Guest blog by Andrew O. Krastins
“Now he is dead. And the most bitter regret that, of so much artistic value, there remains only a memory, such being, unfortunately, the fate of the great performers: to survive only by the virtue of tradition, also fallacious and dying.” – Supplemento al Caffaro di Genova, February 19, 1894, announcing the death of Camillo Sivori earlier that morning.
“This newest invention of Mr. Edison is indeed astonishing. The phonograph makes it possible for a man who has already long rested in the grave once again to raise his voice and greet the present.” – Baron Helmuth von Moltke (1800-1891), spoken into an Edison phonograph in 1889.
***
Photo of Sivori, inscribed to Selene Hofer
Photo of August Wilhelmj
“Sivori is dead. He dies the last of his generation and school” the Violin Times announced on March 5, 1894, “and his death severs the last link that connected the present day with what we may almost call the era of romance in the history of violin playing. . . It is therefore an event of no slight importance to musical history, this extinction of what might be called the Paganini School.” Sixteen mysterious brown wax cylinders which the British Library holds and has provisionally attributed to the great German violinist August Wilhelmj (1845-1908), suggest that this bold but melancholy statement is not entirely true, and that Camillo Sivori (1815-1894) – the only pupil of Niccolo Paganini – long rested in the grave, is ready, in the words of Baron von Moltke, “to again raise his voice and greet the present.”
Before the phonograph, musical performance was ephemeral. If one wanted music, one had either to make it oneself or be in the presence of other humans making it. Once a performance ended, it was lost to posterity, and remained with the auditor only as a fading memory, like the voices of the dead. Hence the bitter melancholy of the Genoese obituary above.
Music is now omnipresent, to be purchased, packaged and consumed at the listener’s whim. Concert audiences, “smart phones” in hand, record rather than listen; musical amateurs easily and routinely document and post their efforts on social media. Students learning a new piece turn first to YouTube to compare a dozen different performances before venturing their own.
It is impossible to “unremember” or “unexperience” technologies to which we have grown accustomed. But to grasp the significance of the British Library cylinders, it is essential to attempt, through imagination, to place oneself back into the musical world of the late 1880s and early 1890s, when performing musicians first experienced the preservation and reproduction of human sounds, when for the first time in human history, a performer’s art could be immortalized as was the art of painters, sculptors and poets; it is essential to contemplate the awe and dread the new technology inspired in musicians suddenly faced with the terrifying prospect of chiseling their own artistic epitaphs, indelible and permanent.
In 1823 and 1824, when Sivori was studying with Paganini, Beethoven completed his Ninth Symphony and Missa Solemnis, and was planning his last string quartets. In 1827, the year of Beethoven’s death and a year before Schubert’s, Sivori was already an international concert artist at age 12, acclaimed in London.1 The earliest-born classical violinists to record commercially were Joseph Joachim (1831-1907) and Pablo de Sarasate (1844-1908), whose recordings date from 1903 and 1904 respectively. If the British Library cylinders are by Sivori rather than by August Wilhelmj, to whom they are presently attributed, then they necessarily date from before February 19, 1894, the date of Sivori’s death.
Should the weight of the evidence point to Sivori rather than Wilhelmj, then the British Library cylinders memorialize performances by a Romantic virtuoso nearly a full generation older than any other classical performer known to have recorded, not from the world of Debussy or even Brahms, but of Paganini, Spohr, Bellini, Berlioz, Chopin, Schumann and Mendelssohn; they are a portal into a lost musical world, a world as strange to us as ours would be to the mysterious performers whose artistry the cylinders preserve.
A personal note and roadmap for the reader.
One winter afternoon in 2008, while my employer was otherwise distracted, I chanced upon the “Wilhelmj” recording of Paganini’s Witches’ Dance on the British Library website, and, with office door closed, listened intently. As a lifelong student of lore pertaining to the great 19th century violin virtuosi, I was excited to learn that the great August Wilhelmj (1845-1908) had made recordings. At the time, the only sample of these “Wilhelmj” recordings available online was the “Witches’ Dance.” To hear the remaining sets required a trip to the British Library’s reading room and listening stations.
In London the following year, I for the first time listened to the whole collection – five classical violin compositions recorded in their entirety: the first movement of Paganini’s First Violin Concerto, the first movement of Camillo Sivori’s unpublished and lost Second Violin Concerto, Paganini’s Witches’ Dance, a major unaccompanied work entitled “The Gypsies” and a “Minuetto Pizzicato.” Because of their early vintage, I knew that these multi-cylinder recordings of lengthy concert pieces were unique among known existing cylinder recordings.2
The Sivori Second Violin Concerto, The Gypsies and the Witches’ Dance created in my mind such a storm of excitement, perplexity and astonishment that I paced several times around the reading room, and listened again and again, and paced again and again, arousing curious glances. I knew that there was far more to these intriguing artifacts than the perfunctory attribution to Wilhelmj suggested. The Gypsies, for solo unaccompanied violin, in particular, stunned me with the unearthly evocation of a Romani chorus in sustained and unbroken three-part chordal passages and hair-raising octave passages at lightning speed – passages that professional violinists who have heard them cannot explain or duplicate. Neither the composer nor the performer is identified anywhere on the cylinders.
Gypsies ex 1
Gypsies ex 2
In 2018, Jonathan Summers, the British Library’s curator of classical music, generously allowed me to examine the cylinders themselves and the boxes they arrived in, and I began solving the riddle of the Mystery Cylinders in earnest. My first stop was the Wilhelmj Archive in the German village of Usingen, where I combed through a lifetime’s worth of Wilhelmj’s collected papers and discovered no evidence that Wilhelmj ever performed a work by Sivori. My further adventures took me to the winding streets of Genoa, to Paris where I stayed in the modest rooming house where Sivori lived for three decades, and to various used book dens, flea markets and auction houses. And, of course, to the British Library, the Bibliotheque nationale de France, and, of crucial importance, Genoa’s Biblioteca Civica Berio.
My subject necessarily deals with three distinct and arcane specialities: (1) the history of early sound recording and related technical minutiae, (2) the interrelationships between Thomas Edison’s various corporations and business entities, and (3) the history of 19th century violin performance practices. A specialist in any one of these fields, alas, might well be bored to tears by the two thirds of my essay pertaining to the speciality of another. But not so the general reader, to whom my essays are respectfully dedicated.
The story we are about to begin is set largely between 1889 and 1894, in London, Paris and Genoa. It is rife with intrigues, implausible coincidences, hair’s breadth rescues, and treasures snatched from oblivion in the nick of time. And it is suffused by the then novel and frightening idea that the human voice could survive death. If you detect the fragrance of Victorian detective and ghost stories found in the pages of old pulp magazines now long-decayed, it is simply the nature of the raw material itself, with the caveat that our story happens also to be true – verifiably so.
Among the characters, in addition to Maestro Sivori and August Wilhelmj, is Thomas Edison, whose broad experimental curiosity and creative impulses were wedded to the purely practical, stingy and bitterly litigious temperament of a business magnate. There is the eccentric Colonel George Gouraud, who controlled Edison’s European Phonograph operations headquartered in London – a blustering, adulation-seeking and hucksterish American entrepreneur Mark Twain might have created. There are important cameo appearances by the eccentric Hungarian violinist and political revolutionary Edouard Remenyi and Mrs. Remenyi, and by the enigmatic Hungarian-Jewish-German-French diplomat, impresario, journalist, novelist, librettist, translator and polyglot Emile Durer, who recorded Italian musical celebrities at the very dawn of sound recording, was Edison’s first French biographer, and had enduring relationships with both Edison and Gouraud.
But most important and least known is Enrico Copello of Genoa, who at 15 fought alongside Garibaldi in the wars for Italy’s unification and independence, emigrated to America to seek his fortune, and in the very earliest years of recorded sound, traveled throughout Italy as Edison’s representative, demonstrating the Phonograph and recording Italian musical celebrities, only to be mangled in the machinery of Gilded Age American “Business.” Misled by Edison and thwarted by Gouraud, Copello lost what money he had invested and, beggared by his former colleagues, returned with his family to America to live in obscurity in a New York boarding house. He is an unsung hero, his legacy never before excavated, his achievements buried and his memory sullied by the enduring falsehoods of a single English journalist. We will visit his grave presently.
Here is the roadmap. Wilhelmj is extremely unlikely to have recorded the cylinders because by the earliest time the cylinders could have been made, he had already given up public performance and none of the pieces recorded but one was ever in his known concert repertoire. The only person presently known to have performed and have had access to the Sivori Second Violin Concerto was Sivori himself. The inscription on the Witches’ Dance cylinder box states: “as played by Paganini.” As Paganini’s only pupil, Sivori knew how Paganini played; Wilhelmj could not because Paganini died five years before Wilhelmj was born. However, at the time of Sivori’s death, the Phonograph was extremely rare in Europe and was not available to the general public. At the earliest date that Sivori could possibly have recorded, he was already in his late seventies. In the last months of his life, his health fluctuated.
Sivori remained an active soloist in public concerts at least until late 1892, and as a musician in private soirees at least up to the spring of 1893. His last private performance likely took place in Genoa, in the weeks before his death on February 19, 1894. Enrico Copello is the only person presently known of to have conducted Phonograph demonstrations in Italy and recorded Italian classical musicians sufficiently early to have recorded Sivori. The Phonograph was exhibited in Genoa in 1892 during the time when Sivori was giving a series of valedictory concerts and private recitals. Sivori returned from Paris to Genoa in October 1893 and remained there until his death on February 19, 1894. The Phonograph reappeared in Genoa in January 1894 where it was demonstrated at the Sala Sivori concert hall and later at the music store of Sivori’s friend Giuseppe Bossola, while Sivori was in Genoa.
Sivori therefore had the capacity to make the recordings, and access to the Phonograph, in 1892, and again in early 1894 when his health by then was precarious. Because Sivori was the only known violinist with access to the unpublished Sivori Second Violin Concerto, he is almost certainly the performer on the cylinders. Documentary and circumstantial evidence also point to Sivori as the performer on the Paganini cylinders as well. Copello was working on behalf of the branch Edison Phonograph enterprise headquartered in London and made numerous trips there. This explains how and why the British Library’s Mystery Cylinders wound up in London.
Evidence is one thing, inferences drawn from it quite another. Claims about events long past and people long dead are necessarily provisional, especially where, as here, little or no prior excavation has been done. This Adventure of the Mystery Cylinders has led to the discovery of hidden treasure chambers long closed off, namely, the previously unknown and unsuspected legacy of Copello’s pioneering recordings of great classical musicians in Italy beginning in 1889. I hope that the curiosity of the general reader, and of specialists and hobbyists in these very interesting overlapping areas, will be sufficiently sparked to unabashedly point out any errors in the facts and any flimsiness in reasoning. The best insurance against persistent historical error is its early detection.
I hope that readers will join in this adventure by adding to our knowledge with their own discoveries. For example, some of Sivori’s closest acquaintances in Genoa and Paris described him as a life-long confirmed bachelor. This is consistent with what is known of his life in in both cities. Some scholars claim otherwise.3 At one point he was rumored to be engaged to marry the French actress Hortense Damain, but news accounts state that the plans were cancelled and there is no evidence that any wedding took place.4 According to Damain’s death certificate, Damain died a “celibataire,” that is unmarried. I have found no reference to Sivori’s children in the Genoa newspapers covering his funeral, nor in the published collections of his letters I have examined. Any documentary evidence that Sivori had children – even just their names – would be immensely useful and interesting, if indeed there ever were any children. Perhaps I have aroused the curiosity of readers with a penchant for genealogy.
Anyone who can identify the performer, composer or even just the tunes in “The Gypsies” cylinders will have solved puzzles that still perplex me. They arrived in the same container and from the same source as the other Mystery Cylinders, but the handwriting on the individual boxes does not match that on the Sivori or Witches’ Dance boxes. Nor is there any evidence I have found of any such composition in lists of Sivori’s lists known works, or the known works of Wilhelmj and Paganini himself. My own wish-driven impulse is to attribute them to Sivori, but without more, this cannot be done. And I hope that additional fragments of Sivori’s Second Violin Concerto will surface. From my own experience, I know that such discoveries are to be expected provided the mind is prepared.
And I hope that the Mystery Cylinders are not only read about but listened to – listened to with at least some of the patience, effort and sacrifice that went into creating them. The uninitiated listener’s first reaction might well be annoyance at the huge amount of noise and difficulty in making out the performance. This is where patience and a willingness to gamble away some of one’s own time are essential. I hope that readers who have never experienced recordings from the early 1890s will indulge me and listen repeatedly, without judgement or expectation, to one of the performances, simply to let the mind and ears adjust. Let the sounds be what they are, like listening to ambient sounds in a forest, by a waterfall or at the seaside. The experience is rather like seeing something one knows to be exquisite and irreplaceable, but seeing it through very dirty and sooty old window of an long-abandoned house. The longer one looks, the more one sees; the longer one listens, the more one hears.
Sivori very possibly made his recordings only weeks and maybe even days before his death, while wracked by recurring illness. If that is the case, the Mystery Cylinders are a true final artistic testament made under circumstances as heroic as they are heartbreaking. I hope that listeners will make a fraction of the effort to reach into the past that Sivori made to reach into the future and out to posterity.
In matters such as this, I follow three maxims: (1) if an artifact is known to have been created, I presume that it still exists absent evidence to the contrary; (2) the truth is in the detritus; and (3) nothing is too preposterous to be true nor too plausible to be false. And with those thoughts, I invite the British Library’s readers to join in the hunt.
The British Library’s Mystery Cylinders
In 2005, an elderly former employee of the London branch of the Schott music publishing house donated a box of 16 brown wax cylinders to the British Library. The donor had retrieved the cylinders from a rubbish bin where they had been discarded in the course of clearing out the Schott firm’s old facilities and moving to a new building in the 1960s. According to the donor, the cylinders were found in the desk of Charles Volkert, the German-born head of Schott’s London branch. There they apparently had remained through Volkert’s death in 1929 and up to the time the old Schott facility was cleared. Had the donor been home sick that day, or on the telephone, or otherwise distracted, and had he not, at the last moment, rescued the Mystery Cylinders from the garbage bin, they would have been forever lost and even their creation would never be known. These are the first of our treasures snatched from oblivion in the nick of time.
The donor informed the British Library that the cylinders were thought to have been recorded by the great German violinist August Wilhelmj, who resided in London from the end of 1894 until his death in 1908. Schott published Wilhelmj’s compositions; during his London years, Wilhelmj edited and transcribed dozens upon dozens of major and minor compositions for Schott. These included Paganini’s Witches’ Dance and Paganini’s first and second violin concerti. Based on this information, the British Library identified the cylinders as “believed to be by August Wilhelmj,” presumed they were made in London after Wilhelmj’s arrival, and estimated the cylinders to date from 1895 to approximately 1900.
The British Library’s attribution to Wilhelmj was eminently plausible in light of the information then available. Wilhelmj lived in London from until his death in 1908. In the first decades of sound recording, London was the hub of Edison’s phonograph enterprise for all of Europe.5 Wilhelmj easily could have made phonograph recordings if he wished, because by the time of Wilhelmj’s death in 1908, sound recording in the United Kingdom was a fully developed and highly competitive commercial enterprise. And if the recordings were made on the Continent by someone else, what were they doing in Volkert’s desk for three quarters of a century?
To the untrained eye, nothing about the British Library’s 16 Mystery Cylinders or any other brown wax cylinders – considered simply as objects – inspires awe; they look like obsolete clutter.
Photos of Cylinders (C1210) at the British Library
But to students of the earliest recorded music, and of the history of violin performance, they present a tangle of mysteries and contradictions which, if one did not know better, seem deliberately crafted to vex posterity. Unlike most recordings of the era, they do not announce the performers. Nor are the performers identified anywhere on the cylinder boxes. Unlike any other early cylinder recordings known to exist, they memorialize entire classical concerto movements and lengthy instrumental compositions recorded over four and even five cylinders, rather than short compositions which fit neatly on a single cylinder. To add to the mystery, one set memorializes a performance of the first movement of Camillo Sivori’s Second Violin Concerto – a composition which was never published and the manuscript score and solo part of which are lost.
Sivori Concerto box
The label of the four-cylinder recording of Paganini’s “Le Streghe,” or “Witches’ Dance” is as puzzling as it is tantalizing: “The Witch’s Dance, - a Song of the Old Woman under the Walnut Tree, as played by Paganini. During the dark ages, the Walnut Tree was believed to be the trysting place of witches. Hence the Old Woman’s Song.”
Witches' Dance box with inscription
Is the phrase “as played by Paganini” mere “AS SEEN ON TV!” hyperbole? Or did the writer instead mean that the performance reflected the actual style of Paganini himself? And what is the significance of the curious language about walnut trees and an Old Woman’s Song? Like the other sets, no performer is identified, either on the boxes or the cylinders themselves. But the version of “Witches’ Dance” on the Cylinders is substantially different from the 1851 first edition of the piece, from Wilhelmj’s edition for Schott published around 1905, and from Paganini’s own manuscript. A puzzle indeed.
Finally, two of the most intriguing performances are also the most mysterious. A four-cylinder set memorializes a major multi-movement piece for unaccompanied violin entitled “The Gypsies.” This set of cylinders contains some of the most astonishing violin playing on record, including sustained three-string legato passages, as well as complicated trills and rapid octave passages that would do credit to any modern virtuoso. The other single-cylinder composition is entitled “Minuetto Pizzicato,” and contains similar sustained three-string passages. Neither set identifies even a composer, let alone the performer. No such compositions are among the known works of Paganini or Sivori or Wilhelmj. While these cylinders are, as objects, substantially identical to the others and arrived at the British Library in the same box and from the same source, there presently is no additional documentary evidence to suggest that they were recorded by Wilhelmj, Sivori or any other particular violinist.
The Mystery Cylinders, by their nature, inspire wish-driven reasoning. The idea that Paganini’s only pupil left a major recorded legacy is inherently appealing; it is all too tempting to bury the absence of actual evidence in “surely would have” and in surmises which are as enticingly plausible as they are unsupported by anything beyond the author’s hunches. Nothing is easier than unwittingly imputing to historical actors presumed behaviors derived from one’s own wishes, fears and prejudices rather than from verifiable evidence; the only thing sillier than telling the dead what to do is expecting they’ll obey. Therefore, our inquiry begins with the artifacts themselves in light of what is known of the history of early sound recording, independently of what might be recorded on them.
The threshold question is whether there is anything about the cylinders themselves that precludes them from dating before February 19, 1894, the date of Sivori’s death. The answer requires a short excursion into early recording technology. The very earliest cylinders for Edison’s 1888 “perfected” phonograph were made of a yellowish and rather soft waxy substance. These are referred to as “white wax” or “yellow wax” cylinders. Some of these very early cylinders had cores of wound string. By late 1888, Edison replaced the wax with a ceresin-based darker-colored “metallic soap” which produced cylinders that were more durable. These latter are called “brown wax” cylinders. These brown wax cylinders were also more conducive to recording music.
There is as of yet no comprehensive and authoritative reference work for the dating of brown wax cylinders. The essential knowledge needed to create such a reference resides primarily in the minds and memories of passionate but aging private collectors and a sprinkling of archivists and academics who have devoted their lives to the hands-on study of these musical artifacts. Just as in the rare violin trade, opinions of experts regarding the age or origin of a particular artifact can vary because their opinions derive from the particular expert’s lifetime of personal experience and the particular artifacts which have passed before the expert’s eyes and ears. Nonetheless, brown wax cylinders have some characteristics which can establish some facts with certainty.
Core of British Library cylinder
String-core and white wax cylinders were manufactured only up to mid 1889 and are now as rare as Mozart manuscripts and Gutenberg Bibles. Edison abandoned string core cylinders because they were prone to break. From late 1889 through approximately 1897, the cores of brown wax cylinders newly manufactured by Edison contained a single spiral by which the cylinder was fitted to the mandrel of the machine. Around 1897, Edison began manufacturing cylinders with double spiral cores. Later, Edison began manufacturing cylinders with concentric circles and discontinued the single-spiral type. Thus, if any of the cylinders have cores with concentric circles or which are double-spiraled, they necessarily were manufactured after Sivori died.
The British Library’s Mystery Cylinders are all made of brown wax and have single-spiral cores. Therefore, it is entirely possible, but not certain, that they were manufactured and recorded prior to Sivori’s death on February 19, 1894. Dr. Michael Khanchalian is one of the world’s leading authorities on early cylinder recording and recording technology and has assisted museums and sound archives around the world. Dr. Khanchalian examined detailed photographs of the Mystery Cylinders. He compared them with exemplars from his collection and those he has examined in his decades of direct experience and concluded that the Mystery Cylinders are consistent with European brown wax cylinders recorded between 1891-1894, including exemplars from Dr. Khanchalian’s own collection.
Another factor pointing to extremely early vintage is the inherently experimental nature of the recording project itself. Almost all known early brown wax musical recordings of the early to mid 1890s, whether commercial or private, consist of short selections of two to three minutes duration – short enough to fit on a single cylinder or disc. This lasted into the early twentieth century. The keyword is “almost.” The Mystery Cylinders, by contrast, record entire classical concert pieces and concerto movements up to some 15 minutes in length on successive cylinders. In this, they are unique, at least among brown wax cylinders presently known to exist, and they present us with the most tantalizing of anomalies. To fully appreciate these anomalies requires questioning some widely accepted ideas about Edison’s attitudes toward music and resulting presumptions about the repertoire contained in the earliest cylinder recordings.
Edison and Classical Music
In much literature about early classical recordings, Edison appears as a caricature – a hard-nosed stone-deaf American philistine businessman interested only in how much money he might leech from his inventions, famously regarding his phonograph primarily as a business dictation machine and “serious” music as a commercially worthless waste of effort. The violinist Carl Flesch, who recorded for Edison from 1914 through 1928, claimed that Edison knew only two types of music: “good seller” and “no seller,” an opinion echoed by other classical musicians in the first decades of the 20th century.6
However, Edison as a crotchety and crankish but ruthless laissez-faire captain of industry is an incomplete portrait. Edison at 80 was not Edison at 40. The Edison of the 1880s surrounded himself with musicians, both in the laboratory and in his private life. His chief recording engineer, Theodore Wangemann, a trained musician from Germany, was the piano accompanist in many of Edison’s earliest recordings. Wangemann described his duties as: “Experimenting on phonograph recording with a view to making better musical records, vocal and instrumental.”7 It was Wangemann who recorded Johannes Brahms in December 1889.
The earliest Edison phonographs were sophisticated precision scientific instruments and works of art created by some of the most skilled engineers and craftsmen in the world. They were powered by electrical batteries and listened to through tubes that channeled the sound directly into the ear. While, with sufficient training, office workers might use them for stenographic purposes, recording classical music successfully was quite another matter. The violin and other stringed instruments were among the most difficult of all to record successfully and required technical expertise. To successfully record entire violin compositions over a span of several cylinders necessarily required at least one, and most likely two, highly trained technicians.
Some time prior to 1880, an abiding friendship arose between Edison and the Hungarian violin virtuoso and political revolutionary Edouard Remenyi. While Edison was working on the phonograph, Remenyi frequently visited Edison’s Fifth Avenue offices after concerts and the two would talk philosophy late into the night. Their jocular correspondence suggests that they were on intimate terms. For example, on August 19, 1881, Remenyi wrote to Edison:
"Since I was with Victor Hugo and Liszt I never was so much in a [sic] intellectual heaven as day before yesterday--I was wide awake, still I was in a dreamland, and I want to remain there, and to nourish myself on that heavenly food--and in the same time I do not wish to be so terribly in debt toward you--otherwise I will be soon bankrupt,--therefore prepare yourself immediately--if not sooner, to a musical assault on your doomed head--and then, only then we will be even.” (Remenyi adds a postscript: "looking at your photo--I invent also all sorts of melodies--you bet.")8
Reminiscing to journalists at age 70, Edison recalled Remenyi as “a long-winded talker . . . a Socialist or something” who would spontaneously take up his violin. “He would sit there talking, and bye the bye start playing the most beautiful things – wailing soft music. He’d play two or three thousand dollars worth every night.” Edison was a pallbearer at Remenyi’s funeral in 1898. Edison, Remenyi and their mutual friend, the musician Julius Fuchs, also were involved in an effort to present German operas at the Metropolitan Opera. Later, Fuchs asked Edison to recommend Fuchs for the position of musical director of the Metropolitan Opera to its Board of Directors.9
In March or April 1889, Edison met personally with the great German pianist and conductor Hans von Bulow, who asked Edison to record von Bulow’s performances in Boston. Wangemann testified in legal proceedings: “I left for Boston on very short notice, as I remember about three hours, von Bulow having asked Edison if his playing of the Beethoven sonatas could be done in one of his concerts, and Mr. Edison ordered me at that time to go to Boston and take them.” Wangemann spent some two months in Boston recording some of Boston’s finest classical musicians. One newspaper reported that Wangemann even “attempted what he considers as the most difficult test, a string quartette (sic), especially with ‘piano’ passages, the Listemann Quartette furnishing the instruments.”10
On May 2, 1889, von Bulow gave his farewell concert in New York, conducting Brahms’ Tragic Overture, Haydn’s Symphony in B flat (no. 12), Meyerbeer’s Struensee Overture, Beethoven’s Eroica Symphony and Wagner’s Meistersinger Vorspiel. According to numerous news accounts, Edison had the entire performance recorded on cylinders, using four phonographs to record the entire concert. Bulow left before he could hear the cylinders, but according to reports, others heard them and marveled at their high quality.11 On June 11, 1889, Wangemann or his colleagues appear to have recorded an orchestral performance by the Wagner disciple and future conductor of the New York Philharmonic, Anton Seidl.12
If one accepts the caricature of Edison the Philistine American Businessman, the idea that Edison, while engaged in intense scientific work and presiding over his numerous litigation-heavy interrelated businesses, would have any interest in the musical directorship of the Metropolitan Opera, or the late-night philosophical musings and improvised concerts of an eccentric Hungarian violinist, or the making of expensive but commercially useless large-scale recordings of entire two-hour classical music concerts is counterintuitive – downright preposterous. However, these documented actions reveal a side of Edison, at least while engaged in creative experiment, quite at odds with the Edison of popular imagination. More important is the freedom and encouragement he gave to Wangemann to experiment, at great cost in time and money, to record the greatest classical musicians then at hand.
But what has any of this to do with the British Library’s Mystery Cylinders? The answer is that Wangemann’s and Edison’s documented experiments with recording classical music, including entire works over numerous cylinders with multiple machines, provide a documented precedent and a context for the multi-cylinder whole-movement performances that the Mystery Cylinders preserve. The only recordings presently known to have been made bearing any similarity to the Mystery Cylinders were the cutting-edge experimental recordings made through Edison at the very beginning.
Is Wilhelmj a plausible candidate?13
The British Library based its provisional attribution to Wilhelmj on (1) what the donor said he had heard four decades earlier, (2) the fact that Wilhelmj was active in London from his arrival in 1894 until his death in 1908, and (3) that Wilhelmj had a decades-long connection with the Schott company as an editor and arranger. At first glance, this provisional attribution seems quite solid. But there are three difficulties which make Wilhelmj a singularly implausible candidate. First, there is no evidence that Wilhelmj ever met Sivori, or had access to, let alone performed, Sivori’s unpublished Second Violin Concerto. Second, the version of the Witches’ Dance on the cylinders bears little resemblance to Wilhelmj’s own edition of the work. Finally, by the time Wilhelmj arrived in London in late 1894, he had given up public performance despite public demand and generous offers from impresarios.
Wilhelmj was born in the village of Usingen, Germany in 1845. He studied violin with Ferdinand David, to whom Mendelssohn dedicated his violin concerto; he studied composition with the composer Joachim Raff. Wilhelmj was on intimate terms with Wagner and his circle; Wagner chose Wilhelmj as his concertmaster at Bayreuth for the premier of the Ring cycle, and again for the 1877 Wagner Festival in London. During his short performing career, Wilhelmj was known for his performances of the Beethoven concerto, the Bach Chaconne and other high classical works, and his heroic, statuesque “Classical” stage presence.
In 1908, Albert Franke, chairman of the “Usingen Beautification Society,” wrote to Wilhelmj asking if he would donate his musical papers to the local history museum. Franke received no reply because Wilhelmj was gravely ill and then had died. Wilhelmj’s widow found Franke’s request and readily sent numerous boxes of letters to Usingen, followed by other memorabilia including more letters, documents, albums, portraits, manuscripts and other items. The town of Usingen formally established and funded a Wilhelmj Archive, which ultimately came to be housed in the Usingen municipal history museum, where the materials are now stored. Franke wrote to Wilhelmj’s other remaining relatives and descendants, who also readily supplied additional materials to the Wilhelmj Archive. Franke also visited Schott’s London office and obtained 73 folders containing Wilhelmj’s various editions, compositions and transcriptions which Wilhelmj published through Schott up until his death.
Over the next century, other Usingen residents passionate about preserving Wilhelmj’s legacy continued collecting Wilhelmj memorabilia and adding to the archive. The Wilhelmj Archive presently consists of twenty-three large storage boxes containing Wilhelmj’s scores, manuscripts, a manuscript autobiography, programs, scrapbooks, and almost five decades of correspondence between Wilhelmj and his family – in essence, all the papers which the emissaries from Usingen gathered upon his death and through most of the twentieth century. It is a true fortune that a town of less than 15,000 inhabitants created and has maintained such an invaluable and irreplaceable resource and made it available to Wilhelmj researchers and admirers. Because Wilhelmj himself was a meticulous collector of printed material pertaining to his own career, including hundreds of newspaper reviews and programs, a thorough examination of these materials necessarily provides an idea of his repertoire and acquaintances.
Wilhelmj’s performing career was notoriously short. His last documented major public concerts as a violin soloist occurred in Germany in 1890. In December 1893, a year before Wilhelmj resettled permanently in London, the secretary of the London Philharmonic Society wrote to Wilhelmj asking whether Wilhelmj wished to be included in the list of soloists for the coming season. The tone of the letter suggests it followed up on prior correspondence: “do you wish to be included or not in this list? I [illegible] be sending the proof to the printer this week and will esteem your reply one way or the other; so please let me hear from you.” The Archive does not contain a response; there is nothing in the 1894 musical news of Wilhelmj performing publicly in London in 1894. Wilhelmj also appears to have agreed to perform at a concert in Nottingham on March 9, 1894 but was “unavoidably absent,” requiring a last-minute substitute.14
From his arrival in London until shortly before his death, Wilhelmj received numerous offers to perform, none of which he accepted. On November 22, 1905, an A. L. De Robert, of De Robert’s Music House in New York, tried to induce Wilhelmj to tour America from October 1906 through April 1907, promising to make it a “most phenomenal success.”15 In 1906, Daniel Mayer, “Sole Agent for Mischa Elman,” wrote to Wilhelmj: “Will nothing induce you to accept a few concerts in Germany? Mannheim especially is very desirous of fixing an engagement with you and I have no doubt they would pay about 2000 marks or even more to have the pleasure of securing your services. Kindly let me hear from you on this matter.”16 Later, Mayer sent an equally unsuccessful follow-up.
Similarly limited was Wilhelmj’s documented performance repertoire. This can be seen by examining the voluminous notebooks and programs documenting his entire performing career, held by the Wilhelmj Archive. Wilhelmj’s active repertoire consisted of the major concerti then in fashion, including the Beethoven, Bruch G minor, Mendelssohn, Vieuxtemps’ fifth, and several other concerti, his own reworking of the first movement of Paganini’s First Violin Concerto in his own arrangement, a violin concerto by his composition teacher Joachim Raff, the Bach Chaconne with and without added accompaniment, several other Bach compositions, his own compositions, and a few other then-standard virtuoso works. Nowhere among these is there any composition by Camillo Sivori, neither the concerto nor anything else. Nor is there any mention of Sivori in the decades of correspondence spanning Wilhelmj’s entire career, from touring virtuoso to London violin teacher, arranger, and violin dealer.
Prominent in Wilhelmj’s own performing repertoire were his Wagner transcriptions and paraphrases, including his transcription of the Albumblatt (originally for piano), the Preislied from Die Meistersinger, the song Traume, and paraphrases on Siegfried and Parsifal. These memorialized Wilhelmj’s roles as an intimate member of Wagner’s circle and Wagner’s concertmaster at Bayreuth and the London Wagner Festival, roles which were central to his musical and public identity. Other exceedingly popular arrangements in Wilhelmj’s performance repertoire were taken up by nearly every major violin virtuoso, by amateurs and students, and are played to this day. These include the Bach “Air for the G String” (extracted from the Bach D major Orchestral Suite and transposed) and the Schubert Ave Maria.
Wilhelmj gained fame as “the German Paganini” and was celebrated for his performances of several of Paganini’s compositions and his technical and tonal prowess. However, his actual Paganini performing repertoire was limited to the first movement of Paganini’s First Concerto in Wilhelmj’s own “modernized” arrangement, the Moses variations, several caprices arranged as an “Italian Suite,” and perhaps a few other pieces. Notably absent from these is Le Streghe, the “Witches’ Dance” memorialized on the Mystery Cylinders.
After abandoning public performance, Wilhelmj devoted his time to teaching, dealing in violins, and editing and arranging dozens of compositions for Schott by other composers, the bulk of which he never performed. This continued almost up to his death in 1908. Wilhelmj’s edition of Le Streghe dates from approximately 1905, well over a decade after he abandoned public performance. Le Streghe was never in his active repertoire. Wilhelmj’s edition of Le Streghe is substantially identical to the first edition published in 1851, with added fingerings and bowings but no substantial changes to the music.
The version of Le Streghe on the Mystery Cylinders bears little resemblance to Wilhelmj’s edition. Even the simple theme on which the variations are based differs in key ways from Wilhelmj’s version and from the 1851 first edition. The variations appear in a different order and are frequently interrupted by a strange wailing repetition of the theme, apparently intended to represent the “old woman’s song” noted on the cylinder boxes. There also are a coda with chromatic octaves and other passages which do not appear in Wilhelmj’s version, or in the 1851 first edition, or in Paganini’s manuscript. This raises the obvious question of why Wilhelmj, a violinist who consistently refused to play publicly, would record a multi-cylinder non-commercial version of a famous virtuous composition which bears scant resemblance to his own edition of the work, and perform it in a manner completely at odds with his reputation as a classicist and Wagner’s concertmaster.
And then there is that curious copperplate inscription on the boxes: “as played by Paganini.” Wilhelmj was born five years after Paganini’s death. If Wilhelmj was indeed the performer, that wording must be mere hyperbole because Wilhelmj could have no direct knowledge of how Paganini played. Perhaps Wilhelmj was performing in accordance with his own notion of Paganini’s style. Or – perhaps – there is another candidate who might give more substance to that curious phrase.
The Sivori Concerto is even more problematic. There is no evidence in the voluminous and meticulously collected papers in the Wilhelmj Archive that Wilhelmj ever met Sivori, or performed any of Sivori’s works, or had access to Sivori’s manuscripts, let alone made a recording of an unpublished 1841 violin concerto movement which had not been publicly performed in London since 1871 or anywhere else since 1877.
Wilhelmj demonstrated a reverence for his earlier colleagues: the Wilhelmj Archive contains a manuscript Suite for solo violin dedicated and inscribed to Wilhelmj by his teacher, Ferdinand David (1810-1873), a manuscript by Heinrich Wilhelmj Ernst (1812-1865), and a manuscript prelude and fugue by Henri Vieuxtemps (1820-1881) which Vieuxtemps inscribed to the great Henri Wieniawski (1835-1880). However, the Wilhelmj papers contain no manuscripts or memorabilia from Sivori. None.
Nor does a thorough review of Wilhelmj’s public performances through the digitized newspaper archives created by the British Library, the Bibliotheque national de France and the Library of Congress disclose any performance by Wilhelmj of a Sivori work. Nor does Sivori appear in Wilhelmj’s four decades of correspondence. This raises the question of how Wilhelmj managed to obtain the Sivori concerto manuscript, and why, after famously abandoning public performance entirely, he would learn it and memorialize it on record.
Much of Wilhelmj’s reputation as the German Paganini derived from his version of the first movement of Paganini’s First Violin Concerto. Wilhelmj created a lush, late Romantic “modernized” orchestral accompaniment and added melodic material to the solo part to comport with the reworked orchestra accompaniment. Nothing in the Wilhelmj Archive, nor in the American or British press suggests that Wilhelmj performed any other version. The version on the Mystery Cylinders is not Wilhelmj’s, and contains none of the obvious changes Wilhelmj made to the solo part and accompaniment. This raises the same question as the Witches’ Dance recording: Why did Wilhelmj record a version other than his own?
Next there is the question of why Wilhelmj, famed intimate of Wagner, did not record any of his own very popular Wagner arrangements with which his name was associated, or why Wilhelmj did not record his most widely played transcriptions, namely the Air on the G String, the Schubert Ave Maria, and the others with which his name is linked to this day. When Joseph Joachim recorded commercially for the Gramophone & Typewriter company, he chose works with which he was intimately associated – two Brahms Hungarian Dances in Joachim’s own arrangements, two movements from the unaccompanied Bach partitas, and Joachim’s own early Romance for violin and piano. Similarly, in 1904, Pablo de Sarasate recorded seven of his own Spanish Dances and his popular transcription of a Chopin Nocturne. In 1912, Eugene Ysaye (1858-1931) recorded virtuoso pieces by his teachers, Wieniawski and Vieuxtemps, and two of his own compositions in addition to one short piece each by Brahms and Fauré. What might have propelled Wilhelmj, long retired from the stage, to make complicated non-commercial recordings of forgotten compositions which were never in his known repertoire – but include nothing for which he was famous – is a question which presently eludes credible explanation.
Presently there is nothing connecting Wilhelmj to the Mystery Cylinders beyond what the donor recalled that he had heard decades ago at the time he retrieved them from the garbage. Unless additional evidence is discovered, there is no basis for attributing the Mystery Cylinders to Wilhelmj, beyond the donor’s vague recollection, Wilhelmj’s presence in London at the turn of the last century and his work for Schott, and we must look elsewhere.
“But if not Wilhelmj, then who?”
This was the question raised over dinner by the British Library’s curator of classical music recordings upon learning of the difficulties with the Wilhelmj attribution in 2018. If one is confronted with a recording of an unpublished composition, the manuscript of which is lost, but of which there is no documented performance by anyone other than the composer, the answer ought to be obvious – namely the composer. But nothing is obvious about the Mystery Cylinders because, by their nature, they lead the investigator perilously close to a nether region of fraudsters and hoaxters, and into a dark historical murk as yet uncharted.
We begin with the Sivori Concerto: unpublished, manuscript lost, cylinders fished from a refuse bin half a century ago, no information about who oversaw the recordings or who wrote the labels, no known trace of the composition in any institutional archive. Unless we can demonstrate that the piece on the Cylinders is indeed the first movement of Sivori’s Second Violin Concerto, there is no more reason to attribute the Mystery Cylinders to Sivori than there was to Wilhelmj. And if it is not the Sivori Concerto, we are left with no plausible candidate because the universe of possible violinists who could have made the cylinders expands to every technically proficient fiddler active when the cylinders could have been made.
This might seem like contrived quibbling. But few guilty pleasures are as universally appealing as a well-wrought pretension-puncturing hoax. The internet is awash with fake “historical” recordings of Oscar Wilde, Walt Whitman and others. There is even a film of Franz Liszt himself giving a master class, in a preposterous white wig, surface noise added to what is obviously a modern film doctored to look old. These Barnumesque wonders are too often embraced unquestioned by people who should know better. Antique phonograph hobbyists and collectors frequently shave defective early brown wax cylinders and use them to make their own cylinder recordings on original early phonographs lovingly restored to pristine working order. They have done so for years, simply for the joy of it.
Anything relating to Paganini is, for reasons unknown, particularly inspiring to tricksters. In 1900, the renowned violin maker Giuseppe Fiorini forged a “daguerreotype” of Paganini himself in an extravagant pose embodying all the cliches regarding Paganini’s appearance, copyrighted it, sold the “original” to a collector, and allowed it to be included in countless books and magazine articles as genuine, as it sometimes is presented to this day. In 2000, Giuseppe Gaccetta, an elderly Genoese carpenter, gained international attention by claiming a direct pedagogical lineage to Paganini via Sivori and Francesco Sfilio, who Gaccetta claimed was one of Sivori’s last pupils and with whom he claimed to have studied.
Gaccetta, who claimed to have inherited Paganini’s “secret,” passed off a 1970s commercial LP recording of several Paganini Caprices as his own, insisting that he recorded them as a teenager in 1931 on wax cylinders in Genoa under the direct guidance of Sfilio, the last exponent of the Paganini tradition. The City of Genoa gave him a medal; the President of the Republic of Italy gave him the title of “Commendatore.” Numerous internationally recognized academics and professionals caught in the whirlwind of wishful thinking fully endorsed Gaccetta, some even after the fraud was discovered.
Had Gaccetta enlisted the help of an antique phonograph hobbyist, simply recorded the 1970s LP through a horn onto brown wax cylinders and rolled them around in some moist dirt, his hoax might yet be undetected. Because we do not know who wrote the labels on the Sivori cylinder boxes and under what circumstances, there is no basis to presume they are true without corroboration. However, if the labels are true, the possibility of a hoax is eliminated because the pool of possible candidates is necessarily limited to those who had access to the solo part of the Sivori manuscript. That pool presently consists of one known person, Sivori himself. However, if the labels are false, then the pool of violinists who, as a logical possibility, could have recorded the cylinders swells to include every fiddler with access to a cylinder phonograph, and we can make no guess about the performer’s identity based on direct evidence.
An implausible coincidence and another treasure snatched from oblivion in the nick of time
And now for our first implausible coincidence. In 2021 a curious manuscript, privately owned, turned up in Italy, a “Cantabile Moderato” which consists of a twenty bar musical fragment in E major consisting of a solo line with piano accompaniment on two pages of hand-lined paper.17 The inscription reads: “Alle Signore Sorelle Branca / distintissime dilettanti di canto / questo semplice saggio musicale / dedicava in segno di distinta stima / Camillo Sivori / Vienna 23 aprile 1841.” (“To the Branca Sisters / distinguished amateurs of singing / this simple musical essay / dedicated in token of distinguished esteem / Camillo Sivori / Vienna 23 April 1841.")
Sivori Cantabile Moderato manuscript, first line
Sivori Cantabile Moderato manuscript, last line and signature.
The fragment resembles a typical musical album leaf of the type once popular among people rich enough to entertain musical celebrities. Sivori does not indicate whether the solo line is intended for voice or violin or some other instrument, or whether it is part of a larger composition. It is, in fact, the second major theme of Sivori’s Second Violin Concerto which he premiered in Vienna three days earlier, notated with careful attention to expression markings and phrasing. We know this because it matches the music on the Mystery Cylinders.
Sivori Concerto Moderato Cantabile
Because Sivori chose not to identify the source of his musical souvenir, it could not possibly be identified now but for the existence of the British Library’s Mystery Cylinders. This highlights the immense importance of these precious musical artifacts.
But for the Mystery Cylinders, this precious and irreplaceable fragment easily could have been forever lost in a dealer’s or collector’s stock of countless similar nondescript “autograph musical quotations signed” and never been identified for what it is. The only way to identify other fragments of the Concerto, should they turn up, is by being thoroughly familiar with the recordings. On April 30, 1863, Sivori inscribed the opening four bars of his Second Concerto to an admirer and clearly identified it as such, also obtained through blind fortune. It matches the opening four bars on the first of the Sivori Concerto Cylinders.
Sivori 2d Concerto opening bars album leaf
Sivori Concerto opening
These manuscript items confirm that the composition on the Mystery Cylinders is indeed as described on the labels.
So far, we’ve only gotten through the curtain-raiser, and our inquiry is far from over. Just because Sivori himself is the only person presently known to have performed his Second Violin Concerto and to have access to the solo part of the manuscript doesn’t establish that, in his late 70s, Sivori had the physical capacity to make the recordings, or that he even had access to a phonograph. Unlike in the United States, at the time of Sivori’s death there was no established French or Italian phonograph recording industry, and the devices in Continental Europe were exceedingly rare. Then, we still have the questions of how and why, if the Mystery Cylinders were indeed recorded in Continental Europe, they were deposited in the desk of a London music publishing executive and forgotten. And what about the “Old Woman’s Song” and the walnut tree and the Witches’ Dance recording that does not match any published edition? There is much still to do.
Wary readers might begin to wonder where exactly they are being led. Is the author going to prove that, instead of being recorded by a violinist nobody’s ever heard of who died in 1908, the Mystery Cylinders were actually recorded by a violinist nobody’s ever heard of who died in 1894? We have already met Wilhelmj. In our next installment, we will get to know Sivori, a violin virtuoso from an earlier and wilder age when the distinctions between “classical” and “popular” were not so sharp, when a Beethoven symphony could occupy the same program as “Kathleen Mavourneen” or some other popular hit without the least condescension, and when a virtuoso violinist’s acceptable palette of emotional and expressive devices was far broader than today.
By Andrew O. Krastins
© 2023 by Andrew O. Krastins. All rights reserved
*****
- The London newspapers and even the Government Gazette in Madras, India reported on the young prodigy, describing him as a pupil of the legendary Paganini, who was then known outside of Italy only through travelers’ accounts and rumour. See the London Morning Herald, July 17, 1827, Morning Post, June 7, 8 and 23, 1827, Evening Mail, May 25, 1827, Government Gazette, November 27, 1827, available online through the British Newspaper Archive. The 1827 program is in the Krastins Sivori Archive.
- The closest examples are the astonishing Julius Block cylinders recorded in Russia and Switzerland between 1889 and 1927. These include truncated movements of Arensky’s D minor Piano Trio with Arensky at the piano and dozens of other fascinating recordings. However, all the compositions on the set issued by Marston Records are short enough or cut to fit on a single cylinder. A superb essay by John A. Maltese and Gregor Benko is available on the Marston Records website.
- Inzaghi, Luigi (2004) Camillo Sivori: Carteggi del grande violinista e compositore allevio di Paganini; Zecchini Editore, Varese, Italy, p. 15; Menardi Noguera, F. (1991) Camillo Sivori, La vita, I concerti, le musiche; Genoa, Graphos, p. 65.
- November 27, 1863, L’Europa, p. 2; See Retro-news/Gallica: https://gallica.bnf.fr/ark:/12148/bpt6k4774224r/f2.image.r=(prOx:%20%22Sivori%22%2020%20%22Damain%22)?rk=107296
- See Andrews, F. (1986); The Phonograph: the British Connection; City of London Phonograph Society; Hope, H. (2021); The Remarkable Life of Colonel George Gouraud: the Man who Brought the Edison Phonograph to Britain, Howard Hope’s pioneering and wide-ranging biography Gouraud.
- Flesch, C. (1957), The Memoirs of Carl Flesch, tns. Hans Keller; Rockliff Publishing Corp., London; pp. 289-291.
- “Legal Testimony, Adelbert Theodor Edward Wangemann, October 1st, 1903,” Edison Papers Digital Edition, accessed September 27, 2018, http://edison.rutgers.edu/digital/document/QP006059.
- Rutgers Edison Papers, digital edition document accessed on September 25, 2023: TAED https://edisondigital.rutgers.edu/document/D8104ZCN
- All of the Edison/Remenyi/Fuchs correspondence is easily accessed through the invaluable Edison Papers Digital Edition through Rutgers University at https://edisondigital.rutgers.edu/
- “Clipping, Boston Journal, April 20th, 1889,” Edison Papers Digital Edition, accessed September 25, 2018, Thomas Alva Edison Digital (TAED) http://edison.rutgers.edu/digital/document/SC89025D
- The Musical Times, June 1, 1889, London; Philip G. Hubert, “The New Talking Machine,” Atlantic Monthly, Feb. 1889; Musical Courier, v. 18, no. 19, May 8, 1889, p. 367; Philip G, Hubert, “What the Phonograph will Do for Music and Music Lovers,” Scribner’s Monthly, v. 46 (1893) p. 152-154; Hubert in Century Magazine, May, 1893 p. 153.
- A. Theo E. Wangemann, Walter H. Miller, Henry Hagen; First Book of Phonograph Records (1889), p. 2. This invaluable historical document can be accessed through the website of the Association of Recorded Sound Collections at https://arsc-audio.org/blog/2017/04/04/firstbook/
- The facts in this section are drawn from (1) my personal review of materials at the August Wilhelmj Archive in Usingen, Germany; (2) Detmar Dressel’s invaluable but extremely scarce memoir, Up and Down the Scale (1937); the Memoirs of Carl Flesch; early articles in The Strad magazine; and various news accounts in the British and American press.
- Nottingham Evening Post, “Concert at the Nottingham German Club,” 10 March 1894, p. 4; https://www.britishnewspaperarchive.co.uk/viewer/bl/0000321/18940310/038/0004 (accessed 25 September 2023).
- August Wilhelmj Archive (AWA) item W756 A.
- AWA, W756a,
- Krastins Sivori Archive/Branca liber amicorum (1841).