published in the Proceedingsof the Royal Institution 15 (1896/97): 467-76. 57GeorgeF. FitzGeraldto Oliver Lodge, June 21, 1897, Lodge Collection,UCL. 'Oliver Lodge, "Telegraphywithout Wires," London Times, June 22, 1897. "Lodge to Thompson,June 1, 1897 (n. 55 above). "FitzGeraldto Lodge, June 21, 1897 (n. 57 above).

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 739 the second floor,Muirhead demonstratedthe same "as practisedby Dr. OliverLodge in 1894."Here, Muirhead used a Branlytube and a Morse inker,and Preece and Marconi used a Morse sounder. The distances betweenthe transmittersand the receiverswere about 100 feet.Accord- ing to the Electrician'sjudgment, "Lodge's systemworked satisfactorily," and "the markingof the signalson the ribbonwere undoubtedly distinct and readable.""61 Fromthis brief description, we can noticethat Alexander Muirhead had collaboratedwith Lodge, ignitingthe competitionbetween Marconi's and Lodge's method.About a monthearlier, Lodge had fileda patenton the "Improvementsin SyntonizedTelegraphy without Line Wires."As the tide indicates,the principleof syntonyor tuningby varying the inductanceof the transmitterand the receiverwas its centralpart. The patentis now famous as the firstpatent on syntony.But its provisionalspecification claimedmore than that. Another important claim was on Lodge's improvementof Branly'stube filingsand itsuse as a detector.Lodge also made a claim on his tappingdevice such as an electricbell and a clockwork.In short,the patentwas on the Lodgian systemof wirelesstelegraphy.62 Marconi had filedhis provisionalspecification on June 2, 1896, about a yearbefore Lodge's patent.There was no doubt thatMarconi's patent was the firstpatent on Hertzianwave telegraphy,but thereexisted much doubt about its power.For Marconi's success to be continued commercially,the patenthad to be strongenough to overcomethe subsequent litigation.But its provisionalspecification shows the immatureMar- conism clearly.For instance, as Aitken points out, it contains such passages as " whentransmitting through the earth or water I connectone end of the tube or contact to earthand the otherto conductors"(emphasis added). This illustratesMarconi's early convictionthat waves from a verticalantenna were differentfrom Hertzian waves." In addition,an automatic tapper of Marconi's own design, operated by the relay current,was described side by side with an independent tremblerof Lodge's clockworktype. If he had committedthe same mistakein the complete specification,he would have invalidatedhis patent.

61"Notes"(n. 45 above), p. 237. See also "The Royal SocietyConversazione," Nature 56 (1897): 185. 6Oliver Lodge, "Improvementsin SyntonizedTelegraphy without Line Wires," no. 11,575, Provisional Specification (date of application, May 10, 1897), and Complete Specification (February 5, 1898; date of acceptance, August 10, 1898). For Lodge's syntony,see Aitken,Syntony and Spark(n. 4 above), pp. 130-42. OAitken,Syntony and Spark,pp. 285-86, n. 12. See also GuglielmoMarconi, "Improve- mentsin TransmittingElectrical Impulses and Signals,and in ApparatusTherefor, " no. 12,039,Provisional Specification (date of application,June 2, 1896). The contentof the patent,of course, had been kept secret until its complete specificationwas accepted on July2, 1897.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 740 SungookHong Without doubt, Marconi could safelypatent two things:a tapper activatedby the relaycurrent," and an antenna, thatis, the aerial and the earth connectionfor the transmitterand the coherer." Except for thesetwo, the matterwas extremelyuncertain. His transmitterwas of the Righitype, his detectorwas an improvedBranly filing-tube coherer, and his relayand inkerwere ordinarytelegraphic devices. The cohererwas most problematic.Even though the Britishpatent on inventionwas given to the one who had firstapplied forit ratherthan to the person who had firstinvented the deviceor publishedit, it was generallybelieved thatMarconi's claim on the coherermust be a modestone, restrictinghis claim to the improvementof its sensitivity.Even expert opinion was vacillating,as is shownby the followingremark of FitzGerald:

Troutonwas sufficientlyimpressed with its [Marconi'ssecret box's] value to venturesome moneyin the concern.Since findingout how the thingis reallyworked he has become much more doubtfulas to the validityof the patentsand has refusedto put any more money into it. It is all a question of patentrights and maydepend on such a questionas thatmercury [in the coherer] is importantin orderto make the thingwork with certainty and thata hammerworked by the relayitself is importantand so forth.If these thingsare of value and patentable, the patents may be of considerable importance. Branly'stube, Righi's emitter &c are all certainlyimpatentable, but so manythings go to make up a workableinvention that Marconi's patentsmay be valuable."

However,FitzGerald's conclusion was optimistic:"As faras I can judge fromwhat I am told it is onlydetails thatare patentableand theirvalue is not proved."The editorialopinion of the Electricianwas similar.This predictedthat Marconi's patent would not be a masterpatent, because the general principlesunderlying the apparatus,as well as the appara- tuses themselves,were not new.67And therewas anotherfactor contrib- utingto such optimism.Since Marconiwas not a man of science,he had probablycommitted an error in describingthe principle of wireless telegraphy(as he did in his provisionalspecification). If such were the

"Even Lodge admittedMarconi's noveltyin the tapping system.See Oliver Lodge, "Reportto the ChiefEngineer of the GovernmentTelegraphs" (June 1900), in ADM. 116. 570, Public Record Office,London, p. 5. 'For a contemporarywitness on Marconi'santenna, see A. Slaby,"The New Telegraphy: Recent Experimentsin Telegraphywith Sparks," CenturyMagazine 55 (April 1898): 867-74, esp. 870-71. Even Lodge admittedthat this was Marconi'shighly original novelty. See Lodge, Signallingthrough Space without Wires (n. 21 above), p. 47. "FitzGeraldto Lodge, June 21, 1897 (n. 57 above). 67""Notes"(n. 45 above), p. 431.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 741 case, the patentwould be invalidated.At the veryleast, this might leave room for another patent. The completespecification for Marconi's patent was filedon March 2, 1897. But, as Aitkencomments, it was a "differentkind of document entirely.""Between the provisionaland completespecification, Marconi had secured the crucial assistanceofJ. FletcherMoulton, certainlythe most famous patent expert in Britain." Moulton's assistancesurprised the Maxwellians. Thompson wrote to Lodge on June 30, 1897, "I happen to know that Moulton was called in to advise Marconi on the claim of his finalspecification of patent,.... and he advisedhim to claim everything.I understand that as the claim was drawn, they claim, for telegraphy,not only coherers,oscillators, & such like details, but even Hertz waves! ... there is nothing new except the Hertz wave, the oscillator& the coherer,and these are not patented nor patentable."70 Marconi's patent was accepted on July2, 1897. Meanwhile,Marconi, who had been under the patronage of Preece and the Post Office, formeda privatecompany to exploit his patent.71 As the contentsof Marconi's patent were publicized, his secretbox was finallyopened (see fig. 4). Marconi detailed his inventionsand attached nineteen claims. To everyone'ssurprise, most of these claims were related to coherersand the variousmethods of connectingthem, such as the ground connection. The claims were not limited to his improvement,but to the coherer itself.There were claims on the ball transmittersof Righitype, relay and hammertapper, even his improved inductioncoils and the antenna (elevated condenser plate, not vertical wire)." In addition,an awkwardexpression like "transmittingthrough earth and water" was replaced by a more refined expression like

6?Aitken,Syntony and Spark(n. 4 above), p. 204. 'John Fletcher Moulton (1844-1921) was the firstSmith's Prizeman and Senior Wranglerof the MathematicalTripos in Cambridge,in 1868. He soon became Fellow of the RoyalSociety as a resultof his electricalresearch and thenengaged in legal works.See Hugh FletcherMoulton, TheLife of Lord Moulton (London, 1922); Dictionaryof National Biography(1912-21), s.v. "JohnFletcher Moulton," pp. 392-94. "Silvanus P. Thompson to Oliver Lodge, June 30, 1897, Lodge Collection,UCL. 71Itwas on July20, 1897, and the company was the WirelessTelegraph and Signal Company. In February 1900 the name was changed to Marconi's Wireless Telegraph Company.For the earlyhistory of the company,see W.J. Baker,A Historyof the Marconi Company(London, 1970), pp. 35 ff. "Under the Britishpatent systemat that time,in whichthe comptrollerof the Patent Officehad no power over the contentsof the patent,an inventorcould claim as many inventionsas he wanted in a single specificationat his own risk.In cases of some new inventions,an inventorcould deliberatelyforge the claimswith the effectof monopolizing the "principle"of thatinvention, rather than merelya specificartifact. Marconi's patent was close to such cases. James Watt'spowerful patent on his new steam engine with a separate condenseris anotherexample. Referto EncyclopaediaBritannica (Chicago, 1961),

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 742 SungookHong

M I '

-,t *,

FIG. 4.-Marconi around 1900 withhis "secret box" open. (Courtesyof the Marconi CompanyArchives, Chelmsford.) transmitting"where obstacles, such as manyhouses or a hill or moun- tains,intervene between the transmitterand the receiver.""In termsof scientificprinciples, there was no mistake.FitzGerald noted that"Moul- ton has drawnhis patentstoo cutelyto commithim to any particular theoryof what he is doing." Even the criticalElectrician appraised the specificationas "a model of perspicuity."" How did Marconi,who was thoughtof as a modestand open youth, dare to claim everythingin the Hertzianwaves? How did he claim an originalityover the Branlytube thathad been used and improvedby Lodge, and over the ball transmitterof Righi type?75Once Marconi's widelyranging patent was accepted, Lodge had to withdrawhis claims on the coherer and tapping device in filinghis complete specification the followingyear. Only the principle of syntonywas leftin Lodge's patent.With thisdefeat, Lodge musthave feltan immensefrustration and a feelingof betrayal. s.v."Patent," 17:372. See also E. Robinson,"James Watt and the Law of Patents,"Technology and Culture13 (1972): 115-39. nGuglielmoMarconi, "Improvementsin TransmittingElectrical Impulses and Signals, and in ApparatusTherefor," no. 12,039, Complete Specification(March 2, 1897). The patent is also printedin J.J.Fahie, A Historyof Wireless Telegraphy, 1838-1899 (New York, 1899), pp. 296-320. 74"Notes"(n. 45 above), p. 665. On FitzGerald'scomment, see FitzGeraldto Lodge, October 30, 1897 (n. 53 above). 'Justafter Marconi's patent was published,Electrician published a seriesof articles on the coherer,including Lodge's "Historyof the Coherer Principle" (n. 19 above).

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 743 An elementof nationalismdeepened the frustration.Marconi was an Italian. The "ether" had been discoveredby greatBritish scientists like Faraday,Kelvin, and Maxwell.The Maxwellianswere theirheirs, but they had lost the priorityof the discoveryof electromagneticwaves to a German, Heinrich Hertz. Maxwell's electromagneticwave was then named the Hertzian wave. Lodge tried to change its name to the "Maxwellian wave" at Oxford,but he failed as a result of the strong objection of another German scientist,Ludwig Boltzmann.76The possibilityof a commercialuse of the ether was then opened by Marconi. This rendered Lodge twice narrowlyanticipated by foreignersin important discoveries. Marconi's comprehensive patent worsened things.The immenseuse of wirelesstelegraphy during wartime and for naval ships seemed obvious. If Marconi's patentwent unchallenged,it would monopolize not only Hertzian waves but also importantBritish national interests.It was thus no accident that,after Marconi's patent, many Britishscientists and engineers such as J.J. Thomson, Minchin, Rollo Appleyard,and Campbell Swintonjoined withLodge in deprecat- ing Marconi's originality.77 As Thompson reported in 1899, "They were evidentlypurposely draftedas widelyas possible to cover all possible extensionsto telegraphy,explosion of mines,and the like,which, indeed, were talked about publiclyin connection with Marconi from the first.... theyare not patentsfor telegraphy, but forthe transmissionby Hertz waves of signals or impulsesof anykind. .... In thissense beyondall questionLodge was using Hertz waves for a wireless'telegraph' in 1894."78For Lodge and Thompson, it was Marconi,with his marvelouslybroad claims,who first violated "the rules of the game." Thus, therewas no need for them to followthe rules.

ConstructingLodge's Priority Now let us examine Aitken'sfirst source, an articlein the Electrician entitled "Dr. Oliver Lodge's Apparatus for WirelessTelegraphy." The articlewas intentionallypublished side by side withMarconi's patentas the "best antidoteof Marconism.""79Aitken apparently thought that the articlecould supportthe claim of Lodge's telegraphyin 1894. But there was in factno mentionof Lodge's telegraphictrial. What the articlesaid

76Forthis episode, see "The BritishAssociation," London Times, August 15, 1894; Lodge, AdvancingScience (n. 31 above), pp. 162-63. Even afterthis, Lodge oftenused the term "Maxwellianwave"; see, e.g., his "Historyof the Coherer Principle,"p. 89. "Pocock (n. 5 above), pp. 103-5. 7SilvanusP. Thompson, "Report of WirelessTelegraph Patents" (1900), in ADM. 116. 570, Public Record Office,p. 38. ""Notes" (n. 45 above), p. 665.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 744 SungookHong was that "Lodge described and exhibited publicly in operation a combinationof sendingand receivingapparatus constituting a systemof telegraphysubstantially the same as that now claimed in" Marconi's patent,and that"Dr. Lodge publishedenough threeyears ago to enable the mostsimple-minded 'practician' to compound a systemof practical telegraphy.""These two strategiesare exactlythe same as Lodge's two strategies,namely, identifying the principlesof his experimentsin 1894 with those in Marconi's wirelesstelegraphy and stressingthe possible influenceof Lodge on Marconi. After1898, the "Maxwell-Hertz-Marconi"genealogy in wirelessteleg- raphywas firmlyestablished. More so, Lodge and Thompson tried all possibleways of refutingMarconi. In order to weaken Marconi'spatent, theyadvertised that, due to the wires("base lines,"as Thompson called them), "there is no such thingas wirelesstelegraphy." They publicized other scientists'success, particularly Adolf Slaby's success in Germany.8" But, most importantfor our discussions,Lodge's 1894 experiments began to be interpretedas telegraphicin nature.Thompson forthe first timeforged the claim that"on severaloccasions, and notablyat Oxford in 1894, he showed how such coherers could be used in transmitting telegraphicsignals to a distance.He showed thatthey would workthrough solid walls.Lodge's greatdistance at thattime had not exceeded some 100 or 150 yards.Communication was thusmade betweenthe University Museum and the adjacent building of the Clarendon Laboratory""82 (emphasisadded). It markedthe beginningof the long storyof Lodge's telegraphyin 1894. Thompson's "telegraphic interpretation"of Lodge's 1894 experi- mentsdid not appear in Lodge's own writings.In 1900, Lodge admitted that "the writer[Lodge] himselfdid not pursue the matterinto telegraphic application,because he was unaware that therewould be any demand for this kind of telegraph.""3In the third edition of his Signallingthmugh Space withoutWires (1900), which Fleming even criti- cized as "a perversionof fact,"84Lodge's recollectionwas essentiallythe same, sayingthat "so faras the presentauthor was concernedhe did not realise that there would be any particular advantage in thus with difficultytelegraphing across space.... In this non-perceptionof the practicaluses of wirelesstelegraphy he undoubtedlyerred."85

80"Dr.Oliver Lodge's Apparatus for Wireless Telegraphy" (n. 9 above). 8"Thompson,"Telegraphy Across Space" (n. 40 above); "Dr. Lodge on Wireless Telegraphy,"Electrical Review 42 (1898): 103-4. 8"Thompson,"Telegraphy Across Space," p. 458. "Lodge,"Report to theChief Engineer of the Government Telegraphs" (n. 64 above). 'MJohnAmbrose Fleming to GuglielmoMarconi, January 12, 1900,Marconi Company Archives,Chelmsford. 'Lodge, Signallingthmugh Space without Wires (n. 21 above),p. 45.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 745 Lodge was at thattime developing an alternativesystem of inductionor magnetictelegraphy to competewith Marconi's. For this,Lodge allied in 1898 withPreece, who feltbetrayed after Marconi had formeda private company."In 1901,Lodge, thenprincipal of Birmingham University, also launched the Lodge-MuirheadSyndicate. At that time, the Marconi Company tried to contractwith the Royal Navy,Lloyd's, and the Post Officebut encountered a seriesof obstacles.The PostOffice had planned a patentlitigation with Marconi's company in 1899 and asked Lodge and Thompson their expert opinion on Marconi's 1896 patent.87Because Preece was skepticalabout the litigation,it was finallygiven up. The followingyear, the Royal Navy,which had been suspiciousof Marconi's connectionwith the Italian Navy,prepared another litigationand was given the reportof Lodge and Thompson fromthe Post Office.This litigationwas finallyabandoned, afterCaptain HenryJackson,an expert in wirelesstelegraphy in the RoyalNavy and Marconi'sfriend, advised the Admiraltynot to pursue it.8 Instead of contractingwith the Marconi Company,Lloyd's had triedto develop itsown system. The year 1901 was a verylucky one for Marconi. Lodge eventually abandoned induction telegraphyafter his failurein the summer.The Royal Navy and Lloyd's contractedwith the Marconi Companyfor the use of Marconi's system.In December 1901, Marconi succeeded in the transatlantictransmission of the signal "SSS" over 1,800 miles.After this event,Marconi's success was too obvious to be challenged. Lodge and Thompson lost theirchance. Moreover,as Fleming,Lodge's friendand professor of electrical engineering at UniversityCollege, London, became scientificadvisor to the Marconi Companyin 1899 and FitzGer- ald died in 1901, the Maxwelliancamp was breakingup. The situation became more and more unbearable to Lodge and Thompson, who thoughtthat Marconi's successes were based on hisviolation of the rules of the game in 1896-97. In April 1902, even before the heat of Marconi's firsttransatlantic success had dissipated,Thompson revivedthe issue of the inventionof

'Oliver Lodge to WilliamH. Preece, March 4, 1898, in E. C. Baker, Sir WilliamPreece, E R S.: VictorianEngineer Extraordinary (London, 1976), pp. 299-300. For Lodge's and Preece's magneticinduction telegraphy,see Oliver Lodge, "Improvementsin Magnetic Space Telegraphy,"Journal of theInstitution of ElectricalEngineers 27 (1898): 798-849; WilliamH. Preece, "AethericTelegraphy," Journal of the Institution of Electrical Engineers 27 (1898): 869-86. 87Lodge,"Report to the ChiefEngineer of the GovernmentTelegraphs" (n. 64 above); Thompson, "Reportof Wireless Telegraph Patents" (n. 78 above). Thompson summarized (on p. 36) thatMarconi's claimswere eitherinvalid or unessential. 'For an account of these corporatepolitics around 1900, see Jolly,Marconi (n. 5 above), pp. 68 ff.Jackson's report on Lodge's and Thompson's reportsis in the formof a letter, Capt.Jackson to J.A. Fisher,in ADM. 116. 570, Public Record Office.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 746 SungookHong wireless telegraphyby attackingMarconi in the SaturdayReview. He wrotethat "Signor Marconi is not the inventor,but the skilledexploiter, of telegraphywithout wires" and that "the original inventorof the wirelesstelegraphy" is "ProfessorOliver Lodge." Lodge was transformed here from a pioneer to the original inventorof wirelesstelegraphy. Thompson'sevidence was nothingnew: Lodge had used a ball oscillator, coherer,relay, and automatictapper in 1894 and delivered"a signal in the telegraphicinstrument.""89 Thompson's motivationwas also unorigi- nal: Marconi, consciouslyor unconsciously,devalued scientists'prior creditsby violatingthe rules of the game. Lodge wroteto Thompson in appreciationof "the wayin whichyou referto myclaims or rightsin the matter."He added that"the opinion of one who is alwaysso well informedon historicalsubjects ought to carry considerableweight."" Marconi, however, was embarrassedby Thomp- son's attack.In his reply,Marconi stressed his priorityin patentsand the noveltyin his antenna and tapper design.9'Thompson rejoined again, criticizingMarconi: "Now the matterdoes not reston any assertionof mine (forthere are scoresof personsliving who witnessedit) thatin 1894 Principal Lodge did publicly transmitsignals from one building to another,through several stone walls, without connecting wires, by means ofHertzian waves which were received perfectly clearly upon a telegraphic instrumentto whichthese waves were relayedby means of an automati- callytapped 'coherer.'If thatis notwireless telegraphy, then the termhas no meaning.""9On thismatter, Marconi, who was not good at controversy and had not been in Britainin 1894, could make no furtherreply. The issue was raised again in 1906, when the second International Congresson WirelessTelegraphy took place. Thompson reiteratedhis claim of Lodge's priority,but this time Fleming refutedThompson's claim: "When it is assertedthat Lodge sent 'signals'by electricwaves in 1894, whatit meant is thathe caused an oscillatoryelectric spark made in one room or buildingto affecta cohererand so move the needle of a galvanometerin an adjacentroom, and showedthese experiments both at the RoyalInstitution inJune, 1894, and at the BritishAssociation, Oxford, in the same year.But therewas not a single trace of any suggestionof applicationto telegraphyin his lectureand in the reprintof it.""9

"Silvanus P. Thompson, "The Inventorof WirelessTelegraphy," Saturday Review 93 (1902): 424-25 (quoted on p. 424). 9Oliver Lodge to SilvanusP. Thompson,April 11, 1902, Lodge Collection,UCL. "9GuglielmoMarconi, "The Inventorof WirelessTelegraphy: A Reply,"Saturday Review 93 (1902): 556-57. 92SilvanusP. Thompson, "WirelessTelegraphy: A Rejoinder,"Saturday Review 93 (1902): 598-99. 9JohnAmbrose Fleming, "WirelessTelegraphy: To the Editor of The Times,"London Times,October 29, 1906. Referalso to the articlesof SilvanusP. Thompson (October 12,

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 747 In a widelyread articleon "WirelessTelegraphy" in the Encyclopaedia Britannica(11th ed., 1911), Flemingagain refutedLodge's telegraphyin 1894." The same year,however, Thompson's and Lodge's effortswere most fruitful;Lodge's syntonicpatent in 1897 was extended for seven more years (despite a petitionby the Marconi Company). This exten- sion was partlydue to Thompson's detailed historicalanalysis of Lodge's wirelesstelegraphy in 1894 and of Lodge's firstsyntonic patent in 1897.'9 Encouraged by thisvictory, Lodge prepared a patentlitigation against the Marconi Company.Preece helped settlethe case on conditionthat the Marconi Company would purchase Lodge's syntonypatent and appoint Lodge as its scientificadvisor. Lodge became the company's advisor,whose advice was never asked. Instead, the Lodge-Muirhead Syndicatewas simplydissolved." After that time, the name of Lodge was no longer importantin the world of commercialtelegraphy. AfterThompson's death in 1916, Lodge took Thompson's role, emphasizinga telegraphicnature of his 1894 lectures.In 1925, Lodge recalled, "It was now [in 1894] possible to use them [experimental successes] for roughlyand imperfectlytransmitting signals in theMorse code,either by the directuse of a Thomson marinespeaking galvanometer or througha relay by operating an ordinaryMorse tape instrument..... In August 1894, I exhibited thismethod of signallingat the BritishAssociation in Oxford"" (emphasisadded). This was modest,but the followingyear, Lodge gave a detailed and much more audacious descriptionof his Oxfordlecture, which became the standardone until his death and has persistedeven to the present:

The possibilityof actual signallingby this method was insistedon at Oxford.... The sending instrumentwas a Hertz vibratoractuated by an ordinaryinduction coil set in action by a Morse key.The apparatuswas in anotherroom, and wasworked by an assistant.The receivingapparatus was a filingtube in a copper hat,in circuitwith a battery,actuating either a Morse recorderon a tape, or,for better demonstration,a Kelvin marine galvanometer,as firstused for Atlantictelegraphy before the siphon recorder replaced it. The instrumentwas lent me by Dr. Alexander Muirhead, whose firm

1906), Kelvin (October 16, 1906), and C. C. A. Swinton (October 29, 1906) under the heading of "WirelessTelegraphy" in the LondonTimes. 'John AmbroseFleming, s.v. "Wireless Telegraphy," in EncyclopaediaBritannica, 11th ed. (Cambridge,1911), 26:529-42. I thankone of the Technologyand Culturereferees for this reference. 95SilvanusP. Thompson, Noteson Sir OliverLodges Patent for Wireless Telegraphy (London, 1911). 96Aitken,Syntony and Spark(n. 4 above), pp. 163-68. 9"OliverLodge, Talksabout Wireless (London, 1925), pp. 37-38.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions 748 SungookHong habituallyconstructed a number of cable instruments.... When the Morse keyat the sendingend was held down,the rapid trembler of the coil maintainedthe waveproduction, and the deflectedspot of lightat the receivingend remained in its deflectedpositions as long as the keywas down; but when the keywas onlymomentarily depressed,a shortseries of waveswas emitted,and the spot of light thensuffered a momentarydeflection. These long and shortsignals obviouslycorresponded to the dashes and dots of the Morse code; and thusit was easyto demonstratethe signallingof some lettersof the alphabet,so thatthey could be read by any telegraphistin the audience-some of whom mayremember that they did so. Trulyit was a very infantilekind of radio-telegraphy,but we found that distance was comparativelyimmaterial; and at Liverpool,where I was then working,the dots and dashes were received with ease across the quadrangle,or fromany reasonable distance."

The same passage was reproduced in Lodge's monograph on the historyof the BritishAssociation, Advancing Science (1931), and in his autobiography,Past Years(1932)." By then,Lodge was over eighty.The experimentshad been performednearly forty years earlier; there were fewliving scientists who could clearlyremember what Lodge had done in 1894. Lodge's argumentwas picked up byW. H. Eccles, a Britishradio engineer,and the incorrectdiagram of Lodge's detectorin 1894, which consistedof a coherer,a telegraphicrelay, a hammer-typetapper, and a Morse inker-all of which, except for the coherer, Lodge had not used-appeared in his popular book on radio.'" Eccles's diagram was noticed by Fleming,who was preparinga lectureon Marconi's lifeand works.Fleming wrote to Lodge to confirmthe diagram,and thatwas the beginningof our story. Conclusion Several elements are intermingledin this long historyof Lodge's "wireless telegraphyin 1894." A typical battle over the priorityof inventionis there, but that does not explain everything.Regarding priority,Lodge's concern had rather been his scientificcredit as a mediator between Hertz and commercial telegraphy.But Marconi audaciouslydenied this.Marconi had neveradmitted Lodge's, nor any

"Lodge, "Reminiscencesof the Last BritishAssociation Meeting" (n. 31 above), pp. 265-66. Past Years 31 9Lodge, AdvancingScience (n. 31 above), pp. 163-64; (n. above), pp. 231-32. to this had 'W. H. Eccles, Wireless(n. 13 above), p. 54. See also fig.1. Prior book, Eccles stated thatLodge transmittedMorse code signalsin Oxfordin 1894. See Eccles, "Physics in Relation to Wireless,"Nature 125 (1930): 894-97, esp. 894.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions The Originsof Wireless Telegraphy Revisited 749 other physicist's,influence on his invention.In addition, there was Preece's counterattack.Preece publiclyclaimed that Marconi's invention was an outcome of the effortsof a "practician"like himself.This broughtall of them (except forMarconi, who did not knowits meaning) to the "Practiceversus Theory" controversyseveral years before. That upset Lodge and the other Maxwellians,particularly since theywere thentrying to showthe significanceof pure researchfor technology and industryin seekinggovernmental support for the establishmentof the National PhysicalLaboratory. Marconi's "secret box" threatenedthe Maxwellians'established he- gemonyin electricaltheory and practice.The genealogyof "Maxwell- Hertz-Marconi"was unbearable. A furtheraggravation was Marconi's patenton "everything,"which virtually monopolized the ether.Marco- ni's sweepingclaims for the cohererespecially infuriated Lodge. Nation- alismand patriotism,which had existedin a rathersubtle form from the beginning,emerged to the surfaceat thisstage. The BritishMaxwellians thoughtthat theyshould not hand over the rightsto monopolize the ether to Marconi, who apparentlyviolated the rules of the game. Marconi was not an inventor,but only an "exploiter" of wireless telegraphy.In thesecircumstances, Lodge's 1894 Oxforddemonstration was recharacterizedas the firstwireless telegraphy. By this maneuver, Lodge became itsinventor, and the BritishMaxwellians were somewhat consoled for theirlosses.

This content downloaded from 209.129.30.134 on Fri, 20 Feb 2015 20:40:51 PM All use subject to JSTOR Terms and Conditions