Twenty-First Century Musicology and the Player-Piano

This evening’s recital takes the form of an introduction to the Pianola. There have not been many such lecture recitals in the USA in the last hundred years, certainly not if one is to compare the statistics with those for other instruments. At least three musicians who have contributed to this rarified process are currently here in this hall, and it is both a privilege and an honor to be representing the instrument on this occasion. The Pianola is one of the most forgotten, most misunderstood, and most misrepresented musical instruments in the world. Those books on which we, as musicologists, might consider that we depend for the foundations of our research, are as full of errors as a decent British tea strainer is full of holes! Over a period of at least half a century, there has been no effective peer review in this field, because the general level of knowledge has been so low. As a result, sources of information that would usually be carefully vetted have been accepted without question, and these include coffee-table books written for the collecting community, and even manufacturers’ own advertising assertions.

According to Professor Craig Roell, in his universally quoted standard work, The Piano in America, 1890 - 1940, the Pianola that makes its appearance this evening should not really exist: “... by 1909 the 'push-up' piano-player market was dead, long overtaken by the self-contained player piano” (p. 41). Well, the American successors to the Welte-Mignon manufactured and sold “push-up” piano-players in the mid-1920s, which is roughly sixteen years later than 1909. Given that the overall period dealt with by Professor Roell’s book is of fifty years’ duration, that is an error of nearly 33⅓ percent. But, although he doesn’t say so, let us assume that Professor Roell means only foot-pedalled piano-players, such as the one on this stage. According to its serial number, this push-up was built somewhere between 1910 and 1911, in Garwood, Union County, New Jersey, in a factory whose buildings still cling to life, with a tall chimney proudly displaying the vertically depicted brand name of “IANOLA”, the initial “P” at the summit having presumably been removed along with its surrounding brickwork, as a result of some long forgotten lightning strike or gentle deterioration. Since we cannot ever record history with absolute accuracy, we remember the lives of our forebears more as a form of agreeably fictional novel, but nevertheless we do our best to encapsulate the spirit of the time with as much exactitude as we can muster. Looking down from his final resting place in Mountainside, New Jersey (near the city of Summit, where he resided for much of his working life), the inventor of the Pianola, Edwin Scott Votey, must be rather pleased that six of the seven letters of his seminal invention are still to be seen by the travelers on the former Central Railroad of New Jersey, not that many trains actually stop at Garwood these days.

The survival of six out of the seven gives us a literary accuracy of roughly 85 per cent, or a 15 percent error, and that seems to me like a sensibly achievable standard by which to judge our own work, and from which to extrapolate further judgments. Do we, as an academic discipline, get close to 85 percent accuracy when we report on the history of the player piano? Do we, indeed, play the Pianola, or cause the reproducing piano to play, with anything like 85 percent fidelity? The answer, regrettably, is that no, we most certainly do not! I don’t mean to single out Professor Craig Roell personally. To judge from his students' comments on the web, he is clearly a lively and friendly professor who has inspired generations of students at Georgia Southern University. In any case, he is by no means the only scholar to invent and pass on such imprecise information. But, if we are to base our own research, and our own performances, on our faith in the printed word and in audio recordings, we need a level of accuracy that secures the foundations of our work and allows us to develop our understanding.

Just to get the provenance of tonight’s particular piano-player about right, the instrument that became the Pianola was first developed, according to Edwin Votey’s own personal papers, in 1895, at his home in Forest Avenue West, Detroit, Michigan. It was manufactured in experimental quantities in 1897, at the Farrand and Votey Organ Company factory in that city, and was put on sale to the general public in the autumn of 1898, as a product of the Aeolian Company, of which Edwin Votey eventually became Vice-President. It remained a flagship product of Aeolian until the time of the First World War in 1914, being used in important concerts with the London Symphony Orchestra under Nikisch, and with the Lamoureux Orchestra under Camille Chévillard, in London in 1912 and in Paris in 1913, respectively. You cannot really argue that the US parent company was disinterested in such foreign developments, not least because it manufactured the instruments in the USA, but also because the President of Aeolian, Harry Barnes Tremaine, was present at the London concert, along with many well-known musicians, including Ignacy Jan Paderewski, and reports of the concert formed a notable part of Aeolian’s worldwide advertising for many years. That certainly is not an instrument that had dropped off the horizon before 1909.

This Pianola plays both 65 and 88-note rolls, which places it after the famous player manufacturers’ convention, held on December 10th, 1908, at the Iroquois Hotel, Buffalo, New York, at which the new industry standard for 88-note music rolls was agreed. Professor Roell dates the convention to 1910, which is out by two years –not so much perhaps– but it is really not a difficult task to ascertain that particular date, since the convention was widely reported in the music trade press. Here is an illustration of the June 1912 concert in London, at which the solo pianolist was Easthope Martin, chief demonstrator for the Aeolian Company in London, and a successful composer of ballads and occasional organ music. The concerted items consisted of the Grieg Piano Concerto and the Liszt Hungarian Fantasy, and in addition the well-known soprano, Elena Gerhardt, was accompanied in lieder by Hugo Wolf and Richard Strauss.

The earliest piano roll catalogue I have at home is the US Aeolian issue from July 1901. It includes music of various categories, in both Title and Composer sections, but also conveniently arranged in numerical order in the final pages. While it is impossible to list the exact totals of the different categories, since one or two titles had already been removed, nevertheless they are accurate to well within our 85 percent limit.

Accompaniment and Special Rolls, 234
Classic and Miscellaneous, 632
Dance, 315
Operatic, 197
Popular and Popular Classic, Songs, etc., 1148
Total = 2526

One should also be aware of what sort of music was included in these various categories, since our own perceptions of popular and dance music are very different from those of our great-great grandparents. We might not be too surprised that Albeniz and Arensky, as well Bach, Beethoven and Brahms, tended to be included under Classic music. But Popular Classics brought together some of the compositions of Elgar, Grieg, Liszt, Mendelssohn, Moszkowski, Rubinstein, and so on, as well as a few marches, and what might be slightly disparagingly referred to nowadays as “Salon Music.” The Accompaniment Roll section is exactly as one might expect, a mixture of classical songs and chamber music, reflecting in large measure the regular Pianola recitals that the Aeolian Company presented, without charge, on a twice-weekly basis throughout the concert season. But Dance Music is by no means predominantly ragtime. At an initial glance, the considerable majority of the content consists of waltzes, with marches and two steps coming in just ahead of the other types of dance: galops, polkas, lancers and mazurkas. At a very rough guess, there are probably just under one hundred cake walks and a few overt ragtimes, out of a total of 2,500 titles in the catalogue, which equates to about 4 percent.

By the time of the July 1905 Pianola roll catalogue, an explosion of titles had occurred, and it is no longer quickly possible to calculate the totals under each heading, though the number of pages devoted to the various sections gives a basic guide to the relative proportions. Classic music takes 63 pages, Operatic 43, Dance music 33, Popular 91, Miscellaneous 22, and Accompaniment music 38. The proportion of two steps to waltzes has definitely increased, but once again dance music is a long way from being the mainstay of the catalogue. There is not the space here to describe the repertoire that came to feature on piano rolls around the world, during the first thirty years of the player piano’s existence, or during the succeeding period in which companies such as QRS and Mastertouch (Australia) came to dominate the field. It seems to be true that those foot-pedalled rolls which appear on Ebay in the USA tend more towards popular music than those found in other countries. Probably Great Britain and Germany had the finest classical catalogues, although France took the lead in special, unplayable arrangements and compositions of contemporary music. But one must be wary of judging repertoire merely from surviving rolls; generations of owners and collectors will no doubt have heated their houses by throwing unwanted music rolls into the furnace, so that each successive decade destroys some of the evidence, according to its own particular likes and dislikes. Ragtime has done well since the Second World War, whereas Metrostyled classical rolls by Rimsky-Korsakov have not. The catalogues bear witness to a much wider repertoire than most musical historians have assumed.

On Tonight’s Program
Chopin, Scherzo in Bb minor
As an introduction, we need one decent length piano work, to make the point that the Pianola is a serious musical instrument. And, as it happens, the Aeolian Company, which made this original roll, prided itself on having perforated the entire opus of Frédéric Chopin. Incidentally, all the rolls in this lecture recital are transcribed and not recorded. The notion that the majority of the world’s music rolls was of the recorded variety is simply not true. Even Gershwin’s rolls, and Scott Joplin’s, were in the main not recorded, or at any rate not a direct representation of those composers’ actual playing. Music rolls are generally made on perforating machines that deal in punch rows, meaning that each horizontal line of punches is followed by a small pulling through of the paper, generally between 1/16th and 1/48th of an inch, prior to the next line being punched. A set number of punch rows per beat, throughout a roll, is an absolutely certain sign that the roll is not hand-played, because no human being can possibly play with the robotic accuracy that such a result would necessitate. Hands up those here who have counted perforations! Not many, I’ll be bound! Craig Roell (p. 41) implies that push-up players were innately unmusical, because they lacked phrasing or volume controls. He is in fact misunderstanding the article in Yale Review that he has read, but it is abundantly clear that from the very outset of the Pianola, in 1898, it had both phrasing (tempo) and volume (dynamic) levers. Subsequent developments undoubtedly allowed for more sensitivity in following the performer’s movements, but in November 1899, at a performance with the Kaltenborn Quartet in Mendelssohn Hall, New York, the instrument was already sufficiently agile that a reviewer in the Brooklyn Eagle noted the different quality of musicianship between two separate pianolists.

Grainger, Shepherd’s Hey
Grainger was a lovely, very sensitive man, with eccentricities, to be sure, but with a youthful enthusiasm for music that he retained throughout his life. I always think he had the best ear I have come across for the spacing of music across the keyboard. Stravinsky and George Antheil had other qualities, but their music is always a little “blocky”, whereas in this saucy arrangement of an English folksong, Grainger always manages to retain a certain lightness, even when playing virtually all the notes on the keyboard. “Shepherd’s Hey,” and its companion, “Molly on the Shore,” both published in 1914 by the Orchestrelle Company (Aeolian’s subsidiary for Great Britain, Europe, Russia, the British Empire and South America), are the first truly musical special arrangements for Pianola.

W. A. Mozart, Sonata in D for Two Pianos, K. 448, First Movement
Denis Hall and I played this entire work at one of our early concerts together in the 1980s. In those far-off days before computers (he still won’t have one in his house!), we took as our starting point a set of hand-punched rolls, made by Frederick Evans, a professional photographer and obsessive roll-puncher of a hundred years ago, who transcribed around 1,500 rolls over the course of his lifetime, and played them by means of a 65/88-note push-up Pianola, just like this one. We got the rolls copied, and I then edited them with Scotch tape, removing all those perforations that I didn’t want, and cutting in those that were missing by means of a single-sided razor blade. I thanked my lucky stars that I had started my career as a tape editor for Decca Records (known as London Records in the USA), and that by the 1980s I didn’t need any razor blades for shaving. Roll copying and creating has come on a long way since those days, but I still use a modified Apple //e (pre-Apple Mac) from around 1980, with a mammoth machine-code program that I wrote in hexadecimal, having de-soldered chips from the motherboard and added extra daughter and extension boards of my own design. Since almost nobody wanted Pianola concerts, I had a lot of time on my hands, so I took the liberty of creating my own operating system, which I called “Rexdos”. You say pianolists are obsessive? Never!

Stravinsky, Firebird - Kastchei’s Dance
Once Stravinsky had settled in Paris during the 1920s, he was taken on by Pleyel as something of a cause célèbre. I doubt his music rolls made much money for the company, but the associated publicity must have helped hugely with the bread and butter business of piano manufacture. Stravinsky did very well from the deal; not only did Pleyel act as his collector of mechanical rights worldwide, affording him a regular monthly salary, but he was also paid on five different counts for all the musical arrangements that he made. He was the composer, the arranger, and the performer of the music, and there were also mechanical rights, plus a payment for exclusivity, since the arrangements for Pleyel represented the first time that the works in question had been produced as recordings. In fact, it can be worked out from perforation counts that the rolls were not produced by means of a recording piano, so that Stravinsky was not the performer, although he was credited as such on the roll labels. At least one of the production manuscripts survives, that of Les Noces (solo Pleyela version), which was for a time at the Pierpont Morgan Library in New York. Having seen it for only five minutes, about twenty years ago, I would say it was a mixture of short manuscript sketches, instructions for how to arrange the roll from both the orchestral and vocal scores, and a few (though not many!) words of encouragement. The probable arranger, under Stravinsky’s guidance in this way, was Jacques Larmanjat, a minor French composer and the head of Pleyel’s music roll department.
In 1924 Stravinsky’s entire contract with Pleyel was bought out by the Aeolian Company in New York, which had the intention of publishing the roll arrangements in its proposed Duo-Art AudioGraphic series. Only Firebird was ever produced publicly, though once in the 1980s I heard a few excerpts of Petrushka on the PianoCorder cassette piano system, which sounded different from the Pleyela version to me. The progenitor of the PianoCorder was Joseph Tushinsky, a millionaire businessman and owner of Marantz, who had a fine collection of Duo-Art rolls. I always wondered whether he might have ended up with a few Duo-Art trial rolls, perhaps from the collection of W. Creary Woods, the main New York Duo-Art recording producer, the majority of which ended up at the International Piano Archive at the University of Maryland. But I doubt they exist any more, since by now they will have been through many hands, nearly all of whom will not have understood their importance. Unfortunately the Duo-Art uses the end four note positions at each end of the tracker-bar for the special coding that controls dynamics. For most music it doesn’t matter, but for Kastchei’s Dance Stravinsky uses some of the very highest and lowest notes, which makes the Pleyela rolls much more gutsy, in all the right places!

Mendelssohn, Piano Trio in D minor, Op. 49 - First Movement
The tradition of playing instrumental chamber music with Pianola goes right back to the beginnings of its history. The very first public performance with a stringed instrument can be traced to a free recital on October 8th, 1898, at the Aeolian premises at 18, West 23rd Street in Manhattan, when the cellist, Hans Kronold, was accompanied in a performance of “Walther’s Prize Song,” with an unidentified pianolist, probably Francis Lincoln Young, who went on to invent the Metrostyle, Aeolian’s main interpretative device from the decade before the reproducing piano took over. To give a feeling of the continuity of musical history, here is a photograph from the same Aeolian premises, of a rehearsal for the Mendelssohn Hall concert mentioned briefly above, which took place just over a year later, on November 15th, 1899. The pianolist is probably Charles Parkyn, formerly cellist and manager with the Boston Philharmonic Orchestra (no connection with today’s band), who briefly became Aeolian’s concert manager at the turn of the twentieth century. The notion that such a performance could have taken place with an instrument that was inherently unmusical is completely out of the question.

Nancarrow, Study no 21 - Canon “X”
Conlon Nancarrow was a friend of Denis Hall’s and mine for a while. He even came to dinner! On one of his visits, BBC Television came to my house in south-east London, but on their return to the studios, alas, it turned out that their camera had been faulty, and they lost the entire day’s filming. Conlon had a way of speaking that made one smile at the end of a paragraph, and I’m aware that I do the same thing with exclamation marks, including at the end of many of these informal program notes. I think that is one of the secrets of his music too, in that there is always something slightly unexpected at the end: it’s a way of bringing humanity to a potentially mechanistic musical world. He wrote according to the most complex rhythmic theories, but in the end if it didn’t sound right, then he altered the result until it did, which I approve of heartily. Study no. 21 is a good example of how Conlon contrives to let all hell break loose, though gradually so, and with a powerful sense of control. It is based on a simple two-part polyphony, with one voice beginning slowly, and the other fast. Gradually the faster voice slows down, and the slower one speeds up. In the middle they run together very briefly, and then the accelerating strand gets faster and faster and faster. Bang! Human pianists (and instrumentalists) sometimes want to play Nancarrow by hand, which generally seems to spoil the music for me, I have to say. They never play it as fast, or as accurately. Conlon wrote his music as absolute music, not as mountains to climb. The forcible erection of a mountain range gets in the way of the music; it compels us to admire the pianist’s climbing techniques, which is by and large a ridiculous stratagem, as it is with all music. Beware modern pianists who wave their arms about, wear outlandish clothing or pull dreadful faces - they are almost certainly charlatans, though in our remarkably naïve world they do seem to make a lot of money!

Mozart/Busoni, Duettino Concertante, from Piano Concerto no. 19 in F, K 459
Denis Hall has a fine 78 rpm recording of this lively work, played by the Portuguese pianist who ended up as Liszt’s last surviving pupil, José Vianna da Motta, plus one of his own pupils as secondo. It is wittily done, and since the duration of 78s is rather short, it is split across two sides. Side two is just a little faster –perhaps they had a schooner of port during the break– but that makes it all the more human and exciting. It seemed like an ideal candidate for music roll, but hitherto I have always played it on my own, so this will be the first time à deux, and I made the roll last week. Appropriately enough, Busoni was for a short while interested in the possibilities of the Pianola. He made a special arrangement of Mozart’s “Magic Flute” Overture for 65-note roll, and there is even a short one-page sketch entitled “Für die Pianola,” which someone ought to use one day as a basis for a new composition. In German the Pianola is actually neuter in gender –Das Pianola– but we’ll forgive Busoni, who being Italian, must have thought it was the same as in his own language, where it has always been La Pianola. Indeed, there is even a village in Italy called Pianola, which it won’t surprise you to learn was once the focus of one of my overseas jaunts. By the way, just to make matters even more complicated, it’s Le Pianola in French!

Nancarrow, Study no. 6
In the late 1980s, Conlon had the intention of writing a concerto for me to play, and indeed he completed quite a bit of it, with no end of sketches, before his age and frailty prevented its completion. Finally, in 2004 it was transformed into “Nancarrow Concerto” by a wonderful British composer, Paul Usher, but it took the Westdeutscher Rundfunk in Cologne to commission it, since BBC Radio has shown almost no interest in the player piano for nigh on forty years. Study no. 6 is unusual, in that it is rather quiet and pensive. It suggests to me someone sitting outside in the warmth of the Mexican sunset, watching the slight irregularity of life passing by. There is the same irregular regularity in the 5/4 construction of Rachmaninoff’s Isle of the Dead, which I reckon represents not the waves of a lagoon, but the inexorability of life in the late nineteenth century Russian countryside –but that’s another story. It’s also in Montañesa, one of Manuel de Falla’s Pièces Espagnoles, which he recorded so movingly for the Welte-Mignon. One can imagine the occasional mule on its way uphill ... but perhaps my imagination is beginning to run away with me!

Rachmaninoff, Tarantella from the Second Suite for Two Pianos, Op. 17
We need to end the first half with a bang. I suppose I’ve transcribed and perforated about four hundred rolls in my time, and I love the intellectual exercise of arranging two piano music for Pianola, which is rather like an emotional form of Sudoku. The Tarantella is one of the pieces that Denis Hall and I sometimes play in concert, on two Pianolas. We have found that it helps to face away from each other, which looks tremendously impressive, but actually it’s easier, because you can really hear what the other player is doing, much more clearly than when the sound of both pianos is coming from straight in front of you. Thank you to all the courageous and accomplished musicians who have volunteered to share their music-making with the Pianola this evening. It’s a fine tradition. And, by the way, do stay for the Rite of Spring in the second half! [Intermission]

Rachmaninoff, Prelude in Eb, Op. 23, No. 6
Rachmaninoff is my life’s musical friend. He writes music that is completely from the heart, and he makes me both laugh and cry, quite literally. Atheist that I am, I occasionally visit him near Lucerne, where his bust stands on a plinth near the lake’s edge. No-one else pays any attention to him, so he and I have conversations for a while, and when no-one is looking, I give him an affectionate pat on the head. He was such a clever man, as the fiendish complexities of the symphonies testify, and yet the compositional technique is always in service to the emotion. And he had his own Pianola, at his country estate at Ivanovka, a day or two’s travel south-east of Moscow. I think I know why and how he had it, but I won’t go into all that detail here. You can ask me if you want to know more. The early Pianola has received such a bad press, mainly from people who should know better, and I want to show you how a very simple 65-note roll can sound. There are some things I can’t do with this roll, like bringing out all the inner melodies as much as I should like, but with Rachmaninoff the timing counts for such a lot, and as in most music, the tempo lever never stands still. Rachmaninoff once contrived to provide a very personal memory of this Prelude: the BBC’s one venture into serious consideration of the Pianola occurred in 2011, when it commissioned “Airplane Cantata” from a British composer, Gabriel Jackson, as a result of one particularly sensitive radio producer, Michael Emery. During the first morning’s break in rehearsal with the BBC Singers, some of the sopranos gathered round and said, “Come on, Rex, show us how it works.” I played the Rachmaninoff Prelude first, and when I turned round to explain what I had been doing, one of the ladies was crying. Instinctively, and never having met her before, I walked straight over and gave her a long and enormous hug. That, for me, encapsulates what music, and Rachmaninoff’s music in particular, should be about. You don’t get on the concert platform to show off, or to earn a living; you do it to share, and when it works, it is utterly magical.

Stravinsky, Le Sacre du Printemps - Pleyela version
You won’t get a description of the ballet this time, not for an audience of musicians! Ever the businessman, Stravinsky managed to place the Rite on to no less than nine separate rolls, Pleyela 8429 to 8437. He made the same fees regardless of the length of the rolls, and I forgot to say earlier that Pleyel paid him for all the rolls they made, rather than just the ones that they sold. Boy, what a deal! Denis and I used to play this piece on two concert grands, alternating, so that I did 1, 3, 5, 7 and 9, and he interposed 2, 4, 6 and 8. Like playing back to back, it looks good, but following on from one another is pretty easy, though the instantaneous creation of regular rhythm from a standing start takes practice. But in the end the drama of the original music is better served by a single person playing it, or so it seems to me. There is more of a sense of continuity, and the audience is not distracted by the technicalities of roll changing. But of course you can’t do it with nine rolls, or the flow of the music would be completely lost. Luckily many of the original rolls are rather short, so I was able to transfer the entire ballet on to two rolls, breaking at the pause between Parts One and Two, though I did have to cheat by compressing some of the movements and therefore playing them at rather slower roll speeds. That is a potential problem in the final Danse Sacrale, where the rhythm needs to be white hot, and at the end of an extremely long roll you have to be very alert. At this point I’d like to mention and thank an invisible helper, Anthony Robinson, who is the finest expert on roll scanning in the world, in my view. If you are seeking to copy rolls, you have in most cases to produce an exact copy of perforation patterns, because so much visible information is included in them for the pianolist to read and comprehend. Anthony has what I regard as by far the best method of reading and analysing perforations, which very sensibly is not restricted to being fully automatic, and he is currently acting as a consultant for the Stanford University roll scanning project. He lives in a quaint little village in the English New Forest, called Nomansland, and is of a similar generation to Denis and me. We were all born and brought up at a time when the Allies had just won the Second World War, and when British ingenuity in particular had played a very significant part in codebreaking and outwitting the German High Command. Sometimes it feels as though there is something of a “Wartime Spirit” in the study and preservation of music rolls, and Anthony shares that stiff upper lip and sense of humor. I couldn’t have created these Sacre rolls without his ready and willing help.

That’s it - I hope you enjoyed the lecture. Thank you to all those who took part, and thank you all for the invitation!

–Rex Lawson