Often hailed as the most-recorded violin concerto of the twentieth century, the origins of Sibelius’s Violin Concerto in D minor, Op. 47, which was first performed in 1904, reach back to the last decade of the nineteenth century. In a pattern that would prove typical across his career, Sibelius quickly became dissatisfied with the concerto as it was initially premiered and withdrew it for substantial revision, which resulted in the final version of 1905 that became an enduring staple of the violin repertory. In 1991, a ground-breaking recording of the withdrawn early version performed by Leonidas Kavakos with the Lahti Symphony Orchestra under the direction of Osmo Vänskä (BIS CD 500) piqued the ears of listeners who thought they knew this concerto. Now, the publication of Series II, vol. 1 in the Complete Works (JSW) series makes the score of the early version publicly available for the first time, together with a newly prepared critical edition of the final version.
The idea of writing a concerto for violin – Sibelius’s own instrument – appears as early as 1890 in a letter he wrote to his wife, Aino. The earliest known compositional sketches likely date from quite a bit later, around 1899–1900, and the initial version for violin and piano was completed in summer 1903. Sibelius finished orchestrating the concerto in late 1903 or very early 1904, and the premiere followed quickly thereafter in Helsinki in February 1904 – not by Willy Burmeister, as Sibelius had originally intended, but rather by Viktor Nováček. Additional performances of this first version occurred in Turku in late March, but critical reception of the piece was mixed at best, and by June 1904 Sibelius decided to withdraw the concerto altogether and undertake serious revisions.
Since the appearance of the early version of this concerto in recorded form, several scholars have investigated its genesis and compositional techniques. Erkki Salmenhaara places the concerto in the context of Sibelius’s biography,Footnote 1 while Jukka Tiilikainen has reconstructed layers of sketches.Footnote 2 More recently, Tomi Mäkelä took up the concerto’s reception history,Footnote 3 and Les Black has investigated progressions of key areas.Footnote 4 Until now, however, limited access to the score of the early version has hindered follow-up investigations.
Building upon existing as well as significant original research, Tomi Virtanen’s extensive introduction, written in English and German according to editorial principles for JSW, guides the reader through the stages of the concerto’s genesis, from sketches to first performances, revisions, performances of the final revised version, and finally the publication of the final version. Virtanen draws richly upon primary source material, allowing Sibelius and his contemporaries to speak through letters and giving voice to critics through direct quotations from their writings. A discussion of the 1929 edition of the solo violin part with bowings and fingerings by Ossip Schnirlin – with which, it appears, Sibelius was not entirely pleased – will be of special interest to performers and scholars of performance practice.
Many works by Sibelius have long paper trails because of the composer’s propensity to revise and even re-revise his compositions both before and after their premieres, and the violin concerto is no exception. The preparation of this edition involved consulting nearly two dozen musical sources from the National Library of Finland, the Sibelius Museum, Ainola, the German publisher Lienau, and under private ownership. Five extant sources – the autograph orchestral score, fragmentary copies of the solo violin part and the piano score, and orchestral parts for the first performance – provided material for the early version. Editing the final version required the consultation of a further 11 primary sources (autographs, scribal copies, proofs, and first editions), several secondary sources with markings Sibelius made later on in the 1930s or 1940s, and more distantly related scores and conversation notes between the composer and his son-in-law, the conductor Jussi Jalas. The critical commentary (in English only) opens with an explanation of general editorial principles, including decisions to regularize inconsistencies in the notation of certain rhythms and the precise placement of dynamic markings. Detailed critical remarks provide further insight into discrepancies between the autograph orchestral score, parts, and the first surviving copy of the piano score in nearly every bar of the early version of the concerto, including several reductions in orchestration. Compare, for example, the absence of winds and horns in bars 230–245 of the third movement with the original version of the same passage printed on p. 275 in the commentary, which is crossed out with pencil in the autograph score and correspondingly omitted from the parts used for the 1904 premiere performance. In addition, a section of critical commentary lists successive alterations made in various primary and secondary source materials for the final version, including markings made by Richard Strauss for the 1905 premiere of the final version (which consist mostly of moderate changes in dynamics). Helpful tables also list precise locations of sketches directly related to each version of the concerto, including HUL codes – catalogue numbers for manuscripts in the Sibelius collection at Helsinki University Library, now known as the National Library of Finland – and page numbers, so that interested scholars can easily locate relevant source material.
In general, Sibelius’s revisions condense, rather than add, material: the first movement is reduced by about eight per cent (from 540 bars to 499), while the third movement is condensed by 18 per cent (from 326 bars to 268). In the first movement, one locus of change is cadenzas. Each version contains two unaccompanied cadenzas for solo violin, but only one of these is common to both versions. In the final version, a new unaccompanied cadenza is created early in the work by silencing a rhythmicized string accompaniment during a passage from the early version (bars 55–70 in both versions). The second solo cadenza in the final version (bars 234–271) also appears in the early version (bars 217–254), but a longer, 59-bar cadenza near the end of the early version (bars 399–457) is cut entirely from the final version.
Of particular interest in this parallel edition is the second movement; in the version for violin and piano, Sibelius pasted the new violin part over the existing one, leaving the piano part intact, but the careful reader will note that numerous subtle changes creep into dynamics, articulation, and, occasionally, pitches in the final version. Comparing the first half of these two versions is facilitated by similar layouts in which the same bars are set on each page in the respective versions, until the rapid trills in the accompanying voice in bars 33–34 of the solo violin part in the early version change to crotchet triplets in the final version, reducing the number of note-heads per bar in that voice from 48 to six and thereby completely changing the width of the bars.
A generous selection of almost 30 facsimile images of excerpts from sketches, autographs and published score materials both underscores the editorial scope of this project and provides visual confirmation for selected entries in the critical commentary, which make occasional reference to facsimiles by number. In future projects, a caption on the page of each facsimile with an explanation of its relevance, or at the least a reference to corresponding page or pages in the critical commentary, would help the reader understand the reasons for inclusion of each image.
Virtanen is preparing a companion volume (JSW II/1A) of additional stages of this concerto, namely an incomplete version for violin and piano from 1903–04 and a complete version for the same instrumentation from 1905. Together, these volumes will offer unparalleled access to the gradual unfolding of a concerto that, as its original publisher Robert foresaw in 1907, would ‘only conquer the world slowly’ (II/1, XII).
