On Tour with Rachmaninoff: Analysing the Programmes from his 1924/25 Season

Abstract

Sergei Rachmaninoff is widely regarded as one of the great pianists of the twentieth century. In a research project that has stretched over two decades, I have compiled data on Rachmaninoff’s performance career, comprising research in archives as well as published sources in Russian and English languages. The resulting Rachmaninoff Performance Diary has been publicly available online since 2011. A missing link in the data has been the complete programmatic details of over 1,080 solo recitals. In 2006, I discovered research that was apparently unrecognized in its completeness in an archive of the Library of Congress, undertaken and donated by Rachmaninoff’s sister-in-law, Dr Sophia Satina. In this article, I examine the details of the 1924/25 season, which was a critical time for Rachmaninoff: after the collapse of his personal fortune caused by the Russian Revolution, he at last had achieved sufficient success and financial security from his hectic touring to allow him to return to composition the following year, his ‘sabbatical’ break of 1926. From the data, a clearer picture emerges of how Rachmaninoff varied his repertoire in his many concert appearances and recording sessions, showing how frequency of performance and, in instances, apparent self-assessment of his own music, were key factors.

Readers of Rachmaninoff biographies might be unaware of the existence of a small notebook in which the musician listed the date and location of every performance he gave from 1907. Prior to this time, Rachmaninoff’s performance activity had been growing in significance, yet his appearances in the main had been as a conductor, frequently of opera at the Bolshoi Theatre in Moscow. ¹ As the frequency of his work as a solo pianist increased, especially after his emigration to the West in 1917, the notebook subsequently recorded the performance career of one of the early twentieth century’s most noted musicians.

After his death, his sister-in-law, Dr Sophia Alexandrovna Satina, came into possession of the notebook, and used it as the basis of research as she attempted to assemble a complete catalogue of every performance he had given. ² She wrote to cities and towns where Rachmaninoff performed, addressing correspondence to newspapers, libraries, and concert venues, asking for details of the programmes played. ³ Following a tradition inaugurated by her brother-in-law, and continued by his wife, Satina gave her materials to the Library of Congress in Washington, DC. ⁴ Attached to a letter preserved in the archive is a newspaper cutting from April 1945: written by one of her small team of helpers, it is a ‘letter to the editor’ of a local newspaper, and it outlines how Satina was attempting to ‘collect a list of all the programmes given by Rachmaninoff at his many concerts’. ⁵

Satina’s data are significant for those interested in Rachmaninoff’s career as a concert pianist. Her materials broaden our knowledge of the pieces that were played on long tours that stretched across North America and into Europe, providing information on how he varied his repertoire, and how he approached the structuring of his recitals over decades. While details of almost every concert given by Rachmaninoff were published at the time (through printed programmes, advertisements, and concert reviews), my research, which takes the form of a digitized chronological listing of concerts, was compiled initially from the scholarship of other authors. The first avenue of research was an appendix to the third volume of Zarui Apetian’s Literaturnoe nasledie, at the time representing the most detailed listing of every known performance date and venue from the 1909/10 season to Rachmaninoff’s final concert in Knoxville, Tennessee, on 17 February 1943. ⁶

Where Apetian supplied details of works played in specific cities, it usually involved naming the concertos Rachmaninoff performed, or the symphonic works he conducted. In terms of solo repertoire, she listed a compendium of works played each season only. Details were collated from Barrie Martyn’s Rachmaninoff: Composer, Pianist, Conductor, which also listed solo repertoire, but more notably supplied excellent data on Rachmaninoff’s conducting and recording activities. ⁷ Data on first and other notable performances were sourced from A Catalogue of the Compositions of S. Rachmaninoff by Robert Threlfall and Geoffrey Norris, ⁸ while the early biography of Rachmaninoff by Sergei Bertensson and Jay Leyda was useful in corroborating concert dates and other details. ⁹ In a more personal way, two separate visits to the Rachmaninoff Archive at the Glinka Museum of Musical Culture in Moscow clarified precise information about certain early concerts. ¹⁰

Yet missing from all published sources on Rachmaninoff’s performance career to date has been a comprehensive listing of the specific programmes from all solo recitals, and it is here that the materials donated to the Library of Congress by Satina offer my project a pathway towards completeness. When viewed as a whole, the listed programmes reveal far more about Rachmaninoff than might be expected, to the extent that a retelling of his biography from the perspective of his performance career has at times seemed warranted. ¹¹ The present article provides a unique glimpse into the daily concert life of Rachmaninoff as a solo pianist in the concert season of 1924/25, the last complete season before he took a ‘sabbatical’ break to complete his Fourth Piano Concerto in 1926, delayed since leaving Russia. ¹² Given that this was his first attempt to return to composition away from his homeland, I believe that throughout the season he likely assessed the works he was playing from the renewed perspective of a fellow composer.

The presentation of these recital data offers new insights into Rachmaninoff studies. They offer a better understanding of how his programmes were ordered, the degree of variation between them as he toured, and the extent to which he chose to highlight his own compositions. ¹³ Through periods of programmatic uniformity, they show that in the main he presented the same works to audiences in the leading cities as he did in smaller towns, and that he had sufficient charge of his performance activity that accommodation of requests from local concert promoters in the main appears to have been avoided. The concert programmes from this season have been incorporated into my online Rachmaninoff Performance Diary, with data for all other seasons soon to follow. ¹⁴

Repertoire Old and New

After holidaying in Europe over the summer, Rachmaninoff’s concert season began on 2 October 1924 in Bournemouth, England. ¹⁵ Since late 1918, his concert activity had centred on North America, yet the first signs of a widening touring pattern were soon evident. For example, the conclusion of the 1921/22 season was marked by two recitals in London. Hence, with the 1924/25 season commencing in the United Kingdom, a sense of Rachmaninoff’s increasing international status further emerges. Owing to this growing fame as a performer, thought would have been given to Rachmaninoff’s concert schedule and the programming of works both old and new. While the summer cessation of concert-giving was long (March to October), at some point work was required to learn new repertoire. ¹⁶ As a consequence, further consideration regarding when – and where – new works might be premiered was necessary.

Providing definitive answers to such questions has been problematic without access to comprehensive recital data. Even so, in a footnote regarding recital programming, Martyn asserted that ‘trying out’ new repertoire in ‘out of town’ venues ‘became a hallmark of Rachmaninoff’s tours’. ¹⁷ While the following programme details largely confirm this claim, the instances in which major works were presented in important venues without any ‘warming up’ are, by contrast, significant. ¹⁸

In regard to the season under discussion, the return to the stage after his summer break appears not to have been overly pressured. The Bournemouth programme began with a collection of works by Chopin, opening with the Fantasy in F minor, op. 49 (Figure 1). ¹⁹ The Fantasy had been a feature of many programmes in early 1923, where it also began concerts. The Chopin set continued with the Nocturne in E major, op. 62, no. 2, and the Waltz in A flat major, op. 64, no. 3. As many programmes in the 1922/23 and 1923/24 seasons featured unnamed Nocturnes and Waltzes by Chopin, in this instance it is not possible to ascertain whether these specific pieces were frequent inclusions. ²⁰ The major work which followed – Chopin’s Sonata in B flat minor, op. 35 – had been a regular inclusion in programmes from late 1922, and had been a mainstay in concerts in the early months of 1924. ²¹

Figure 1. Recital, 2 October 1924.

As is typical in a majority of the programmes that Satina collected, there is no indication of where the interval occurred. ²² In programmes which included a major sonata in the first half (as all do in this season), it is assumed that the interval followed. ²³ Closing the programme with a work by Liszt was a feature of numerous programmes in earlier seasons, with La campanella and the Second Hungarian Rhapsody appearing regularly. The Rapsodie Espagnole in this programme had featured as a concert-closer in programmes from late 1923.

Of his own works, preludes from the op. 23 set had been performed extensively on Russian tours after their initial composition, and had featured in concerts given in the West in 1911 and 1914. Regarding the unnamed Etude-Tableau, more will be discussed about incomplete and problematic references later, yet both the op. 33 and op. 39 sets also had been performed in Russia. Rachmaninoff first performed his transcription of Mussorgsky’s Hopak in 1921, his initial experience of it having been the orchestral version, which he conducted in Russia on 27 April 1909 (Julian calendar). ²⁴ The Bournemouth programme was given identically in Liverpool (4 October) and at the Queen’s Hall in London (6 October).

In the fourth concert of the tour in Eastbourne on 9 October, a second programme was initiated, beginning with J. S. Bach’s English Suite in A minor, BWV 807, Mendelssohn’s Variations sérieuses, op. 54, and Beethoven’s ‘Appassionata’ Sonata, op. 57 (Figure 2). The pairing of the English Suite and the Mendelssohn to open a programme had been established in recitals from late 1923, while the Beethoven Sonata had featured multiple times across the 1922/23 season.

Figure 2. Recital, 9 October 1924.

Liszt’s Funérailles had also appeared in programmes from 1923, and would feature in programmes throughout the 1920s and up to 1935. Godowsky’s transcription of Johann Strauss Jr. melodies, known generally as Künstlerleben (‘An Artist’s Life’) but more formally as Symphonic Metamorphoses no. 1, concluded the concert. A popular work with audiences at the time, the Godowsky was in fact a new work for Rachmaninoff this season, and the concert on 9 October appears to have been his first performance. The piece would be included with great frequency this season, reaching a total of twenty-two performances, but thereafter seems never to have been performed publicly by him again. This programme was repeated in Leeds on 14 October, and again in London on 16 October.

After sailing from England, Rachmaninoff performed a slightly altered programme in his first US appearance of the season in Urbana, IL, a relatively small town. ²⁵ His new programme began with Mendelssohn’s Variations sérieuses from the concerts in England, which now acted as a curtain-raiser for Liszt’s masterpiece, the Sonata in B minor, S. 178 (Figure 3). Rachmaninoff would perform this great sonata a further seventeen times before the end of the year, but after this appears never to have then played it in public again. ²⁶

Figure 3. Recital, 12 November 1924.

Also incorporated into the programme in Urbana were further works by Chopin, including the Scherzo no. 4 in E major, op. 54, and the Ballade no. 3 in A flat major, op. 47. While Chopin’s Ballade had been a regular inclusion in programmes dating back to the 1921/22 season, the Scherzo no. 4 – like the Liszt Sonata – appears to have been a first performance. ²⁷ The Etude in E major, op. 10, no. 3, had been performed in Scandinavian concerts in October 1918, as too had the Etude in C minor, op. 25, no. 12, which would be added to the programme in Terra Haute, IN, two days after the concert in Urbana.

It is conspicuous that Rachmaninoff’s ubiquitous C sharp minor Prelude is listed on the programme for these concerts, given the many anecdotes about him being unable to leave the stage without including it as an encore in later years, much to his frustration. ²⁸ In Urbana, however, Satina noted that Rachmaninoff performed as an encore Giovanni Sgambatti’s arrangement of the famous ‘Dance of the Blessed Spirits’ from the second act of Gluck’s Orfeo ed Euridice, published as Mélodie. Along with the work by Liszt, this also may have been a first performance.

In Terra Haute a similar programme was given two days later, albeit in this instance commencing with two new works by J. S. Bach, the first being an arrangement by Liszt (Figure 4). In the second work, it might seem plausible that the Fugue was played after the Prelude, and that its absence in Satina’s list of programmes was due to her abbreviated nomenclature. ²⁹ This might especially seem to be the case given the preceding work included a fugue, yet there is no evidence for it. While original compositions by Bach, and arrangements of his music by others (including by Rachmaninoff himself), were a feature of programmes for the remainder of his career, this amended Liszt Sonata programme appears to represent the only time beyond his student years that Rachmaninoff programmed an item from the Well-Tempered Clavier. ³⁰ Among the selection of the composer’s own works, the two preludes are unremarkable, yet the inclusion of the Etude-Tableau in D major, op. 39, no. 9, is more noteworthy due to its advanced harmonic and stylistic elements, these being emblematic of his later style, as will be commented on further below. ³¹

Figure 4. Recital, 14 November 1924.

As a format, it is significant that the programme would be repeated with only small changes until the end of the year. ³² From late 1921, Rachmaninoff had increasingly been drawn into playing a well-honed programme of works as he toured, perhaps unsurprisingly given the increasing frequency of his performances. ³³ Yet from Terra Haute until the end of the year, these repeated programmes represent the longest instance of an unbroken stretch in his career thus far, proceeding without an alternate programme or an intervening concerto appearance. Hence, the programme was identical in Indianapolis, IN (16 November); Youngstown, OH (17 November); Buffalo, NY (18 November); Detroit, MI (20 November); Rochester, NY (21 November); Boston, MA (23 November); Pittsburgh, PA (26 November); and New York, NY (Carnegie Hall, 30 November). The only variation appears to have been on 25 November in Cleveland, OH, where two Chopin waltzes (op. 64 no. 2 in C sharp minor, and ‘A flat major’) replaced the two Chopin etudes. ³⁴

It might be surmised that these consecutive performances reflect the significance of later dates in Boston and New York, there being an added pressure with the B minor Sonata of Liszt on the programme. Yet at the same time, this programme marked Rachmaninoff’s 44th appearance at Carnegie Hall, and it seems unlikely he would have been particularly anxious by this point in his career. ³⁵ Similarly, it might also be supposed that the intensity of this tour presented limited opportunities to introduce new works.

Following the performance in New York, the Liszt Sonata programme continued its unbroken run of performances in concerts in New London, CT (1 December); Worcester, MA (2 December); Haverhill, MA (4 December); Manchester, NH (5 December); Providence, RI (7 December); New Haven, NJ (8 December); and Middletown, CT (9 December) – all of which were likely day-drives from Manhattan for Rachmaninoff, an avid motorist. ³⁶

Among the legion of successive concert dates, Rachmaninoff was in demand to record music for both piano roll and the gramophone. At times, one sees clear correlations between the recording of a work and its frequency in programmes of the time, while on other occasions there appears to be no obvious link. ³⁷ While Satina’s research does not focus on this activity, the ability to read her materials in parallel with published data about recording sessions offers an opportunity to gain a new perspective on Rachmaninoff’s work in the studio. The following details, including information on ‘takes’ and subsequent releases, have been sourced from the latter pages of Martyn’s Rachmaninoff: Composer, Pianist, Conductor. ³⁸

On 22 December 1924, Rachmaninoff returned to Trinity Church in Camden, NJ, with Leopold Stokowski and the Philadelphia Orchestra, to conclude a recording begun a full year earlier (Figure 5). The recording of the Second Piano Concerto had commenced with the second movement on 31 December 1923, and had concluded a few days later with the third movement. No attempt at the first movement had been made; rather, it was recorded the following year. ³⁹ The most recent performance of the Second Concerto had been in Cleveland in late March 1923, while it would be performed again with Serge Koussevitzky in April 1925. ⁴⁰

Figure 5. Recording session, 22 December 1924.

As was noted for the concert in Urbana, IL, on 12 November 1924, Sgambatti’s arrangement of Gluck’s famous tune from Orfeo was first introduced into programmes as an encore. Many of the recordings Rachmaninoff made of small piano pieces, which are still highly praised for their pianism, indubitably benefitted from frequent performances on stage, and this small work by Sgambatti would become a focus of recording in numerous sessions over the coming months. ⁴¹

A little over a week after he recorded the concerto movement, Rachmaninoff returned to the studio to record solo works (Figures 6 and 7). Along with the Sgambatti he played his more-frequently performed transcription of Mussorgsky’s Hopak, while the Chopin Waltz was not from the current season. Remarkably, Rachmaninoff appears never to have programmed Liszt’s transcription of the Schubert song, ‘Das Wandern’, nor Siloti’s arrangement of ‘The Swan’ by Saint-Saëns, in previous concerts, and neither would appear in later programmes. ⁴² None of the takes were marked for release, although the recording of ‘The Swan’ was eventually published. ⁴³

Figure 6. Recording session, 30 December 1924.

Figure 7. Recording session, 31 December 1924.

Following a fortnight’s break from performing early in the new year, Rachmaninoff returned to the studio mid-January, this time to record two piano rolls for Ampico (Figure 8). ⁴⁴ In this instance the Sgambatti arrangement, along with ‘Das Wandern’, were both released.

Figure 8. Recording session, 14 January 1925.

Conflicting and Incomplete Data

Despite Satina’s attempt to create a complete account of Rachmaninoff’s concert activity, in some instances there are discrepancies among sources. Similarly, for some concerts she was unable to ascertain precise details, such as for the reference to the unnamed Etude-Tableau in the preceding 2 October concert, while in other instances she was unable to find any programme details at all. Accordingly, across her collected data there are certain lacunae. Yet with the increased ease with which historic newspapers can be searched and read, and the accessibility of other online databases, it is now possible to resolve some lingering questions presented by her research.

Within this season, a discrepancy among sources arises in relation to a concerto performance in London on 11 October, which Satina assumed to be of the Third Piano Concerto although she was unable to supply details of the other works in the concert. Her note for this concert references a comment by ‘Mr Tillet’ (John Tillet, Rachmaninoff’s UK agent), indicating that the work could, in fact, have been the Second Piano Concerto, op. 18. Apetian included the date and location of the concert, also naming the conductor, but was unable to name the concerto. ⁴⁵

Given a more detailed and, hence, reliable programme that included the Third Concerto one week later (Figure 9), there is perhaps a tendency to view Satina’s assumption as correct. ⁴⁶ Yet confusingly for this second concert, Apetian indicated a belief that the Second Concerto was performed. ⁴⁷ It is salient to recall that Rachmaninoff by now had considerable experience performing both concertos, sometimes in very close proximity. ⁴⁸

Figure 9. Concerto concert, 18 October 1924.

However, modern research presents further information. The evening concert on Saturday 11 October 1924 at Queen’s Hall was a Promenade concert, led by their resident conductor Sir Henry Wood, who was correctly named. ⁴⁹ In a programme that ranged from Italian operatic excerpts by Rossini, Donizetti, and Verdi, to works by Granados and Alexander Mackenzie, the concerto was in fact the Piano Concerto no. 3 in G minor ‘Fantasia’ by York Bowen, performed by Bowen. ⁵⁰ Yet Wood was a busy conductor that day, as a review in The Times indicates that earlier in the afternoon Rachmaninoff had, indeed, performed his Third Concerto, in a concert that included Berlioz, Strauss, and Tchaikovsky. ⁵¹ While there is no indication whether Satina based her claim solely on the word of John Tillet, it seems more likely that Rachmaninoff’s notebook provided foundational information that she sought to corroborate with later sources. ⁵²

In one instance where Satina was unable to ascertain precise details about works, the concert itself presents a riddle. On 16 January 1925, Rachmaninoff gave a performance for President Coolidge and the First Lady at the White House (Figure 10). ⁵³ Given that the programme comprised many works that had not featured in any earlier programmes of this season, it is intriguing to speculate whether they were personal choices or requests from his hosts. ⁵⁴ His White House programme the previous year had also comprised works considered ‘favourites’, yet those pieces had mostly appeared within other programmes of the season, or in studio recording sessions. Of note, his arrangement of Mussorgsky’s Hopak was included in the White House programmes in both 1924 and 1925.

Figure 10. Recital, 16 January 1925.

In terms of the general references to works by Chopin, it could be speculated that the unnamed nocturne was in E major, op. 62, no. 2, as it was included earlier in the season. Yet two nocturnes from op. 15 had been recorded at the end of the previous season, as, too, had the C sharp minor Mazurka, op. 63, no. 3. ⁵⁵ It is logical to assume that the ballade was no. 3 in A flat major, while the Chopin waltz is ultimately unknowable given the number of these Rachmaninoff had performed. Hence, given that the data for this concert were supplied directly to Satina, and that other public records for the concert are unlikely to exist, in this instance it might never be known with certainty which works by Chopin were played. ⁵⁶

Of the other works, Tchaikovsky’s ‘Troika’ had been a programme favourite dating back to concerts in Scandinavia in 1918, ⁵⁷ while the Humoresque is a more interesting inclusion: between 1921 and 1923 Rachmaninoff had on five occasions attempted to record it, yet appears never to have programmed the work publicly before, nor again afterwards. ⁵⁸ His transcription of Kreisler’s Liebesleid had featured in many earlier programmes, and had also been recorded. Similarly, Moszkowski’s La jongleuse had been a feature of the 1922/23 season, and had been recorded; yet after the performance in the White House, Rachmaninoff appears not to have programmed it in a public concert for the rest of his life. The Waltz from Faust, in Liszt’s arrangement, is a work Rachmaninoff knew well, having first performed it in a recital in September 1892. ⁵⁹ In 1904, he had played Sarasate’s violin transcription of the same operatic number with Stanisław Barcewicz, while the version by Liszt featured again in the first American concerts of 1919, reappearing in programmes as late as 1938. The numerous questions prompted by this sundry collection of ‘favourites’ will likely remain unanswered.

Another potential discrepancy can be noted in a recital in Seattle on February 19 1925, when Rachmaninoff was in the middle of a long tour (Figure 11). ⁶⁰ He was alternating between two programmes, one of which included Chopin’s Scherzo no. 4 (Figure 3), yet Satina has listed Chopin’s ‘Scherzo in B minor, op. 20’. No prior performances of this work had been listed by her, nor did she note a discrepancy. While it is therefore tempting to dismiss the record as an inadvertent error, it seems possible that the B minor Scherzo was in Rachmaninoff’s repertory, as it appears to have been included in a concert in 1892, and it would be included in programmes from 1930 onwards. ⁶¹

Figure 11. Recital, 19 February 1925.

Yet another reason to perhaps doubt the accuracy of the Seattle programme is the sudden appearance of an Etude in A flat major by Chopin. The source for the 1892 programme is Martyn, who stated confidently that the B minor Scherzo and Etude in A flat major, op. 10, no. 10, were performed, yet he did not supply a reference . ⁶² As a point of comparison for the 1892 concert, in Satina’s data there are references only to a ‘Scherzo’ and an ‘Etude’ by Chopin being performed, both without identifying details.

In the rare instance when Satina was unable to find any details for a concert at all, speculation is best avoided, as for consecutive concerts on 2 and 3 February in Tucson and Phoenix, AZ. At the time, Rachmaninoff was alternating two programmes, one featuring Schumann’s Sonata in G minor, op. 22, the other being the ‘Appassionata’ Sonata programme from earlier in the season. Given that the Schumann programme had been performed in St Louis, MO, on 27 January, the following day in Topeka, KS, and in El Paso, TX, on 30 January, one might assume that it was repeated in Arizona; yet the Beethoven programme would be performed two days later in Los Angeles on 5 February.

Despite Satina’s attempt to construct a complete record of Rachmaninoff’s numerous concerts, it seems inevitable that discrepancies and incomplete data will remain an issue. Nevertheless, her research has added immeasurably to our knowledge of the works he played in concert. Perhaps unintentionally, her data also highlight a very human exigency: a growing need to keep concertizing fresh.

The Need for Programmatic Variation

Rachmaninoff was very aware of the stresses of extended touring. Responding in an interview at the time of his first American tour in 1909, he pondered out loud: ‘Imagine giving an almost daily concert for three whole months’, further noting that it was a ‘strain’. ⁶³ When his tour resumed in January 1925, he was quite conscious of the vast stretch of dates before him (lasting until the beginning of April), jokingly referring to them in a letter as a ‘perpetuum mobile’. ⁶⁴

Accordingly, it is likely that both prior to the tour and while it was in progress Rachmaninoff looked for ways to keep his performances fresh through varying his repertoire. The reasons for programmatic variation can be complex due to the nature of the touring that was undertaken. For example, at times there was little chance to practise between concerts, and altering a programme may have presented issues. ⁶⁵ Yet in other instances Rachmaninoff’s itinerary afforded a few days without travel, and it is possible that he used this time to ‘re-tune’ a programme, or to include a work that he contemplated introducing at later venues. Importantly, however, there is no indication that Rachmaninoff altered a programme on account of a smaller venue, or an arguably less sophisticated audience; his creative ideal, as expressed through programming, was largely constant across the range of venue types, albeit with occasional exceptions. As will be outlined later, the few notable occasions when a programme was altered potentially relate to his expectation of a more sophisticated audience, such as in Boston and New York.

In the first concert on 14 January in Washington, DC, ⁶⁶ the Schumann Sonata in G minor, op. 22, was programmed for the first time (Figure 12). ⁶⁷ Remarkably, this would appear to be a relatively large city at which to introduce a new work. This programme would alternate with the Beethoven Sonata programme up to March, with the former appearing more frequently until the Beethoven Sonata eventually dominated programmes by the end of the season. Schumann’s Sonata was preceded by a new concert-opener, Saint-Saëns’s transcription of themes from Gluck’s ballet Alceste. ⁶⁸ Chopin’s Scherzo no. 4 and Ballade no. 3 remained on the programme, and the Godowsky transcription closed it.

Figure 12. Recital, 14 January 1925.

In Charlottesville, VA, on 17 January, and again in Norfolk, VA, on 19 January, Rachmaninoff played the Schumann Sonata programme, while in his first appearance in Knoxville, TN, and a subsequent New Orleans, LA, concert on 23 January, he returned to the Beethoven Sonata programme, all works having had numerous performances earlier in the season (Figure 13).

Figure 13. Recital, 21 January 1925.

As was noted earlier, the Schumann Sonata programme was repeated in St Louis, Topeka, and El Paso, with the details for concerts in Tucson and Phoenix unknown, while the Beethoven Sonata programme returned for the concert in Los Angeles on 5 February. After returning to the Schumann Sonata programme on 6 February in Pasadena, CA, Rachmaninoff began to alternate how he closed the programme, choosing the music of Liszt in San Diego (Figure 14). This appears to be the first performance Rachmaninoff gave of Liszt’s Polonaise no. 2, a work which would feature in many later programmes, and one of the few mid-length solo piano works for which a recording exists. ⁶⁹ The unnamed Liszt Sonetto del Petrarca was likely no. 104, given that it had appeared frequently in previous seasons. ⁷⁰ In San Jose, CA, two days later, the Polonaise appeared alone (Figure 15).

Figure 14. Recital, 9 February 1925.

Figure 15. Recital, 11 February 1925.

In Stockton, CA, the following day, at a venue Satina noted merely as ‘High School Auditorium’, the programme was varied again with the Godowsky transcription returning as the concert-closer, while Rachmaninoff reverted to Liszt’s Sonetto in both Fresno, CA, on 13 February and Oakland, CA, on 14 February. ⁷¹ As Rachmaninoff travelled up the West Coast, he continued alternating his programmes, playing the Beethoven Sonata programme in San Francisco, CA, on 15 February, while in Portland, OR, he performed the Schumann Sonata programme, although he chose to close with Godowsky. As indicated previously, the potential discrepancies of the programme he offered in Seattle, WA, raise questions, and could represent the most significant sudden change of music in his tour so far. In Tacoma, WA, two days later, Rachmaninoff added a new work between the concert-opener and the Schumann Sonata, Beethoven’s 32 Variations in C minor, WoO. 80 (Figure 16).

Figure 16. Recital, 21 February 1925.

Notably, Tacoma may also be considered a ‘smaller’ type of venue. The work first had been included in a programme in his Scandinavian concerts in 1918, yet had not been played since concerts in America and Canada in 1919. He again switched the works that concluded the concert, finishing with the Polonaise and Sonetto of Liszt, the format being repeated without further variation in concerts across Canada beginning in Victoria, BC (23 February), then on to Vancouver, BC (25 February), Calgary, AB (27 February), Edmonton, AB (2 March), and ending in Winnipeg, MB (4 March). Given that the return journey across the continent involved many hours of travel, hence offering limited time to practise, one imagines there may have been practical reasons for not varying the programme.

After six successive iterations of the programme, Rachmaninoff gave a further performance of it in Chicago, IL, on 8 March, albeit with the apparent addition of another Chopin Etude. ⁷² The following day in St Paul, MN, the Schumann Sonata was abruptly dropped, and Rachmaninoff would not return to it again for many years (Figure 17). ⁷³ While Beethoven’s ‘Appassionata’ became the major sonata on his programme, he similarly abandoned the companion pieces from earlier in the season (the English Suite of Bach, Mendelssohn’s Variations, and Liszt’s Funérailles). Rather, he created a hybrid or ‘combined’ programme, performing the ‘Appassionata’ alongside the works typically found in the Schumann Sonata programme, including the Beethoven Variations, which were added on the West Coast, although at times omitting the opening Caprice by Saint-Saëns.

Figure 17. Recital, 9 March 1925.

The new ‘combined’ programme from St Paul was performed in Des Moines, IA, on 11 March, while in Grand Rapids, MI, on 13 March the Saint-Saëns returned as the concert-opener before the Beethoven Variations. ⁷⁴ At a concert in Kalamazoo, MI, on 16 March the ‘combined’ programme, albeit now without the Saint-Saëns but concluding with Godowsky, marked the final performance before he returned home to the East Coast. With the exception of two concerts in Boston and New York, which are discussed below, a further five concerts rounded out the ‘perpetuum mobile’ tour, and variants of the ‘combined’ programme were given in each. In Toronto, ON, on 23 March he began with the Caprice but ended with Liszt’s Polonaise; in Wilkes-Barre, PA, on 27 March without the Caprice and ending with Godowsky; in Philadelphia, PA, on 28 March with both the Caprice and the Godowsky; in Plainfield, NJ, on 30 March without the Caprice but ending with Godowsky; and in Brooklyn, NY, on 31 March, at last in his home state, including both the Caprice and the Godowsky but dropping Beethoven’s Variations.

Over eleven weeks, Rachmaninoff had given thirty-eight recitals, and it is clear that adroit variation in programming not only occurred throughout but was also likely needed. While it is necessary to set to one side the two programmes for which there are no details, the longest unbroken stretch of an identical programme on this tour appears to have been the six concerts given while making the long return journey from the West Coast across Canada. Elsewhere, the same programme was performed in succession no more than three times, and by the end of the tour no two successive programmes were the same.

Two further performances rounded out the concert season, and they highlight how Satina’s collected data reveal the playing-out of personal friendships and acquaintanceships on the concert stage. When Rachmaninoff returned to Carnegie Hall for an afternoon performance of his Third Piano Concerto on 2 April, and again the following evening, it was with Walter Damrosch conducting the New York Symphony Orchestra (Figure 18). ⁷⁵ Rachmaninoff had premiered the concerto with Damrosch and the orchestra in the same hall on 28 November 1909. In the interim, the two had performed Rachmaninoff’s First Piano Concerto (1922), the Second Piano Concerto (1919, 1921, 1923), Tchaikovsky’s Piano Concerto in B flat minor (1921), and other performances of the Third Piano Concerto in 1919, 1920, and 1922. ⁷⁶

Figure 18. Concerto concert, 2 April 1925.

When Rachmaninoff performed his Second Piano Concerto in Boston, MA, on 17 and 18 April, it rekindled a partnership that had been far less frequent over recent years (Figure 19). ⁷⁷ Serge Koussevitzky had been a regular partner with Rachmaninoff in concerts in Russia, although their first performance together had been at the conductor’s famous premiere with the Berlin Philharmonic in Germany on 23 January 1908, where the Second Piano Concerto had also been performed. ⁷⁸ Their last appearances together had been in Moscow on 13 and 20 March 1917 (Julian calendar), playing first the Tchaikovsky B flat minor Concerto (the ‘February’ Revolution was underway at the time), and the Liszt E flat Concerto on the later date. ⁷⁹

Figure 19. Concerto concert, 17 April 1925.

Rachmaninoff’s third recording session of the season came at the conclusion of the long tour, as he returned to again attempt recordings of works from earlier seasons, as well as those from recent concerts (Figure 20). Yet of the smaller works attempted on 13 April, only the Hopak was approved for release. ⁸⁰ Of the larger works, the Liszt Polonaise was recorded successfully, yet the first section of the Beethoven proved unsatisfactory. While the recording of Chopin’s Ballade was marked to be held over, it was eventually released. When Rachmaninoff returned to the studio the following day, he managed a successful take of ‘Das Wandern’, yet could do no better when trying to improve on his previous recording of the Liszt Polonaise (Figure 21).

Figure 20. Recording session, 13 April 1925.

Figure 21. Recording session, 14 April 1925.

After a total break of almost a month, Rachmaninoff returned again to Camden where he at last completed his recordings (Figure 22). The Rondo alla Turca was accomplished to his satisfaction, along with the opening section of the Beethoven Variations. After introducing it first as an encore in November the previous year, there was at last a gramophone recording of Sgambatti’s arrangement of Gluck’s Mélodie, preserving what the present writer believes is one of Rachmaninoff’s finest. ⁸¹

Figure 22. Recording session 14 May 1925.

Given the degree to which I have been able to document the works that Rachmaninoff was playing in recital, the interaction between the frequency of concert performance of a work and its eventual recording can now be ascertained more precisely. In the case of works such as the Polonaise by Liszt, one senses that the recording captures exactly what audiences heard in the concert hall. Yet, it also appears that the recording of a piece was not necessarily related to its recent frequency of performance, with Rachmaninoff returning to record successfully certain works from much earlier seasons. In regard to the Sgambatti transcription, his best effort appears to have come after a complete break from performance.

The Programmes of Rachmaninoff’s Contemporaries

It is appropriate at this juncture to assess the degree to which Rachmaninoff’s programmes were structured differently to those of his contemporaries. Rachmaninoff had predominately played only his own works in recital prior to leaving Russia, which was not entirely atypical for a Russian composer-pianist. Alexander Scriabin, a classmate from the Moscow Conservatory, appears to have played only his own works after his student years. For example, in the anthology of close to 15,000 piano recital programmes compiled by George Kehler, all ten of Scriabin’s concerts listed by the author between 1907 and 1915 comprised his own works. ⁸² Faubion Bowers, in his two-volume biography, lists numerous programmes from venues in Russia and the West, these also attesting to Scriabin’s singular focus on the concert platform. ⁸³ Sergei Prokofiev, when planning to travel to America in 1918, thought it sufficiently important to note in his diary his ‘first time in public not playing my own compositions’. ⁸⁴ While his later concerts in the West frequently included music by other composers, a collection of programmes in the Prokofiev Archive of Columbia University highlights that the majority of the music performed remained his own. ⁸⁵

It is difficult to know with certainty why Rachmaninoff had drastically altered his standard way of programming recitals soon after leaving Russia. ⁸⁶ In recitals in 1918 in Scandinavia over March, April, and May, he exclusively had played his own works; yet from recitals commencing in September, and then as the norm once he arrived in the United States, his programmes predominantly featured the music of other composers. In understanding his thinking, the ‘Historical Concerts’ of Anton Rubinstein potentially offer context: the series created a significant impression on Rachmaninoff’s formative years, and likely influenced repertoire choices later on. ⁸⁷ Inherent in their structure, the programmes of Rubinstein’s concert series highlighted that a pianist in the late nineteenth century could present music from an array of composers and periods, even when the performer was also a composer. ⁸⁸

To assess Rachmaninoff’s 1924/25 season more broadly, a trove of digital programmes in the archive of Carnegie Hall has been used to create comparative data. ⁸⁹ The programmes indicate a total of sixteen solo piano recitals were given at the venue, two of which featured Rachmaninoff. ⁹⁰ In a clear majority of cases, it has been shown that Rachmaninoff presented very similar programmes in both regional towns and larger cities although, as discussed later, in the case of Boston and New York he occasionally offered a few of his more challenging works, as he did this season. Even so, the Carnegie Hall archive serves generally as a suitable basis for comparison. As a note of caution, the provision of comparative data on concert programmes can be complex, given that true statistical equivalences require an assessment of the scope of each piece to determine a precise weighting, in particular its duration. Yet simple approaches still highlight trends, and in the following data I have followed a methodology used by John Gould in his research into piano recital programmes, where the process records only the instances in which a composer is named. ⁹¹

Gould expressed his data in tables featuring eleven ‘named’ composers who dominated programmes, comprising J. S. Bach (both original works and transcriptions), Mozart, Beethoven, Schubert, Mendelssohn, Chopin, Schumann, Liszt, Brahms, Debussy, and Bartók, with a further column for ‘other’ composers (Table 1). Across sixty different recital programmes in the period 1921–50, Gould calculated that the music of Chopin, Beethoven and Liszt dominated, respectively accounting for 12.2%, 10.6%, and 8.9% of programming, with the music of Schumann trailing closely at 7.6%. Overall, the music of the ‘named’ composers featured 67% of the time, while the music of ‘other’ composers accounted for the remaining 33%. ⁹²

Table 1. Gould’s data for 1921–1950.

	Bach (original)	Bach (trans.)	Mozart	Beethoven	Schubert	Mendelssohn	Chopin	Schumann	Liszt	Brahms	Debussy	Bartok	Total ’named’	’Other’
Gould’s data 1921–1950	15 (5%)	11 (3.6%)	11 (3.6%)	32 (10.6%)	13 (4.3%)	6 (2%)	37 (12.2%)	23 (7.6%)	27 (8.9%)	11 (3.6%)	16 (5.3%)	1 (0.3%)	203 (67%)	100 (33%)

In my analysis of the sixteen recitals in the 1924/25 Carnegie Hall season, programmes were similarly dominated by these same composers in the majority of cases (Table 2). The music of Chopin was played by most pianists, while the music of Liszt, Beethoven, and Schumann featured prominently. The average of instances of the ‘named’ composers accounted for a slightly higher figure of 76.2% in the Carnegie Hall season. A clear exception was a performance by Percy Grainger in which ‘named’ composers accounted for only 27.2% of the music, thus highlighting the more diverse nature of his programming. ⁹³ At Rachmaninoff’s second recital on 21 March in which he played a broad set of his more challenging works, a comparatively high percentage of ‘other’ composers featured (including his own music); yet, in reference to the issue noted earlier concerning comparative data, this slightly masks the relative shortness of his works.

Table 2. Comparison of Carnegie Hall programmes, 1924–1925.

	Bach (original)	Bach (trans.)	Mozart	Beethoven	Schubert	Mendelssohn	Chopin	Schumann	Liszt	Brahms	Debussy	Bartok	Total ’named’	’Other’
Brailowsky 10 January 1925							16 (100%)						16 (100%)
Brailowsky 14 February 1925			1 (7.1%)		1 (7.1%)		8 (57.1%)	1 (7.1%)	1 (7.1%)				12 (85.7%)	2 (14.2%)
Grainger 5 January 1925		1 (9.1%)								2 (18.2%)			3 (27.2%)	8 (72.7%)
Hofmann 22 November 1924		1 (9.1%)			1 (9.1%)		4 (36.7%)	1 (9.1%)					7 (63.6%)	4 (36.7%)
Hofmann 28 March 1925				1 (10%)		1 (10%)	4 (40%)		3 (30%)				9 (90%)	1 (10%)
Leginska (Munz) 26 January 1925							6 (75%)		2 (25%)				8 (100%)
Levitsky 13 January 1925				2 (15.3%)		1 (7.7%)		1 (7.7%)	1 (7.7%)		2 (15.3%)		7 (53.8%)	6 (46.1%)
Levitsky 14 March 1925				1 (7.7%)			6 (46.1%)	1 (7.7%)					8 (61.5%)	5 (38.4%)
Lhevinne 2 November 1924			1 (10%)				3 (30%)	1 (10%)			1 (10%)		6 (60%)	4 (40%)
Lhevinne 12 January 1925				1 (8.3%)	2 (16.7%)		4 (33.3%)		2 (16.7%)			1 (8.3%)	10 (83.3%)	2 (16.7%)
Pachmann 17 October 1924	1 (8.3%)		1 (8.3%)				7 (58.3%)	1 (8.3%)	1 (8.3%)	1 (8.3%)			12 (100%)
Pachmann 13 April 1925							14 (100%)						14 (100%)
Rosenthal 4 January 1925	1 (11.1%)						4 (44.4%)	1 (11.1%)		1 (11.1%)	1 (11.1%)		8 (88.9%)	1 (11.1%)
Rosenthal 17 January 1925				1 (8.3%)			4 (33.3%)	3 (25%)	1 (8.3%)	1 (8.3%)			10 (83.3%)	2 (16.7%)
Rachmaninoff 30 November 1924	1 (8.3%)	1 (8.3%)					4 (36.3%)		1 (8.3%)				7 (63.6%)	4 (36.3%)
Rachmaninoff 21 March 1925				2 (22.2%)					2 (22.2%)				4 (44.4%)	5 (55.5%)

My analysis of all recitals given by Rachmaninoff in his 1924/25 season indicates that the music of Chopin also dominated to a very high extent, accounting for 30.6% across all programmes (Table 3). The music of Liszt accounted for 10.7% of programming, while the music of Beethoven stood at 6.8%. Potentially more surprising is the overall statistic for ‘other’ composers, which accounts for 42% of programming. This figure can be separated into Rachmaninoff’s own music (26.3%) and all ‘other’ composers (15.7%), the latter being less than the Carnegie Hall season average (32.8%). Again, the method of data collection perhaps masks that many of Rachmaninoff’s works were relatively short (as, indeed, were some of Chopin’s).

Table 3. Rachmaninoff’s programmes, 1924–1925.

	Bach (original)	Bach (trans.)	Mozart	Beethoven	Schubert	Mendelssohn	Chopin	Schumann	Liszt	Brahms	Debussy	Bartok	Total ’named’	’Other’	Rachmaninoff	Actual ‘other’
Rachmaninoff 1924/25	15 (2.8%)	9 (1.7%)		36 (6.8%)		8 (1.5%)	162 (30.6%)	20 (3.7%)	57 (10.7%)				307 (58%)	222 (42%)	139 (26.3%)	83 (15.7%)

While the methodology for collecting data is limited, various trends nevertheless emerge, the strongest pointing to the typicality of Rachmaninoff’s programming. In another sense, even taking into account his lower focus on his own compositions in comparison with Scriabin and Prokofiev, it would seem problematic to argue from the analysis that Rachmaninoff’s programming represented a specific lack of confidence in his music.

While Paderewski did not play at Carnegie Hall in the 1924/25 season, he remained the leading pianist on the North American stage at the time. ⁹⁴ A comparison of Paderewski’s programmes from earlier Carnegie Hall seasons offers insights regarding the structure of recitals and the diversity of composers featured, revealing that many share elements in common with both Rachmaninoff’s programmes and those of other pianists. For example, Paderewski’s 18 March 1893 programme opened with J. S. Bach, featured a Beethoven sonata, was followed by a collection of works by Chopin (plus a single one of his own), and finished with a Hungarian Rhapsody by Liszt. ⁹⁵ On 8 March 1902, Paderewski began with Beethoven, followed with Haydn and a Schumann sonata, before a collection of Chopin pieces and a concluding work by Liszt. ⁹⁶ Towards the beginning of the 1925/26 season, on 25 November, he began with two works by Beethoven, followed with Schumann and Chopin, again finishing with Liszt. ⁹⁷ It appears that Rachmaninoff, along with many of his peers, was willing to steer programmatic pathways proven as commercially viable over decades. ⁹⁸

Assessing His Worth as a Composer

When Rachmaninoff performed his own smaller works, determining their precise details has often been difficult, as with earlier references to the unnamed Etude-Tableau. ⁹⁹ While one might be led to assume this to be a quirk of Satina’s recording of data, many of the printed programmes she amassed similarly lacked these details (Figure 23). ¹⁰⁰ Indeed, it is possible that the absence of detail in materials sent to concert promoters might have been intentional, allowing Rachmaninoff to substitute his own compositions at will. Similarly, with smaller works by Chopin a degree of intentional vagueness perhaps allowed him the same liberty.

Figure 23. Recital programme, 9 February 1925.

Yet at other times, the inclusion of precise details regarding his own works – such as for the named Etude-Tableau in D major, op. 39, in programmes beginning in Terra Haute – is notable. As indicated previously, shortly after emigrating, Rachmaninoff abruptly reduced the appearance of his works in programmes. Further, when he did then programme his own works – almost always in a small set – he preferred playing his early and less challenging compositions for audiences. Hence, in the case of the Etude-Tableau in D major, op. 39, it is significant that a piece from his final years in Russia was included and named in programmes, a work which, as was noted previously, is marked in terms of stylistic progressiveness.

The Etude-Tableau had been named in concerts in Boston and New York in November 1924, and when Rachmaninoff returned for concerts at these leading venues on 19 and 21 March 1925, the programmes are significant due to the sudden and dramatic variation of the earlier ‘combined’ programme, with the inclusion of the B minor Prelude from op. 32 and a selection from his Etudes-Tableaux, op. 33 and op. 39 (Figure 24). ¹⁰¹ Rachmaninoff had last played an extended set of his Etudes-Tableaux in concerts in 1920, although the precise details were not then given. ¹⁰²

Figure 24. Recital, 19 March 1925.

The abrupt inclusion of new works in programmes for major venues in Boston and New York stands in contrast to what other writers have regarded as Rachmaninoff’s typical pattern of introducing new pieces to programmes in smaller venues, as has been outlined. That they were his own works, and more compositionally advanced, invites conjecture, given that Rachmaninoff was soon to pause his concert activity to return to composition: following the short season that resumed on 29 October 1925 (only twenty-two concerts), his ‘sabbatical’ year would begin. ¹⁰³ In my view, Rachmaninoff either used these opportunities to test audience reaction to his music, or to gauge the response of critics as a form of assessing, again, his worth as a composer.

It is notable that in a concert in the 1920/21 season Rachmaninoff had included without any immediate prior performances both his Second Sonata and Scriabin’s Fifth Sonata at concerts in New York and Boston. ¹⁰⁴ One might speculate that the venues were sufficiently ‘high-profile’ to merit inclusion of his own sonata. Yet one might also postulate that Rachmaninoff was deliberately presenting for comparison his music alongside that of Scriabin, his feted compatriot, to these discerning audiences.

If questions about Rachmaninoff’s assessment of his own worth as a composer are evident in his programming choices, it is remarkable that debates about his reception history remain prominent today. For example, the issue has been highlighted in Glen Carruthers’ reference to the music’s popularity remaining ‘a real liability’, ¹⁰⁵ while Charles Fisk noted that its popularity and ‘old-fashioned’ qualities stood as an exemplar of unsophisticated taste. ¹⁰⁶ More recently, Emily Frey noted that in twentieth-century textbooks Rachmaninoff stood as a poster boy for epigonism, one ‘who remained stubbornly […] resistant to the demands of musical “progress”’. ¹⁰⁷ By comparison, Rachmaninoff’s place among competing themes of twentieth-century music-writing, along with a contemporary assessment of the polemics involved, are portrayed vividly in Richard Taruskin’s aptly titled ‘Not Modern and Loving It’. ¹⁰⁸ Moreover, in her recent book, Rebecca Mitchell has argued for reassessing Rachmaninoff in light of his ability to adapt to his changing age, viewing him not necessarily as ‘modernist’ but fundamentally as ‘modern’. ¹⁰⁹

Two works were completed in the sabbatical year of 1926, the more significant being the Fourth Piano Concerto, op. 40. Yet reviews of the concerto were mainly negative, at times caustically so, and the composer ultimately created a further two versions of the work, neither of which has enjoyed the success he likely hoped for. ¹¹⁰ Yet the new concerto represented a further advancement of his compositional style, a nod towards the ‘modernism’ of the age and, hence, a potential attempt at validation of his relevance as a composer as highlighted by the inclusion of his more advanced works in this season. Over the remaining decade and a half of his life, Rachmaninoff returned to composition infrequently, with two of his later works finding immediate success (the Rhapsody on a Theme of Paganini, op. 43, and the Symphonic Dances, op. 45), while the Third Symphony, op. 44, was viewed unfavourably. ¹¹¹ All the while, his career as a performer remained steadfastly strong, ¹¹² with concert activity continuing at an almost unabated rate until shortly before his death. ¹¹³

* * *

In all, Rachmaninoff gave sixty-nine concerts in the 1924/25 season (Figure 25). In addition, he undertook seven recording sessions, one of which was for piano roll (Figure 26). ¹¹⁴ While this was not his busiest season, maintaining such a schedule remains a feat to this day. ¹¹⁵ The collected data for this season have offered an opportunity to reappraise Rachmaninoff’s work as a performer. Insights have been gained into how he structured his programmes, showing that certain traits – such as programming a formally more-complex sonata in the opening half of a concert – were common at this time. Further, on an especially long tour, it is evident that he increasingly needed to vary his programme, even if in only small ways.

Figure 25. List of concerts in the 1924/25 season.

Figure 26. List of recording sessions in the 1924/25 season.

However, there are further ways that the data may be utilized so as to offer greater insights. A question has long lingered as to whether Rachmaninoff performed more exoteric programmes in smaller and regional venues, yet in a general sense this has been disproved by data showing similar programming across vast tracts of the United States and Canada. Even so, a process of determining the relative historicism of programmes through collating and averaging the dates of compositions may highlight whether in certain cities over time he played more modern works, perhaps especially in the case of his own compositionally advanced pieces in New York and Boston. ¹¹⁶

Similar methods may be employed to compare the diversity of musical forms in programmes from across his career, identifying if the second halves of programmes maintained the degree of comparative structural simplicity that is evident in the season analysed here. That Rachmaninoff chose to end concerts with works of outstanding difficulty at this point in his career, such as by Liszt and Godowsky, is unsurprising, yet the data can further be analysed to determine if this remained the case in his later years.

In evaluating Satina’s materials for the 1924/25 season, we encounter a unique view of Rachmaninoff, one which is further amplified when assessing the data across his entire career. What they allow us to see goes beyond a scholarly summary: in reading through consecutive webpages of concert programmes, there is almost a sense of living Rachmaninoff’s life. From the date and location (often a different city on a subsequent night), and in contemplating the works he was performing on stage, the view is revelatory. This, perhaps, was Satina’s intention all along, and her tireless scholarship seems immensely worthwhile. Her research opens a new window through which Rachmaninoff’s career as a concert artist can, again, be assessed.