Faculty of Performance, Visual Arts and Communications

School of Music


Paganini: 24 Capricci per il Violino solo, dedicati agli Artisti Op. 1 - Robin Stowell

Robin Stowell

Paganini’s twenty-four Capricci per violino solo Op. 1 set new standards of technical brilliance for the instrument, establishing a platform in the development of his legendary diabolical stage persona for his concert tours of 1828–34, when he took continental Europe by storm. They have long been considered among the supreme challenges for the advanced violinist, a culminating technical benchmark to which to aspire. The Norwegian virtuoso Ole Bull remarked that they had no equal ‘either in beauty, originality or difficulty of performance’; and Louis Spohr even alleged that they were ‘unplayable’ and ‘against the nature of the instrument’ [Alpheus B. Crosby, The Art of Holding the Violin and Bow as exemplified by Ole Bull (London: William Reeves, 1909), pp. 37-8; François-Joseph Fétis, Notice biographique sur Nicolo Paganini, Eng. trans. Wellington Guernsey (London: Schott, 1852), p. 79]. Nevertheless, they inspired many other Romantic composers, not least Robert Schumann who heralded Paganini as ‘the turning point in the history of virtuosity’, raising the significance of virtuosity to what Liszt described as ‘an indispensable element of musical composition’ [Boris Schwarz, Great Masters of the Violin (London: Robert Hale Ltd., 1984), p. 181; Owen Jander, ‘Virtuoso’, in Stanley Sadie (ed.), The New Grove Dictionary of Music and Musicians, 29 vols. (London: Macmillan, 2001), vol. 26, p. 789]. Schumann evidently saw in these caprices ‘so many gems’ (‘so viel Demanthaltiges’) that he elaborated some of them for the piano [Robert Schumann, Gesammelte Schriften über Musik und Musiker (Berlin: Volksverband der Biicherfreunde, Wegweiser Verlag, 1922), p. 164].


There is some mystery and confusion over when these caprices were actually composed. Paganini’s correspondence evidently includes no mention of the caprices until 1836 [See Philippe Borer, The Twenty-Four Caprices of Niccolò Paganini (Zurich: Stiftung Zentralstelle der Studentenschaft der Universität Zürich, 1997), p. 8]. The Polish violinist Karol Lipiński, who met Paganini in Piacenza in 1818, evidently informed Schumann that the caprices were originally composed as gifts for friends. However, when Giovanni Ricordi requested them for publication, Paganini is believed to have reconstructed them from memory ‘in a great hurry and frenzy’, reinforcing views that these works are essentially notated improvisations, articulated in lucid formal structures [Schumann, Gesammelte Schriften, p. 164].

As there is no evidence of any surviving copy of the caprices prior to the autograph fair copy submitted to Ricordi and received by Signor Tomaso on 24 November 1817, many musicologists consider the above to be an unlikely course of events [The inscription appears on the bottom left-hand corner of the title-page and reads: ‘Sr Tomaso, li 24 9bre 1817’]. Further, Edward Neill believes that this manuscript suggests an integrated collection linked ‘by a remarkable thread of continuity’ rather than a collection of pieces written at various places and times [Paganini, Capricci Op. 1, ed. Edward Neill and Salvatore Accardo (Milan: Ricordi, 1988), pp. III and VII]. Some commentators propose that these works were probably composed over a span of time before being submitted to Ricordi [for example, Claudio Casini, Paganini (Milan: Electa, 1982), p. 62]; others are teased by Paganini’s own declaration to Lichtenthal that, upon his return to Genoa in 1796, ‘he composed difficult music and worked continuously at difficult problems of his own invention’ [In Allgemeine musikalische Zeitung, no. 20 (May, 1830), p. 325]. Israil Yampolski suggests 1798-1800 as Paganini’s initial period of inspiration, ascribing it to Paganini’s discovery in Marquis Di Negro’s library of Locatelli’s 12 concertos Op. 3 (L’arte del violino) and their 24 capricci – essentially cadenzas appropriate for the first and last movements of each concerto [Israil M. Yampolski, in Konstantin Mostras (ed.), 24 Kaprisa dla skripki solo N. Paganini (Moscow: Gosudarstvennoe Muzikal’noe Izdatel’stvo, 1959), Preface, p. 5]. The hypothesis that Paganini may have composed the caprices during his residency at the Napoleonic court of Lucca (1801–09) is also popular [see Jeffrey Perry, ‘Paganini’s quest: the twenty-four Capricci per violino solo, Op. 1’, 19th-Century Music, 27/3 (2003—4), p. 208], as is also one that he worked on a final version of the set sometime between 1812 and 1817.

The inscription ‘dedicati agli artisti’ (‘dedicated to all artists’) on Paganini’s manuscript implies that he intended the caprices as serious essays in composition, combining both musical substance and demanding technical content. A copy of the first Ricordi edition published in Leipzig (Leipzig, Breitkopf und Härtel, [1823?]; plate: 3936), once owned by the antiquarian bookseller Albi Rosenthal, suggests that Lipiński’s statement that the caprices were originally written for friends or close acquaintances may be true [See Albi Rosenthal, ‘An intriguing copy of Paganini’s Capricci’, in Nicolò Paganini e il suo tempo (Genoa: Comune di Genova, [1984]), pp. 235-46]. The copy is inscribed ‘Proprietà di Niccolò Paganini’ (‘property of Niccolo Paganini’) and includes the name of a musician in brown ink in a hand contemporary with the composer at the head of each caprice, as if in dedication. The inscription ‘A Parigi l’inverno’ (‘In Paris, in winter’) suggests that Paganini may have annotated this copy some time between 1832 and 1840. The ‘dedicatees’, most of them violinists, are as follows:

Caprice 1. Henri Vieuxtemps; 2. Giuseppe Austri; 3. Ernesto Camillo Sivori; 4. Ole Bornemann Bull; 5. Heinrich Wilhelm Ernst; 6. Karol Józef Lipiński; 7. Franz Liszt; 8. Delphin Alard; 9. Herrmann; 10. Théodor Haumann; 11. Sigismond Thalberg; 12. Dhuler; 13. Charles Philippe Lafont; 14. Jacques Pierre Rode; 15. Louis Spohr; 16. Rodolphe Kreutzer; 17. Alexandre Artôt; 18. Antoine Bohrer; 19. Andreas Jakob Romberg; 20. Carlo Gignami; 21. Antonio Bazzini; 22. Luigi Alliani; 23. [no name]; 24. ‘Nicolò Paganini, sepolto pur troppo’

Assuming that these annotations are actually in Paganini’s hand, one may speculate that he was preparing a new edition of his Op. 1, possibly for one of his Parisian publisher acquaintances Pacini or Troupenas.

The jury is out regarding any question of a particular caprice reflecting its dedicatee’s playing style, especially with respect to national schools of playing. However, it seems reasonable to align no. 16 with what we know about Kreutzer’s powerful tone and broad bowing style, as well as the characteristics of some of his etudes, while Ole Bull’s experience with Norwegian hardanger violin playing makes him an appropriate dedicatee for he richly polyphonic no. 4. No. 1 may certainly reflect Vieuxtemps’ ‘complete mastery of […] elegant “bouncing” strokes.’ [In Lev Ginsburg, Vieuxtemps: His Life and Times (Neptune, NJ: Paganiniana Publications Inc., 1984), p. 62]. However, no. 15 only relates in part to Spohr’s noble, broad style and the bold double and multiple stopping of no. 14, including some fingered unisons, does not seem to align well with Rode’s style of performance, if reviews and anecdotes are to be believed.

The Autograph

The autograph, comprising 22 folios in oblong format and currently housed in the Archivo Storico di Casa Ricordi, subdivides the twenty-four caprices into three groups, each with its own title page [A facsimile edition, with an introduction by Yehudi Menuhin and a preface by Federico Mompellio, was published by Ricordi (Milan: 1974)]. The first two groups each comprise six caprices and are headed respectively ‘Opera Ima’ and ‘Opera 2da’; the third group, ‘Opera 3za’, consists of caprices nos. 13-24 inclusive. The fair copy manuscript, which is by no means free of error, served as the engraver’s copy for the first edition (Milan: Ricordi, 1820; plate: 403), the publication of which was announced in the Gazzetta di Genova on 23rd June, 1820. This publication presented all 24 caprices together, numbered consecutively, in a single volume. The manuscript contains a number of errors and irregularities, ranging from problems with incorrect pitches (e.g. no. 7 b. 17, no. 11 b. 97), accidentals (e.g. no. 12 b. 52), missing notes (e.g. no. 10 bb. 3 and 39; no. 14 b. 36), rests (e.g. no. 2 b. 15; no. 18 b. 8), rhythms/note durations (no. 21 bb. 18 and 26), octave displacement (e.g. no. 3 b. 14, no. 6, b. 15) and string/timbral indications (no. 4, bb. 9-11; no. 9, bb. 10-12) to those involving alternative performance possibilities (e.g. no. 4 bb. 75-6), altered tempo indications (e.g. in no. 6 ‘Lento’ is changed to ‘Adagio’) and inaccuracies emanating from passages or phrases written in abbreviated form (e.g. no. 4 bb.22-3, no. 8 bb. 8-11), or with differences of slurring between analogous phrases (e.g. no. 2 b. 37) [See Borer, The Twenty-four Caprices, Appendix C ‘The Diplomatic Manuscript’, pp. 233-45; see also Neill/Accardo edition, Preface, p. IX and ‘Critical apparatus’, pp. 53ff.]; thus, it was not an impeccable source for the 1820 edition, which maintained Paganini’s ambiguous symbols and inconsistent notation and itself unsurprisingly introduced further inaccuracies and misunderstandings. Especially noteworthy in the autograph, though, is the sporadic sprinkling of fingering annotations provided by Paganini, notably in no. 2, the quasi-cadenza introduction to no. 5 and bars 1 and 23-5 of no. 6; most are associated with timbral characteristics, especially the choice of string.

Nineteenth-century Editions: an overview

The impact of Paganini’s caprices is reflected in the myriad editions of the opus that have been published since its first issue in 1820; one catalogue lists nearly forty separate editions [Harry Edlund, Music for Solo Violin Unaccompanied (High Wycombe: Peter Marcan Publications, 1989)] Subsequent editions issued in Milan by Ricordi in c.1836 (plate: 9036) and 1872, the latter with performance annotations by Giorgetti-pupil Guido Papini (1847-1912; plate: E.R. 43043), were based on the 1820 print, as is evident from the many errors common to them. So, too, was a pirated edition issued by the publisher Giuseppe Lorenzi in Florence in 1830. As the firm of Ricordi was also responsible for distributing the work abroad, the 1820 print also served as the basis for the c.1823 edition issued under the name of the Leipzig firm Breitkopf und Härtel, the c.1825 and 1829 editions published respectively by Richault (with annotations by Bonaventure Henry; plate: 1028) and by Pacini in Paris (plate: 950), and the 1839 edition (with annotations by the Anglo-Italian violinist Nicolas Mori (1796-1839)) issued in London by Wessel & Co. Some errors in the first edition were rectified during this sequence of publications, but others were also introduced, including wrong notes, missing accidentals and/or tempo indications and incorrect bowings and articulation indications, such that a firmly entrenched but error-ridden performance tradition developed, becoming more and more distant from the autograph and first printed edition. The critical commentary of the Urtext edition of Paganini’s Op. 1 by Alberto Cantù and Ernst Herttrich (with Renato de Barbieri) lists the most significant discrepancies between the autograph, the first three Ricordi editions (1820, 1836 and 1872) and those issued under the names of Breitkopf und Härtel (Leipzig: 1823) and Richault (c.1825) [(Munich: Henle, c.1990), pp. 60-69 (pp. 63-6 in English)].

Discussion of qualities of the myriad editions issued in the almost two-hundred-year history of these caprices must inevitably be selective, not least because only a proportion of these editions have proved accessible. However, one major milestone in the succession of annotated editions of Paganini’s caprices was the edition produced by Ferdinand David (1810-73). Technical developments such as those sparked by Paganini, amongst others, clearly influenced David; a pupil of Louis Spohr, David did not let his mentor’s somewhat conservative views limit his own progress as a violinist, prompting Ferdinand Hiller’s description of David’s playing as combining ‘the sterling qualities of Spohr’s style with the greater facility and piquancy of a later school’ [Ferdinand Hiller, ‘David, Ferdinand,’ in George Grove (ed.), A Dictionary of Music and Musicians, 4 vols. (London: Macmillan, 1879-89), vol. 1, p. 433]. Further, his various associations with celebrated European violinists in Leipzig, particularly those of the Franco-Belgian ‘school’, allowed him to mould his own style of playing.

David heard Paganini in concert in Berlin in March and April 1829 and was so bowled over by these performances that he even considered renouncing violin playing altogether. His appointment as violin professor at the newly established Leipzig Conservatorium (1843) coincided approximately with the start of his editing duties – an edition of Bach’s Sonatas and Partitas for unaccompanied violin specifically for use at the Conservatorium (1843); and his pedagogical role clearly extended to copious annotation of the copies of his pupils. However, it was not until the early 1850s that David’s prolific and highly influential regime of editing others’ works for publication started with any regularity [For more detailed background to Ferdinand David’s development as an editor, see Clive Brown’s article ‘Ferdinand David as Editor’ on this website].

David’s first edition of Paganini’s caprices was published in Leipzig by Breitkopf and Härtel in 1854. Evidence suggests that he used Breitkopf and Härtel’s 1823[?] publication (plate no. 3936) as his principal source, as his musical text incorporates errors and omissions in common with those identified in that edition. Among these, for example, the second note of bar 25 of no. 1 is given as b’ instead of c’’ and the piano indication in bar 72 is omitted; in no. 2, the dolce at bar 67 is omitted; in no. 7 the tempo/character heading ‘Posato’ is lacking, as is the term ‘minore’ heading bar 23 of no. 17 and bar 32 of no. 18; and the accent is missing in bar 4 of no. 24. However, the text is ssubstantially accurate regarding note lengths and pitches, rhythms and articulations and his copious editorial additions (fingerings, bowings, indications of expression, bow division etc.) became a model for subsequent editors, exploring a broad expressive canvas and demonstrating the most thorough and detailed practicable solutions thus far to the formidable technical challenges posed. In addition to negotiating all the complexities of cross-string position work, double and multiple stopping, chromatic fingerings and timbral requirements, he broke up some of Paganini’s long slurs/phrasings (for example, in no. 6) for ease of performance, introduced slurs across the bar line in order better to shape phrasing within a legato context (for example, no. 12), emphasized and articulated multiple stopping with successive down bows (for example, nos. 20 and 23) and brought more consistency to Paganini’s approach to bowing in, for example, no. 7. Some characteristic hallmarks of David’s fingering style also stand out, notably his use of open strings and extensions to facilitate/avoid shifts or unnecessary cross-string activity, his use of harmonics where appropriate and convenient (e.g. no. 19) and his employment of semitone shifts and fingering patterns which involve successive use of one and the same finger and/or invite portamento (e.g. no. 21), even though the exigencies of Paganini’s notation often impose certain restrictions on the performer’s range of fingering options.

David also issued an edition of Paganini’s caprices with piano accompaniments by Robert Schumann. Evidently, as with J. S. Bach’s unaccompanied works, David ‘would not be moved by any fee whatsoever to step onto a stage with only a naked violin. Only when Mendelssohn surprised him one day with the accompaniment he had prepared for the [Bach’s] chaconne did David declare himself ready for a performance in that company’ [Andreas Moser, ‘Zu Joh. Seb. Bachs Sonaten und Partiten für Violine allein’, Bachjahrbuch 17 (1920), pp. 45-6]. It was also commonly believed that the public required such accompaniments to facilitate their comprehension of the music. Schumann’s accompaniments (1855) date from the final years of his life in Endenich and were published in a number of later editions [See, for example, the edition by Georg Schünemann (1884-1945; Frankfurt: Peters, 1941)]. As a 19-year-old law student, Schumann had attended one of Paganini’s concerts in Frankfurt in 1830; that exposure to the Italian violinist’s spell-binding powers exerted a profound influence, as is evidenced also in Schumann's six Etudes pour le pianoforte Op. 3 (composed in 1832) and his six Etudes de concert Op. 10 (1833), both of which are based directly on Paganini's caprices. However, his accompaniments for the caprices offer important evidence of his ‘take’ on these works, especially in terms of harmony, texture, rhythm and form.

Another edition involving Ferdinand David was issued in two books of twelve caprices each ‘mit hinzugefügter Begleitung des Pianoforte von Ferdinand David’ (‘with an additional piano accompaniment by Ferdinand David’; Leipzig: Breitkopf und Härtel, n.d. [c.1860?]; plates: 12200 and 12201). The violin parts appear unchanged from David’s 1854 edition (plates 8790 and 8791). The accompaniments have occasional thematic relevance (as in no. 6, bars 1ff.) but are largely light and supportive, often adopting an ‘oom-pah’ pattern (as in no. 1, bars 28ff.), marking the rhythm, maintaining the pulse and filling in the harmony (sustained chords, for example, underpinning the violin’s arpeggios and scale passages for the opening of no. 5). The English pianist, composer, singer and conductor John Liptrot Hatton (1809-86) followed David’s example by publishing a version of the 24 caprices ‘avec accompagnement de pianoforte’ (London: J. Hart, n.d. [1870]). At the end of the century the piano’s accompanying role was raised to concertante significance in a version (Berlin: Simrock, 1898; plate: 11014) made by Russian violinist Vasily Bezekirsky (1835-1919), who spent most of his career as leader of the Bolshoi Theatre Orchestra.

Among other influential nineteenth-century editions of the caprices were those by Lambert Massart (1811-92) [See Borer, The Twenty-four Caprices, pp. 268-70], Jean Becker (1833-84) and Edmund Singer (1830-1912). Becker’s edition (Leipzig: C. F. Peters, n.d. [c.1875]); plate: 6386) is closely modelled on David’s, perpetuating various textual errors in the process, but Massart’s (Paris: Schonenberger, 1851) and Singer’s (Braunschweig: Henry Litolff, 1892; plate: 2009) are the products of their renown as pedagogues. Singer’s edition was specifically designed for study purposes – ‘zum Studium genau bezeichnet’ (‘precisely annotated for study’) and incorporates extremely detailed performance instructions regarding bowings and fingerings, articulation indications, and additional marks of expression, leaving little for the performer to contemplate. Philippe Borer also mentions an edition of the caprices by the Austrian violinist and last bastion of the Viennese tradition of violin teaching Arnold Rosé (1863-1946; Vienna: Universal, n.d.; no. 315), which the present writer has thus far failed to locate [Borer, The Twenty-four Caprices, pp. 268-70].

Some nineteenth-century pedagogues used Paganini’s caprices as an ultimate technical goal for their pupils and composed various preliminary studies or exercises for them gradually to attain that goal. One such was Ferdinando Giorgetti (1796-1867), an heir of the Nardini school of string playing via his mentor Gian Francesco Giuliani. Giorgetti met Paganini in Florence and described Paganini’s performance practices in some detail in an article in Florence’s Gazzetta Musicale (16 March, 1840) and later published his Sei Studi per violino per servire di esercizio preliminare a quelli di Paganini ed dedicati ai suoi allievi Op. 28 (Milan: Ricordi, [1870]) (‘Six Studies for Violin to serve as Preliminary Exercises for those of Paganini and dedicated to his students’).

Emil Kross (1852-1917) is also well known for his pedagogical literature. His series of Classische Studienwerke für die Violine, nach den Technischen Ansprlichen der Neuzeit bearbeitet, mit systematischen Fingersätze, dynamische Zeichen und erläuternden Anmerkungen versehen included a Grosse praktisch-theoretische Violinschule in 3 Theilen, three volumes of Systematische Scalen-Studien Op. 18 and original etudes, as well as detailed annotated editions of studies by, amongst others, Baillot, Mazas, Kreutzer, Rode, Spohr, Gaviniès and Paganini. Kross’s Ueber das Studium der 24 Capricen Paganinis, und die Art und Weise, wie diese durch Paganinis Hand- und Arm-stellung auch von kleineren Händen überwunden werden können (Mainz: Schott, 1897; Eng. trans. Gustav Saenger, New York: Carl Fischer, [1908]) provides instruction for the study of Paganini’s caprices especially by violinists with small hands. It includes fifteen photographs demonstrating posture, hand position and other such technical details and presents the caprices in what is, in his view, their progressive order of difficulty: 16-5-11-10-15-21-22-7-14-13-12-9-8-1-23-6-19-18-2-3-20-4-17-24 [A. Pacini’s 1829 publication also presents the caprices in a different order – see Cantù/Herttrich edition (Munich: Henle, c.1990), Preface, p. V]. Only nos. 18 and 24 retain their original positions in the sequence. Kross provides recommendations regarding fingering, positions, bowings and performance indications as well as textual observations regarding their execution. He preserved his order of presentation in later independent ‘newly revised’ editions of the caprices (Mainz: Schott, 1912 and New York: Carl Fischer, 1922), recommending some of his own pedagogical publications for their preparatory study.

Twentieth-century Editions: an overview

Despite the steady stream of editions of Paganini’s caprices in the nineteenth and early twentieth centuries, in line with these works’ established position as benchmarks of a violinist’s technical and interpretative pedigree, it is fair to record that no single publication from that period ‘does complete justice to the autograph in either form or substance’ [Neill and Accardo (eds.), Preface p. VIII]. The vast majority of these publications perpetuate former textual errors while demonstrating some of the often widely divergent stylistic hallmarks of their celebrated editors. In typical fashion, editors such as Singer and Enrico Polo (1868-1953; Milan: Ricordi, 1913; plate: E.R.226) altered tempo indications in several caprices (e.g. in no. 6 ‘Lento’ becomes ‘Adagio’; in no. 7 ‘Posato’ becomes ‘Moderato assai’ and in no. 13 ‘Allegro’ becomes ‘Allegro non troppo’) and incorporated substantial modification of several passages, including alteration of various rhythmic values.

Edouard Nadaud (1862-1928), a pupil of Dancla at the Paris Conservatoire who returned to his alma mater as a professor in 1900, evidently published an edition of Paganini’s caprices around the time of his appointment (Paris: Costallat, n.d. [c.1900]) [See Edlund, Music for Solo Violin Unaccompanied. Arrangements by Nadaud of three of Paganini’s caprices also survive in a pedagogical publication by the Paris firm Billaudot, [n.d.]; CC 2249)]; and his professorial colleague in Paris, (Narcisse)-Augustin Lefort (1852-1925), brought out a version in 1916 (Paris: Durand, 1916; plate: D&F9467). French violinist Emile Sauret (1852-1920), who studied under de Bériot, Vieuxtemps and Wieniawski and gained the admiration of Liszt (who partnered him on an American tour in 1872), published his edition of the caprices in London (London: Augener, n.d. [1917]); plate: 15071), where he served as violin professor at the Royal Academy of Music between 1890 and 1903 and from 1908 at Trinity College of Music. Incorporating various misinterpretations of the original sources, Sauret’s version also includes modifications regarding articulation, particularly in his approach to Paganini’s staccato markings (no. 1, bars 25-6 and 50-52) and his addition of slurs for more sostenuto outcomes (e.g. the cross-beat slurs in bars 16-20 of no. 16) and caesuras (e.g. in no. 13) to articulate phrasing. Two noteworthy editions appeared in Paris in the 1920s, one by Gaston Marchet (Paris: Maurice Sénart et Cie, 1920; plate: 5280) and the other by Alberto Bachmann (Paris, Emile Gallet, 1921), this latter incorporating accompaniments for a second violin.

Roughly contemporary German editions included one by Oskar Biehr (1851-1922; Leipzig: Friedrich Hofmeister, 1908) and one by the American child prodigy, Florizel von Reuter (1890-1985; Leipzig: Eulenberg, 1924). The title-page of Reuter’s reads, ‘Revised – often freely – for works of study and in the concert execution’ [Borer, The Twenty-four Caprices, p. 269]. Reuter also provides a detailed preface and a short introduction to each caprice. However, Philippe Borer’s implication that Adolf Busch published an edition of Paganini’s Op. 1 set has no grounding in fact [Ibid.; see below for details of Busch’s arrangements of Paganini’s caprices].


Hungarian violinist Carl Flesch’s (1873-1944) edition of Paganini’s opus (Leipzig, C. F. Peters, n.d. [1910?]) has survived in several reprints issued by a variety of publishers. Although his musical text perpetuates many errors, Flesch’s intellectual approach to editing and the thoroughness and logicality with which he attends to fingerings and bowings, strikes that vital balance between technical and musical concerns; such qualities have consistently marked his edition as one of the most practicable available (illustrated, for example, by the manner in which he breaks up Paganini’s very long original slurs in both the opening and Presto sections of no. 3, his use of an asterisk above the e’’’ in the second descending scale figure in the opening section of no. 5 to indicate that that note ‘need not be taken at all, as in the quick time the open string sounds an octave higher’ – Yampolsky and other successors do likewise), even if it does bear some style hallmarks of its time. In no. 9 Flesch provides an ‘ossia’ passage in harmonics for the flute imitations in the final statement of the rondo idea; and in the Presto of no. 11 he replicates faithfully the ‘contro’ bowing prescribed by Paganini, one of the few editors to do so for the dotted rhythm rather than ‘hook’ the dotted semiquaver and demisemiquaver in one and the same articulated bow stroke. His annotations in his edition of the caprices are supplemented by several observations on technical and interpretative matters in his other pedagogical writings, notably in the first volume of his Die Kunst des Violinspiels (2 vols., Berlin, 1923 and 1928; Eng. trans. Frederick H. Martens, as The Art of Violin Playing, London, 1924 and 1930) and in his Violin fingering: its theory and practice (London: Barrie & Jenkins, 1966). Among the issues explained in some detail are the role and placement of the little finger in the thrown arpeggios of no. 1, the requisite contact-point compromises for optimum tone production for the left-hand tremolos in no. 6, recommended octave fingerings in nos. 3, 17 and 23, the execution of the saltato bowing in no. 5, options for fingering the double stopping in sixths in no. 21 and the realization of the left-hand pizzicatos in the ninth variation of no. 24 [Flesch, Die Kunst des Violinspiels, Eng trans., vol. 1, pp. 78, 142, 139-41, 138, 50].

Flesch’s compatriot Jenö Hubay’s (1858-1937) pedagogical approach initially comprised a symbiosis of French, Belgian and German elements, but Hubay’s later espousal of Paganini’s caprices indicated a shift of emphasis towards the development of virtuosity in his pupils. Nevertheless, his edition (Vienna: Universal-Edition, [c.1925]; plate: UE7159) is typically geared towards the cultivation of artistic interpretations that are as respectful as possible to the spirit and style of the composer [Hubay also arranged caprice no. 13 for violin and piano].

The ‘modern’ Soviet violin school came to prominence in the 1920s and 1930s, founded by the Ukrainian violinist-pedagogue Abram Yampolsky (1890-1956), who became a professor at the Moscow Conservatory in 1926. Developing the teaching legacy of Auer, Yampolsky published an edition of Paganini’s Op. 1 dating from around the period of his elevation to the headship of the Conservatory’s violin department in 1936 (Moscow, Muzyka, n.d. [c.1935]); plate: 14178). Spaciously presented over 61 pages, it incorporates many of the by then customary misreadings, but it also includes practicable fingerings, many of which show similarities with those of Flesch (e.g. in no. 5), some individual approaches to bowing (no. 4, b. 21; no. 7; no. 11, sporadically resorting to Paganini’s ‘contro’ bowing when convenient; no. 24 var. 8 using down bows throughout), additional interpretative signs such as dynamics, hairpins, crescendos and diminuendos and other expressive markings (nos. 1, 4, 16), accents or lines (nos. 2, 3, 10, 12, 17, 19, 20, 22) for different emphases, commas for articulation (nos. 14, 18), suggestions for tempo rubato (no. 13), and timbral indications regarding the particular string to be used. There are instances, too, where Paganini’s annotations have been modified (many of the fortes are transformed into sforzandos in no. 16) or omitted, and Yampolsky offers an ossia passage in harmonics for the flute imitations in the final statement of the rondo idea of no. 9. He also stresses ‘la melodia ben cantando’ in no. 6, implying at the same time the use of a quasi portato stroke rather than a wholly legato slur to realise such cantabile goals.

The Athens-born medical doctor Demetrios Dounis (1886-1955), a violin pupil of César Thomson, concentrated on the physical aspects of violin playing and promoted efficiency in practising in his The Artist’s Technique of Violin Playing Op. 1 (New York, 1921), 'a new scientific method for achieving absolute mastery of advanced left-hand and bowing technique in the shortest possible time.' Its focus and the thrust of his other pedagogical work was on the cultivation of finger independence and mastery of a wide range of bow-strokes, sometimes pursued at the expense of musical logic. In his edition of Paganini’s caprices Dounis claims that ‘primary consideration has been given to the expression of the musical thought through correct phrasing, thus enabling the player to cultivate and develop a pure musical sense’ [London: J. H. Lavender, The Strad Edition, 1949, plate: S.E. 164; Author’s preface]. Dounis considers that ‘any phrasing, that is bowings and fingerings, that is influenced by technical limitations, instrumental considerations or traditional routine is faulty as far as musical expression is concerned’ [Ibid]. Some technical guidance is occasionally provided with the musical text, which nevertheless incorporates a number of errors or modifications, especially with regard to rhythms (e.g. in no. 7 bars 61 and 74), additional dynamic markings (notably in no. 17) and some unconventional bowings, notably the common occurrence of re-taken down-bows (as in no. 7 bar 33 onwards and also in bars 62 and 64).

Some editors of the period emphasised the study element of Paganini’s caprices. Renzo Sabatini, for example, published his L’arte di studiare I 24 capricci di Paganini per violino (Milan: Ricordi, 1937), and Harold Berkley's (1896-1965) edition of the caprices (New York: Schirmer, c.1944; plate: 40428) is described as a ‘study-version’. Initially mentioning studies which facilitate the minimum preparation for the performance of the caprices (among them Dont’s Op. 35, de Bériot’s 60 Studies Book 1 and some of Sauret’s Grandes Études), Berkley warns players that they should not overlook the existing demands made on the right hand in order to master the technical challenges posed for the left hand by Paganini’s caprices. He focuses especially on adding appropriate expressive indications ‘as a guide to the student in developing his own interpretations’ and because ‘Paganini inserted comparatively few’ [Paganini, Twenty-four Caprices for the Violin Op. 1, ed. Harold Berkley, (New York: Schirmer, 1944), Preface, p. iii]. He provides extensive guidance regarding the preparation and interpretation of selected caprices, including slow practice (for no. 1), ‘practice’ or alternative bowings (for nos. 1, 2 and 13), exercises (e.g. for developing strength, independence and flexibility of the fingers in no. 6), editorial tempo indications, additional performance fingering suggestions (including fingered octaves for the scales in the A minor section of no. 9 and artificial harmonics (bars 76-81) and double artificial harmonics for the reprise of the theme towards the end (bar 95), and a special sign to indicate ‘advance fingering’, that is ‘the placing of a finger on the string in preparation for a note which immediately follows’ [Ibid., p. iii].), and bow division. He points out that the double stopping in the da capo of no. 17 is occasionally played in double harmonics, but does not include that variant ‘as its use is far-fetched and not to be recommended.’ [Ibid., p. v]. Berkley also perpetuates Kreisler’s upside-down bowing (starting with a down bow) for the opening of no. 13 and suggests adding embellishments for certain passages in concert performance, including, for example, adding extra octaves in scales and arpeggios, notably extending the final arpeggio of no. 24 to four octaves in order to create an ‘additionally brilliant ending’ [Ibid.].

Instead of preparing an edition of all 24 of Paganini’s caprices, some early twentieth-century violinists were selective regarding the caprices they edited and, in many cases, arranged quite freely for performance. Leopold Auer (1845-1930), John Dunn (1866-1940), Fritz Kreisler (1875-1962), Arthur Catterall (1883-1943), Rowsby Woof (1883-1943), Albert Spalding (1888-1953), Jacques Thibaud (1880-1953), Adolf Busch (1891-1952), Joseph Szigeti (1892-1973), and Joseph Achron (1886-1943), among other violinists, published arrangements of selected caprices, mostly for violin and piano [Eddy Brown, Edward Behm, Erich Itor Kahn and Max Vogrich, among others, have also arranged selected works from Paganini’s Op. 1].

Auer, for example, published arrangements of nos. 13 and 24 ‘for concert use’ with piano accompaniment (Berlin: Bote & Bock, [n.d.]; plate 11457 [no. 24 was recorded by, amongst others, Jascha Heifetz [available on YouTube: http://www.youtube.com/watch?v=vPcnGrie__M last accessed 15/08/2012]. While the first five variations bear fairly close resemblance to the original in the violin part – there is an eight-bar additional introduction for the piano, based on the principal melodic motive, before the announcement of the theme itself – some noteworthy modifications are made in variations 6-9 inclusive. For example, the first section of variation 6 is repeated, complete with a second-time bar, and 16 bars for the piano (Più mosso) are added; and both sections of variations 7-9 inclusive are repeated. Variation 10 features artificial harmonics as opposed to Paganini’s original stopped notes and Variation 11 is almost entirely of Auer’s own composition, including left-hand pizzicatos and bearing little resemblance to the autograph.

Typically, Kreisler relished adding his own personal touches to others’ work. His personal library of music for violin and piano, housed in the Library of Congress, includes ‘a proliferation of alternative ideas written in or pasted over the original printed editions’ [Eric Wen, ‘Miniature masterpieces’, The Strad vol. 98 no. 1161 (January, 1987), p. 53]. His arrangements for violin and piano of Paganini’s caprices nos. 13, 20 and 24 (Mainz and Leipzig: B. Schotts Söhne, 1911?; London, Schott & Co., c. 1913; plate: 30962) demonstrate considerable artistic licence. He makes major structural changes in no. 20, omitting the da capo reprise of the opening D major section, adding a second time bar and some left-hand pizzicato and, somewhat oddly, concluding the caprice in the relative key of B minor. Further, his version of no. 24 comprises only eight, as opposed to Paganini’s eleven variations. The theme is marked ‘Moderato’ and ‘forte’ instead of Paganini’s ‘Quasi presto’ and ‘piano’ and, while the first two variations are mostly faithful to Paganini’s original, Kreisler’s third and seventh (poco più mosso) variations equate largely with Paganini’s fourth and ninth, his fifth variation is similar to Paganini’s sixth but is headed ‘Andante con moto’ and marked ‘con sordino ad libitum’, his fourth, sixth and eighth variations are almost entirely newly composed, and his original coda is written in harmonics throughout, concluding mysteriously rather than fortissimo in Paganini’s virtuoso manner. Elsewhere, notes have been altered (e.g. no. 13 bars 15-16, 36-37; no. 20 bars 17ff); artificial and natural harmonics have been introduced (in no. 20 bars 28, 32, 34 and 36); tempo headings have been modified; numerous subtle inner tempo changes and rubato indications have been included; dynamic and other expressive markings have been added, along with guidance for bow division (‘à la pointe’, ‘du talon’ etc.); and the overall approach seems to be more of a legato one than Paganini’s in the use of slurs and, in some cases, lines as opposed to the original dots. His down-bow start to no. 13 seems particularly unconventional.

Kreisler’s piano accompaniments vary in significance: that of no. 13 is largely supportive of the violin’s thematic material and bolsters up the middle section towards its climax at bar 38; that of no. 20 involves semiquaver movement in the first section, providing more momentum, while the second section is lightly scored and includes occasional motivic interest; that of no.24 is largely complementary and supportive, giving the pianist occasional opportunities to shine without getting in the way of the violinist’s virtuosity.

Of the other violinists of the period who contributed independent versions of selected caprices by Paganini for violin and piano, Arthur Catterall arranged no. 13, Rowsby Woof no.14 (London: Anglo-French Music Company, 1921)], and Jacques Thibaud (http://www.youtube.com/watch?v=bpWLDeJZppg]), John Dunn (London: Hawkes & Son, 1924)] and Albert Spalding (Boston: Oliver Ditson & Co., 1918; plate: 5-149-72046-8) each arranged no. 9. Spalding’s version of no. 9 is substantially faithful to Paganini’s notes but prescribes more interpretative detail regarding dynamic indications and tempo modifications; and it gives the piano part more than a supportive role by including some effective imitation, as well as figuration that interconnects the caprice’s three principal sections.

Adolf Busch evidently performed Paganini’s caprices unaccompanied at soirées at his home but favoured performing arrangements with piano as part of his celebrity recitals. Although he did not publish an edition of Paganini’s Op. 1, he made several arrangements of independent caprices and for various instruments and/or instrumental combinations. His arrangements for violin and piano were made at various times and include nos. 9, 13, 17, 19 (1941) and 24 (1918). All but one are in manuscript and housed in the Paul Sacher Stiftung in Basel; the arrangement of no. 17 was published in Leipzig by Breitkopf und Härtel in 1926 (EB 5339). Busch also arranged no. 24 for chamber orchestra [Information kindly provided by Tully Potter in correspondence with the author 10/12/2012].

Joseph Szigeti contributed a ‘concert edition’ of no. 2, but for unaccompanied violin (New York: C. Fischer, 1937). Including several additional markings, his version provides: extensive fingerings, including various alternatives, above and below the stave, for certain passages; bowings, including an unconventional ‘upside-down’ bowing at bars 24ff and 35ff; recommendations regarding the part of the bow to be employed; additional accents and cautionary accidentals; copious dynamic indications; natural and artificial harmonics (bars 59 and 60); and ossia passages – for example, for bars 29-31, involving natural and artificial harmonics, and for bars 62 onwards, involving octave passagework, accents, harmonics, and trills with turn terminations, and changing the character and notation of the original to create a more ‘energico’ conclusion, with a dotted rhythm anticipatory of only a double-stopped final chord [Szigeti recorded versions of Caprices nos. 2 and 9 (Columbia 68555D)]. In one of his pedagogical publications Szigeti points out how ‘the many accretions and simplifications' of previous editions had ‘falsified the picture we had of these masterpieces of instrumental ingenuity’ [Joseph Szigeti, Szigeti on the Violin (London: Cassell, 1969), p. 199]. He claims that Caprice no. 9 (‘La Chasse’) corroborates his view regrading ‘a regression in our use of the bow. In spite of Paganini’s explicit instructions sulla tastiera imitando il flauto and sulle terza e quarta corda, imitando il corno, which presupposes the bow alternation [down-up] and longish strokes, we nowadays only hear it with the monotonous [up-up-down-up-up-down-up-up] bowing, that the most widely used editions of the last decades have popularized. Gone is any attempt to imitate flutes or horns’ [Ibid.]. He asserts that Paganini’s original bowing [down-up-up-up-down-down-down-up-up-up etc.] is far more demanding (NB ‘the difficulty of alternating down-bow and up-bow chords, using the (presumably bouncing) bow at the lower half for the two semiquavers that precede each chord and keeping down-bow and up-bow indistinguishable in weight and quality’) than the ‘workaday solution’ advocated. ‘The labour-saving device advocated by all the editors follows the line of least resistance’ [Ibid., pp. 199-200]. Szigeti points out another feature of the Paganini manuscript – that the saltando episode is indicated ‘alternating down-bow and up-bow and not, as in current editions, each group of demisemiquavers down-bow, each quaver up-bow’ [Ibid., p. 200]. Singer’s edition evidently did not ‘suppress (or rather censure) the original bowing’ but gave the choice of both. Thus Szigeti played it throughout his career with the original alternating saltando [Ibid., p. 200-1].

The Lithuanian violinist-composer Joseph Achron, a pupil of Leopold Auer in St Petersburg, made transcriptions of Paganini’s caprices between 1919 and 1923 [Their publication by the Joseph Achron Society with an historical introduction by Maiko Kawabata and a performance introduction by violinist and conductor Yuval Waldman is forthcoming]. Originally commissioned by Jascha Heifetz, they are described as ‘highly creative and complex transcriptions [that] shed new light on the many harmonic, contrapuntal, structural, and performance possibilities of the famous Paganini Caprices’ [http://www.josephachron.org/achron-paganini-caprices.html accessed 9/12/12]. Achron’s approach to fingering is the principal focus, not least because he ‘rediscovered’ and promoted a one-finger-per-note fingering for chromatic scales previously advocated by Francesco Geminiani in his The Art of Playing on the Violin (London: J: Johnson for the Author, 1751).

Paganini’s long cantabile melodies have inspired numerous composers to add harmony and stylistic definition to selected caprices, in some cases re-composing the original. Karol Szymanowski was assisted initially by the violinist Viktor Goldfeld in arranging his Trois caprices de Paganini Op. 40 (Vienna: Universal Edition, c.1926) and the duo premiered these versions of Caprices nos. 20, 21 and 24 in Elisavetgrad on April 25, 1918 [The published revision of the violin text is by Szymanowski's violinist friend and the work's dedicatee, Pawel Kochánski]. Szymanowski’s piano accompaniment for no. 20 has been likened to parts of his Violin Sonata of 1904 [see http://www.answers.com/topic/paganini-caprices-3-for-violin-piano-op-40-m42#ixzz20yK3jCYo accessed 5/12/2012] and his re-arrangement of no. 21, involving a recomposed slow first section and omission of Paganini's fast second section, has similarities with the exotic, mystical and sensual style of his First Violin Concerto (1916). This is developed further in his complete transformation of no. 24, which comprises only ten variations. Szymanowski omits three of Paganini’s eleven variations (vars. 5, 10 and 11), adds two of his own composition by way of conclusion (an Andante dolce exploiting violin artificial harmonics throughout and a ‘vigoroso’ variation involving much double and multiple stopping and fleet passage-work) and re-orders several others. He reverses the order of Paganini’s second and fourth and eighth and ninth variations, his fifth and sixth variations are based on Paganini’s sixth and seventh variations, and his seventh and eighth variations use material from Paganini’s ninth and eighth variations respectively. The violin part of variations 1, 3 and 6 is fairly faithful to Paganini’s original text, although the third variation has an additional opening bar for piano and a modified final bar, and the sixth includes some note-changes, different dynamic and tempo markings and a da capo ad libitum. Other modifications in Szymanowski’s text include: the introduction of harmonics and descending glissandos (bar 16 of the theme, headed Vivace); timbral recommendations; additional expressive indications and tempo changes; a repeated second section and a da capo ad libitum for the theme; an obligato con sordino for variations 3-5 inclusive; tonal and structural modifications (e.g. in the second section of variation 4); and changes in content (bowing and dynamic markings in the first section of variation 5, with the second section amended considerably, maintaining the bowing changes but requiring mostly double stopping in thirds instead of Paganini’s tenths and ending in A major with extra bar).

Darius Milhaud’s Trois caprices de Paganini Op. 97 (Paris: Heugel, 1927), arrangements for violin and piano of Paganini’s caprices nos. 10, 13 and 22, are in a similar category to Szymanowski’s versions, but in this case extending the virtuosity to the pianist while remaining substantially faithful to Paganini’s violin text. As Barbara L. Kelly puts it, Milhaud ‘makes small changes to the bowing but otherwise remains faithful to the Paganini violin part. Milhaud sets up a dialogue to Paganini’s monologue. This is reinforced by the way in which he attributes authorship on the first page of the score. Instead of violin and piano, he marks “Paganini” and “Milhaud” next to the respective parts, suggesting a context between the two men’ [Barbara L. Kelly, Tradition and Style in the works of Darius Milhaud 1912-1939, (Aldershot, Ashgate, 2003), p. 170].

Around the middle of the twentieth century there was a flurry of editions of the 24 caprices, among them publications by the violinist of the celebrated Trio Italiano, Alberto Poltronieri (Milan: Carish, 1945 [The other two members of the Trio were cellist Arturo Bonucci and pianist-composer Alfredo Casella], Michelangelo Abbado (1900-78; Milan: Edizioni Suvini Zerboni, [1946]; plate: S.4165.Z), violin professor at the Milan Conservatory, the Prague violin professor Jindřich Feld (Kraków: PWM, 1951), French violinist-composer Raymond Gallois-Montbrun (1918-94; Paris: Leduc, 1952) and Paris Conservatoire violin professor René Benedetti (Paris: Choudens, 1952). Franz Schmidtner (1913–1969), violist of the Hamburg String Quartet and the Radelow Streichquartett, included an edition of the caprices amongst his varied selection of edited works (Hamburg: Hans Sikorski, 1954; plate SIK 0189), as did also the Italian violinist-composer Remy Principe (1889–1977; Milan: Edizioni Curci, n.d.[1958]; plate E. 6496 C) and the Russian pedagogue (and teacher of Ivan Galamian) Konstantin Mostras (1886-1965; Moscow: Murghiz, 1959). Szigeti and Edward Neill single out Abbado’s edition for rectifying some previous misinterpretations, Neill describing it as ‘philologically rigorous’ in closely following the autograph [Paganini, Capricci Op. 1, ed. Neill and Accardo, Preface, pp. VII-VIII].

Early twentieth-century trends towards free adaptation of Paganini’s caprices continued into the latter part of the century and beyond. For example, players such as Austrian-born Max Rostal (1905-91) and Polish violinist Waclaw Niemczyk (1917-95) arranged some caprices for violin and piano [Rostal published an aarrangement of no. 20 (London: Novello, 1955?) and Niemczyk versions of nos. 13, 20, and 24 (London: B.V.F. Anglo-continental Music Co., 1965)].

Rostal transforms no. 20 into an even more virtuoso concert piece for violin and piano, modifying its structure by omitting its da capo and adding a brief and witty codetta based on its characteristic trill figure. He also adds expressive descriptors to Paganini’s text and expands its dynamic range, transposes some passages up an octave, spells out performance details such as spread multiple stopping and trill execution, and introduces occasional natural or artificial harmonics and left-hand pizzicato. He suggests an unusual (even unnatural) bowing for the opening of the second section, taking the first three semiquavers of each group of six in a slurred up bow and the second three strongly detached in one down bow and requiring the strong beats of the bar to be accented. The piano part is unusually busy and thick, with several indications for sustaining pedal usage.

Throughout the century numerous arrangements were also made for solo instruments other than the violin; and William Zinn even made a convincing arrangement of the opus for string quartet, distributing its virtuoso demands between the four players and taking licence to make changes of register, add harmonies and counterpoints, change articulations and even, in the case of no. 24, extend it by adding some bars by way of an introduction to the theme [Recorded by the Wihan Quartet, NIMBUS 6113 (2009)]. Ruggiero Ricci recorded a version of the caprices for solo violin and orchestra based on Schumann’s piano accompaniments [80th birthday concert 1998 with Weiner Chamber Orchestra, Richard Weninger (conductor) issued in 1999 by Dynamic]; and, encouraged by some editors and various performing traditions surrounding Paganini, many modern performers have adopted a free approach to their interpretations of the caprices. This was often of necessity in the case of early recorded performances, due to the exigencies and limitations of the recording process [The first recordings of any of the caprices date back to Franz von Vécsey’s of nos. 2 and 14 in 1911 and Jan Kubelík’s of no. 6 in 1912; the most readily accessible early recordings are the acoustical waxings of nos. 13 and 20 by the teenage Jascha Heifetz in 1918 and 1920. Heifetz re-recorded nos. 13 and 20 in 1934 and 1956 and made a recording of no.24 in 1934]. The first ‘complete’ recording of the twenty-four caprices was made by Ossy Renardy (with pianist Walter Robert) in 1940, the centenary year of Paganini's death. It uses Ferdinand David’s mostly skeletal piano accompaniments and includes heavy cuts because of the need to record two caprices on each side. Ruggiero Ricci (1918-2012) made the first recording of the original solo violin version in 1947 – and without splicing or editing!

Ricci possessed formidable technical facility and was renowned for his daredevil approach to violin playing, making Paganini’s caprices very much his own. Despite the public availability of a facsimile of the autograph [with an introduction by Yehudi Menuhin and a preface by Federico Mompellio (Milan: Ricordi, 1974)], his edition of these works (Budapest: Editio Musica, n.d. [1984]; plate: Z.12 552) incorporates several inaccuracies; however, it does reflect some of his views regarding Paganini’s ‘secret’, which involved keeping a very close contact with the instrument and adopting a manner of holding the violin in which the left wrist maintains a position against the ribs of the instrument and the hand and fingers are extended backwards or forwards in an arc, both in front of and behind the thumb, which acts as a pivot point; playing in first position thus involves a backwards extension [Ruggiero Ricci, ‘Secret history’, The Strad vol. 115 no. 1374 (October 2004), p. 1041]. The very opening of no. 19, the una corda passage from bar 27ff. of the same caprice, and bars 55-56 of no. 3 provide instances where such a philosophy is enormously beneficial, spawning some unusual but workable fingerings, and such economy of hand movement and use of pivot technique contribute substantially to realising the sustained quality of the first section of no. 11. Some of Paganini’s bowings are also amended for more comfortable performance, the four-bar phrasings in no. 3 (bar 25 onwards), for example, being taken in two-bar groups, and many of the original dynamic and other indications have been modified to fulfil Ricci’s expanded expressive regime, most notably in nos. 4, 6 and 14. Additional trills are also introduced (e.g. in no. 13), along with two ossia passages, one including an artificial harmonic to replace the original doubled-stopped octave (no. 19) and one (no. 24, var. 9, bar 115) in which both Ricci’s versions veer away from Paganini’s simpler pattern of alternating bowed and left-hand pizzicato notes.

Ivan Galamian’s (1903-81) analytical and rational approach to violin playing is mirrored in his edition of the caprices (New York: International Music Company, 1973; plate: IMCO 2292), which reminds one to some extent of Flesch’s in its principles. Despite some implied reverence to the facsimile of the autograph, several errors are reproduced in the musical text, but Galamian provides some practicable guidance regarding bow division (adding instructions such as ‘at the point’ (no. 2, bar 24) or ‘at the frog’), dynamic indications (including ‘hairpins’), trill terminations (including a boldly bowed out trill termination in the penultimate bar of no. 24) and, in particular, viable fingerings. In line with the philosophy outlined in his treatise, his fingering principles embrace both musical and technical considerations in order to realize ‘the best sound and finest expression of the phrase’ and ‘make the passage as easy and as comfortable as possible’ so long as the musical purpose is not ‘sacrificed to comfort’ [Ivan Galamian, Principles of Violin Playing and Teaching (London: Faber and Faber, 1964), p. 31]. Consequently, his recommended fingerings include many extensions beyond the frame of the left hand and many examples of what he calls ‘creeping fingering’ for producing clearer and cleaner articulation, involving a ‘change-of-position technique which eliminates the shift and is based on extensions–occasionally on contractions–followed by a readjustment of the hand’ [Ibid., p. 34].

Franco Gulli was greatly infuenced by the opinions and playing style of Joseph Szigeti, taking on board also, via Arrigo Pelliccia, some of Carl Flesch’s theories [See Franco Sciannameo, ‘A conversation with Franco Gulli’, The Violexchange, 2/1 (1987), p. 2]. His edition of Paganini’s Op. 1 (Milan: Edizioni Curci, 1982) is a carefully prepared attempt to follow the composer’s autograph with substantial fidelity, heralding the contemporary fashion for Urtext critical editions, which purportedly correct the many mistakes that have accumulated over time and present the original musical text without editorial addition or emendation but with scholarly critical commentary. This fashion spawned three notable publications of the caprices under that banner in the 1980s and 1990s. The editors concerned, Klaus Hertel, the duo Edward Neill and Salvatore Accardo, and the trio Alberto Cantú, Ernst Herttrich and Renato de Barbieri, considered it worthwhile to determine an original text in a scholarly manner, even though Paganini would doubtless have considered his notated version as one among a number of equally possible performance alternatives.

Hertel’s edition (Leipzig: Peters, n.d. [1988]; plate: E.P. 13406) is presented as an Urtext ‘based on the facsimile of the autograph…as well as the first edition of 1820’ [Editor’s preface]. It includes editorial suggestions for fingering and for phrase and dynamic markings – these are clearly distinguished from those of Paganini by dotted lines or brackets – and editorial notes, referenced by asterisks in the score, in both German and English at the end of the volume. Although harmonic effects are not included in this edition’s primary sources, Hertel suggests an ossia passage in the opening section of no. 9 for a version in harmonics of the principal theme, possibly because Paganini often flaunted them in performance. Indeed, Tadeusz Wrónski claims in a foreword to his own edition of the caprices that Paganini is believed to have executed the final refrain of no. 9 in double harmonics [See Tadeusz Wrónski’s edition of Paganini’s caprices (Krakow: Polskie Wydawnictwo Muzyczne, 1977), Foreword, pp. 4-5]. However, despite Hertel’s Urtext claims, numerous errors of transcription remain in his edition.

A much more scholarly edition combining the critical editing skills of Edward Neill and the technical knowledge and creative artistry of Salvatore Accardo (Milan: Ricordi, 1988; plate: E.R.2876) was issued around the same date as Hertel’s, but the modern Urtext edition most faithful to the Capricci’s original sources is that by Alberto Cantú and Ernst Herttrich (Munich: G. Henle, c. 1990), with extensive performance annotations by Renato de Barbieri (1929-91), a violin pupil of Guido Ferrari, Antonio Abussi and Vasa Prihoda. This edition, aiming to rectify the misinterpretations of publishers and editors spanning some 170 years, lists by caprice the numerous errors included in many previous editions, caused either by non-reference to primary sources or by simple misreading, and provides two versions of the musical text – an unannotated one which represents the composer’s autograph and one which includes Barbieri’s thoroughly practicable annotations on interpretation, fingering and bowing.

In spite of the dissemination of the various twentieth-century Urtext editions mentioned above, artistic licence in Paganini interpretation is still rife and will doubtless remain so, galvanised largely by the various reports that have come down to us of the Italian virtuoso’s performing persona. Julia Fischer, for example, has recorded Caprice no. 6 con sordino, although it is not so marked, because it ‘sounded better and more logical.’ [CD booklet, Decca 478 2274, p. 5.] Thomas Zehetmair claims that ‘in Paganini’s music there has to be something of the circus ring’, [CD booklet, ECM New Series 2124 4763318, p. 20] and his free approach to the performance of the caprices extends to adding double harmonics (nos. 9 and 13), changing pitch registers, and varying da capos, decorating them, adjusting harmonic resolutions (no. 15) and even shortening one by a few bars (no. 20).


Detailed study of the various editions and arrangements of Paganini’s caprices demonstrates the complete about turn that the editing process of printed music has undergone through history – from the autograph and initial prints through editions by celebrated violinists with backgrounds in several very different ‘schools’ of playing, to transcriptions, arrangements and complete re-compositions of individual works within the opus and finally back to the original principal sources, the so-called resulting Urtext publication underpinned by increasingly significant critical commentary and scholarly apparatus. Along the way, some editions have demonstrated performing traditions that have particular relevance to Paganini’s mature style and perhaps even reflect versions that people heard Paganini play, as Tadeusz Wronski, for example, suggests in his edition [Kraków: PWM, 1977, Preface]. Among these traditions may be examples of bariolage in Caprices nos. 2 and 12 and the inclusion of double harmonics in no. 9.

Although the fingerings and bowings annotated by eminent performers and teachers through history often have practical value within only a limited life-span – they are, after all, very personal, ephemeral responses to the musical text – one cannot overemphasise their importance as historical documents, icons of a previous era; because, as Szigeti remarks, we ‘are so poor in documentary evidence about performance traditions of a century ago … that scraps of information … or comparisons [of editions] are of great value to us and deserve the space we devote to what may seem to some minutiae’ [Szigeti, Szigeti on the Violin, pp. 202-3]. All these editions are of interest and immense value in providing us with an opportunity to recapture ways of hearing, appreciating and understanding these caprices that go back before the advent of recording technologies. For, as Szigeti maintains, one can depend little on the information from Urtexts ‘without oral tradition supplementing it’ [Ibid.].