1. Introduction

A published thematic catalogue has traditionally been seen as a prerogative of composers already firmly planted in the musical consciousness of performers and audiences. From a strictly practical perspective, it serves the twin purposes of providing the means to distinguish between two or more compositions liable to confusion and of attaching some kind of agreed label to each piece, even if this is merely a serial number. But this twofold act of identifying and naming also provides a very convenient peg on which to hang all kinds of related discourse, whether analytical, critical, biographical or bibliographical. Merely by defining the nature and extent of a composer's known oeuvre, a thematic catalogue sets a vital parameter for its critical and aesthetic reception.

Bitti, admittedly, has not quite yet reached that degree of familiarity among the modern public, or even among scholars, that would make the absence of such a catalogue intolerable. However, he is definitely already ‘on the march’ towards it, and for that reason I hope this proleptic act aimed at advancing his rediscovery will be welcomed. The major part, qualitative as well as quantitative, of Bitti's surviving music – the 27 known violin sonatas – is, as I write, in the process of appearing in a multi-volume critical edition, while modern editions already exist of all the remaining instrumental works. So the preconditions for a comprehensive revival and reappraisal are already in place, even if there has to date been no major initiative to launch it, such as the recording of a substantial group of works.

An important factor underlying the neglect of Bitti's music in our age has been the paucity of biographical information on him and the apparent uneventfulness of his career. Had he worked in glamorous Rome or Venice or led the life of a flamboyant travelling virtuoso in the manner of so many other prominent violinists of his time, including his younger Florentine contemporary Francesco Maria Veracini (1690–1768), his path towards present-day fame might have been opened by Burney or Hawkins. In fact, Hawkins makes no mention of him whatever, while Burney, almost perversely, remembers him only as one of the few Italian musicians who, alongside Giuseppe Sammartini and the two Besozzis, ‘brought the oboe and the bassoon to very great perfection’.Footnote ¹ Bitti certainly composed – and composed well – for the treble recorder and the oboe and/or transverse flute, but his employment and contemporary reputation were exclusively as a violinist. Nevertheless, the compendious nineteenth- and twentieth-century studies of the violin and its literature from Wasiliewski to Moser, Van der Straeten and latterly Apel have little or nothing to say about him; a similar silence informs William S. Newman's volume on the Baroque sonata.Footnote ² However, studies of music in Florence and especially at the Medici court have been more forthcoming. Particularly meritorious is an entry for Bitti in Warren Kirkendale's study of Florentine court musicians, which introduces a wealth of new information based on local archival records in addition to assembling facts already known.Footnote ³ As for evaluation of the music itself, there was until recently next to nothing: the dismissive sentence ‘His surviving music shows that he was a composer of modest talent and limited imagination’ closing the very short article devoted to him in the 2001 edition of the New Grove, and its paraphrase (or should one say translation?) in the latest edition of MGG, which reads ‘Die überlieferten Werke zeigen Bitti als einen Komponisten von mässiger Talent und begrenzter Vorstellungskraft’, have the look of hasty inferences rather than informed opinions.Footnote ⁴

Introduced by chance to Bitti's music a few years ago through browsing in the digital library of the SLUB in Dresden,Footnote ⁵ I was sufficiently impressed to research and write an introductory article on his career and musical style as expressed in the violin sonatas in particular.Footnote ⁶ In my missionary zeal I certainly acted too quickly, since I based this article on the six violin sonatas I had by then uncovered, whereas the current total stands at 27. In an effort to make amends for the ‘missed’ works, I wrote a follow-up article, similar in scope to the first, which was designed to serve as background information for purchasers of the volumes in the critical edition mentioned earlier.Footnote ⁷ But even this account was premature in its original form (prior to its update in April 2014), since I have subsequently unearthed three more Bitti violin sonatas among the anonymous manuscripts in the Dresden collection.

The experience of publishing two earlier articles on the same subject complicates my task of providing introductory material for the present catalogue. Obviously, simply to repeat what I have written before with the occasional update or correction would be unpardonably lazy. On the other hand, the reader needs an adequate context, biographical and musical, in which to situate the catalogue without necessarily consulting further sources. My solution is to provide slightly condensed and refocused versions of the biography and musical evaluation, taking full advantage of the opportunity to add relevant new information where available. The main weight, however, will fall on matters having a direct bearing on the thematic catalogue itself. These include the nature and status of the sources, the criteria for decisions on authenticity, the rationale for the numbering system adopted and the methodology applied in transcribing the musical incipits. The catalogue proper will consist of two components: a set of data for each individual composition and a matching series of musical incipits.

2. The sources of Bitti's music: I – Manuscripts

The primary, in almost all instances unique, sources for the 29 surviving instrumental works by Bitti that remained unpublished in his lifetime (and for a long time afterwards) are conveniently distributed among a mere three libraries, which are located in London, Dresden and Cambridge, respectively. There are, however, significant differences in the nature of these three collections. The 12 violin sonatas in Cambridge constitute a carefully organized set, whereas the seven similar sonatas in London, while very possibly forming a large portion of an intended set, cannot be so categorically defined. The less homogeneous Dresden collection, comprising seven violin sonatas, one trio sonata and one violin concerto, was acquired (by Pisendel) at a definite place (Florence) and time (1717), but is not necessarily unified in any other way.

The seven ‘London’ sonatas, notated in two-stave score in the manner normal for violin sonatas, occupy seven consecutive openings (ff. 40v–47r) of a large anthology of violin sonatas by 15 different composers that, to judge from the repertory contained, was probably compiled in England by two unidentified co-operating scribes around 1705.Footnote ⁴³ It offers a conspectus of the violin repertory by British and continental composers known and performed in England a short time after Bitti's presumed visit to London.Footnote ⁴⁴ The 80 folios comprising this manuscript are followed by a bifolio containing an engraved sonata for recorder and bass by William Williams, which was evidently added at the time of the original (or perhaps a later) binding. Since new sonatas begin without exception on verso sides in order to eliminate or at least minimize the need for page turns, it is clear that the order of the 66 sonatas it contains (ignoring the Williams addition) reflects the sequence of compilation exactly; and since the seven Bitti sonatas, numbered from XXVIII to XXXIV, form an unbroken block (the longest such block in the anthology), they were probably copied one after the other from a single source. Seeing that they are very uniform in style and musical structure and that each sonata is in a different key, with major and minor tonalities well balanced, one has the impression that they were conceived as the basis for a published set. The strong possibility of Bitti's earlier presence in London leads to a suspicion that the composer intended to have them engraved there, a project that for some reason – perhaps an unexpectedly early recall to Florence – did not materialize.Footnote ⁴⁵ Bitti's name appears at the head of each sonata in the form ‘Martino’ – a fact that until recently misled RISM (and the present writer) into identifying their composer as Giuseppe ‘St Martin’ (Sammartini).

The volume's title page (f. 1r), which may have been added at a later date, since it features the ‘mixed’ hand (fussily oscillating between secretary, humanistic and round letter forms) more characteristic of the middle and end of the century, reads: ‘Sixty Six | Solo's or Sonata's | FOR | A Violin and a Base Viol or Harpsichord. | Composed | BY | Several Eminent Masters’.Footnote ⁴⁶ There is also an index of the sonatas (headed ‘The Table’), on f. 84v.

Two former private owners of the volume are identifiable from inscriptions at the front of the volume. One is James Mathias (1709/10) a prominent merchant, amateur singer and collector of music; the other is Julian Marshall (1836–1903), who sold it to the library of the British Museum in 1880–1. The manuscript has suffered considerable deterioration over time, particularly from ink bleedthrough, and its folios were individually encased in plastic for protection in 1972.

The nine works in Dresden can most easily be introduced via a table (Table 1):

The SLUB shelfmarks in column 1 have the universal prefix ‘Mus.’. In column 2 upper case signifies major keys, lower case minor keys. The papers in column 3 are classified according to their watermarks, using the library's specially created codes. Similar codes identify the anonymous scribes in column 4. Column 5 gives transcriptions of the original titles, and column 6 any subsequent additions to them.Footnote ⁴⁷

The autograph status of the first item becomes evident from the presence of composition corrections in the score, and the initials ‘MB’ confirm the authorship. The manuscript of the trio sonata is a later copy made in Dresden, presumably from a lost score brought back by the violinist. The seven violin sonatas include four (those identified in column 4 by a code devised by the library) in the hand of copyists, who were probably employees of a commercial copisteria in Florence. In two instances, the copyist did not identify the composer, although this omission was made good by annotations (shown in column 5) supplied by Pisendel.

More problematic are the three manuscripts, two copied by Pisendel himself and one by an unidentified Italian copyist, that fail to name the composer altogether. Fortunately, the B flat sonata (Mus. 2-R-8,3) can be attributed to Bitti without hesitation, since its concluding Giga (Pisendel prefers to use the French form, ‘Gigue’) turns up in identical form to close the first ‘London’ sonata (the movement coded SV1.4 in our catalogue); in addition, its first movement has thematic affinities with SV1.1. Where sonatas and concertos in Dresden are concerned, one always has to beware of pasticcio compositions stitched together locally from movements belonging to works by different composers, but the two middle movements of this sonata appear on the basis of their style to be no less genuine.

The anonymous A major and C minor sonatas resemble each other in several ways. First, the copies – one by Pisendel and the other by a different hand – both employ a paper of Roman provenance (W-Dl-240) that appears in Dresden mainly in manuscripts of works by Roman composers such as Giuseppe Valentini, Antonio Montanari and Giovanni Mossi. However, Pisendel is likely to have travelled on from Rome to Florence with a stock of music paper awaiting use, so that is no bar to Bitti's authorship. Second, the two works are alike in taking to an extreme Bitti's predilection for polyphonic writing for his instrument: the second movements of both sonatas (and, additionally, the fourth movement of the C minor sonata) are for the most part rigorously fugal, recalling the finale of the G minor sonata (Mus. 2362-R-3) and the second movements of both the D major sonata (Mus. 2362-R-2) and the ‘other’ A major sonata (Mus. 2-R-8, 54).Footnote ⁴⁸ Third, the fugal second movements of both anonymous sonatas (and also the fourth movement of that in A major) both feature an extended coda taking the form of a kind of fantasia (or perfidia) for the violin involving the non-stop repetition, in different guises, of a brief semiquaver figure. Moreover, a very distinctive cadential figuration used in bar 54 of the second movement of the A major sonata resurfaces in bar 53 of the equivalent movement in the C minor sonata. The figuration of interlocking descending fourths used in the second section of the final movement of the A major sonata is identical with one employed in the episodes of the fugue ending the mentioned G minor sonata. As a glance at the incipits of the internal slow movements of Bitti's violin sonatas will immediately confirm, the examples in the anonymous A major and C minor sonatas display the same familiar, rather old-fashioned melodic style highly reminiscent of seventeenth-century ‘bel canto’ (and, like it, notated for preference in a 3/2 metre that sometimes employs double-length bars in order to point up the frequent hemiola effects). There are simply no dissonant structural or stylistic features, only similarities. Ultimately, it is a matter of fine judgement whether to accept works as authentic on the basis of contextual and stylistic criteria alone, but in the present instance the balance of probability is so unequivocal that I have no hesitation in admitting these two sonatas to the catalogue. In due course, the question of their authenticity will naturally need to be examined at greater length.

Although the fact that Pisendel collected these nine works in 1717 is not in doubt, how old they were by then requires case-by-case consideration. If we take the parallel instance of Vivaldi, we find that the sonatas and concertos in the composer's hand brought back by Pisendel to Dresden include not only ones written for, and performed by, the German during his sojourn in Venice but also what one might uncharitably call Vivaldi's ‘discards’ – older, no longer very serviceable works such as the quartet sonata RV 779 and the concerto ‘in due cori’ RV 585, both of which date back to the opening decade of the century. The more numerous Vivaldi works that Pisendel copied on his own initiative are mainly ones of more recent vintage, but not necessarily brand new. By that token, the autograph concerto by Bitti in Pisendel's collection could be a much older work, as its style certainly suggests, while the violin sonatas could be a mixture of older and newer pieces. The trio sonata, copied in Dresden from a lost original, could also be old.

The reason why Bitti's 12 autograph violin sonatas in Cambridge were the last to come to notice is that the library holding them, the Fitzwilliam Reference Library, is still not covered by RISM's database at the time of writing.Footnote ⁴⁹ Their autograph status is evident from a comparison of their hand with that of the Dresden concerto, and their function as a presentation copy, already deducible from their content and organization, is equally clear from the calligraphic nature of the handwriting and the extreme care taken over such notational details as bowing marks, which are always meticulously drawn from note head to note head in a strikingly modern manner quite untypical of their period.Footnote ⁵⁰

It is possible to trace the provenance of this manuscript volume right back to the beginning. Its first owner was Cardinal Pietro Ottoboni (1667–1740), the music-loving Roman prelate especially well known for his patronage of Corelli and Handel.Footnote ⁵¹ Ottoboni attracted, and doubtless often solicited, countless similar musical offerings, published or manuscript. In 1694, we have the trio sonatas of Corelli's Op. 4 and Albinoni's Op. 1; in 1701, T. A. Vitali's Concerto di sonate, Op. 4; in 1714, Michele Mascitti's Op. 5 violin sonatas; in 1715, Caldara's Op. 4 motets; in 1717, Boni's Op. 1 cello sonatas; in 1726, G. B. Somis's Op. 4 violin sonatas; in 1733, Mossi's Op. 6 violin sonatas. In addition, we know of two other post-1720 manuscript sets of violin sonatas acquired by the Cardinal: Vivaldi's 12 ‘Manchester’ sonatas of c.1726 (named after their present-day location) and a set of ten sonatas by Gaetano Maria Schiassi (1698–1754), also held by the Fitzwilliam.Footnote ⁵² After Ottoboni's death his property was sold off in order to settle his extensive debts. The English traveller Edward Holdsworth bought a huge job-lot of Ottoboni's music ‘on spec’ for his friend Charles Jennens in 1742. This passed on Jennens's death to Heneage Finch, 3rd Earl of Aylesford, and was inherited by successive Earls until, in 1918, it was sold at auction by Sotheby's.Footnote ⁵³ The Bitti sonatas, believed to be part of the 16 items making up lot 313, were bought by the collector William Barclay Squire, who sold them on to the Fitzwilliam Museum the following year.

The Cambridge manuscript comprises 32 folios of music in the customary oblong quarto format. Most of the sonatas occupy independent gatherings, as a result of which they usually end with void staves or even a void page. This was probably done both as a precaution (in a calligraphic copy one hopes to limit the amount of matter that has to be removed, should a mistake occur) and as a means of making the start and finish of the sonatas more visible. Page turns have all been made to occur at convenient places.

The date of this volume is hard to establish precisely, but its musical character offers a few clues. The frequently ornate violin line (some movements are in fact merely heavily decorated new versions of ones found in the ‘London’ group, as a study of incipits in the present catalogue will swiftly reveal), is liberally peppered with grace-notes (i.e. appoggiaturas written, with careful distinction, either as small quavers or as ‘slashed’ semiquavers) but not containing galant traits such as the ‘Lombardic’ rhythm popularized by composers of the Neapolitan school (Porpora, Leo, Vinci et al.) from the mid-1720s onwards.Footnote ⁵⁴ One important pointer to a date of composition no earlier than the 1720s is the presence of three variation movements: a gavotte-like allegro with five variations closing the ninth sonata, a minuet with four variations closing the eleventh sonata and a ‘Tempo di gavotta’ with two variations occupying third (i.e. penultimate) position in the last work. Variation movements, usually functioning as finales, first appear in the late-Baroque Italian sonata and concerto repertory during the 1720s, early instances occurring in sonatas by G. B. Somis, Carbonelli and Locatelli and in concertos by Vivaldi.Footnote ⁵⁵ Tentatively, therefore, we may date Bitti's ‘Cambridge’ sonatas to the period 1720–40, with the likelihood that they were composed towards the start of this time frame.

A further manuscript needs to be considered to conclude this section. It transmits a three-movement violin concerto in D major of no particular musical merit, vaguely Vivaldian in style and from its appearance belonging to the period 1710–30.Footnote ⁵⁶ The concerto is the eighty-third item in the large collection known as the ‘Manchester Concerto Partbooks’.Footnote ⁵⁷ This is another fragment of Cardinal Ottoboni's library whose history can be traced via Holdsworth, Jennens and the Aylesford family to the 1918 auction at Sotheby's, where the group of 13 partbooks (a binder's collection comprising in fact two discrete sets of parts) was bought by the Handel scholar Newman Flower, from whose estate Manchester Public Libraries acquired it in 1965.Footnote ⁵⁸ Each of this concerto's five parts is headed ‘Concerto del Sig:^e [or another abbreviation of ‘Signor’] C: B:’, and the Organo part carries the later, evidently explanatory, inscription ‘C. Bitti Giovani’.Footnote ⁵⁹ For some investigators, this has sufficed to attribute the concerto, as RISM does, to the otherwise unknown composer Giovanni Bitti or even to wonder whether ‘Giovani’ is perhaps in error for ‘Martino’. There are three objections to this postulated ‘Giovanni’. First, the inversion of surname and forename, introduced to Italy only under the influence of Austrian bureaucracy in the nineteenth century, is unidiomatic for the period. Second, the name contradicts the initial letter ‘C:’ of the original heading. Third, the reduction of the double consonant in ‘Giovanni’ to a single one is uncharacteristic. However, the later heading makes perfect sense if ‘Giovani’ is a misspelled ‘Giovane’, meaning ‘young’ or ‘junior’. The ‘C.’ could in that case be short for ‘Cristofano’, whereupon everything, including the general musical style, would fall into place. Cristofano Bitti, who performed with his father in Lucca in 1730, will not have migrated to Britain before that year, so may well have been active as a composer within Italy over a considerable period. The case for his authorship is far from proven, but looks promising. At any rate, this is not a composition to add to our thematic catalogue.

3. The sources of Bitti's music: II – Prints and derived copies

In January 1704 the London publisher John Walsh, in association with John Hare, embarked on a new venture that marks a watershed in the reception of Italian instrumental music in the British Isles.Footnote ⁶⁰ This was a periodical publication whereby subscribers would each month receive an Italian or Italianate trio or ‘multivoice’ sonata in parts plus a solo sonata by a different composer in score. The pagination within each category was made continuous so that purchasers could later bind the complete collection together. The solo sonata selected by Walsh for January, as partner to a trio sonata by Torelli, was the fourth of the Bitti sonatas belonging to the ‘London’ group (SV4). Walsh advertised its publication in the Post Man for 22–5 January 1704 as follows:

An example of the original edition of this sonata is preserved in Durham Cathedral Library.Footnote ⁶¹ Printed on pages numbered 1–4 (but beginning, unusually, on a verso side), the sonata has in addition to its brief title on page 1 a longer, prefatory title stamped on a separate, unpaginated folio. This reads:

This label provides the vital link to Gasparo Visconti (‘Signior Gasperini’) and through him to the ‘Signior Petto’ of the 1703 advertisement. The last line of the title refers to the common practice in Britain, particularly among amateurs, of performing solo sonatas for violin and basso continuo on the keyboard – with the option of inserting improvised ‘filling’ harmonies.

A second violin sonata in A major by Bitti (SV8) was brought out in April 1704 by Walsh, as partner to the well-known trumpet sonata in D major by Corelli.Footnote ⁶² The separate stamped label for this work, is more succinct, reading simply: ‘A SOLO in A# for a VIOLIN by Sig.^r Martino Betti | The SOLO Proper for the Harpsicord or Spinett’. Unlike the ‘January’ sonata, this one has no concordance among the seven sonatas in the British Library manuscript. That it nevertheless belongs in some way to the ‘London’ group can hardly be doubted, however: in fact, by bringing the size of the group up to eight works, it invites one to speculate whether the series SV1–8 constitutes an exact counterpart to the set of eight published recorder sonatas (SR1–8) that we shall shortly be discussing.

In 1706 Walsh and Hare reissued in a single volume six solo sonatas from the 1704 periodical publication, using the original plates.Footnote ⁶³ The new publication was advertised by Walsh on 23 November 1706 as follows:Footnote ⁶⁴

This volume presents in correct chronological order the six sonatas originally published in January–June 1706, with one important difference: Torelli's ‘December’ sonata in G minor is substituted for Bitti's ‘April’ sonata. Walsh's obvious motivation for the switch was to make the number of ‘eminent masters’ equal to the number of ‘choicest works’. If there was a complementary ‘second collection’ of solo sonatas uniting between two covers the six remaining sonatas for the year 1704, it has gone unrecorded.

The second and fourth movements of Bitti's ‘January’ sonata were much later included in John Cluer's anthology of simple pieces for treble instrument and bass Medulla Musicae (c.1727), but not under Bitti's name: the corrente-like second movement is headed ‘Torelli’, while the final movement is described as ‘Giga Torelli’.Footnote ⁶⁵ The error may have come about because the title of the trio sonata by Torelli published in January 1704 does not make it clear that the partner work is by a different composer.Footnote ⁶⁶

The Amsterdamsche Courant of 8 May 1710 carried an advertisement by the music publisher Estienne Roger for ‘Sonate a flauto traverso, haubois o violino solo van Mrs. Haim e Martinetto Bitti, 2 gl.’. The title page of this publication reads: ‘VI SONATE | DA CAMERA | a Flauto Traversa [sic], Haubois | o Violino Solo | di | Nicola Francesco Haim Romano | e M. Bitti | A Amsterdam| Aux depens d'ESTIENNE ROGER Marchand Libraire’.Footnote ⁶⁷ This is a multi-authored publication in which Sonatas I–IV are attributed to Haym and Sonata V to Bitti; Sonata VI is left without an attribution, but looks stylistically foreign to both named composers and contains hints of a possible English provenance. Sonata V, at any rate, looks perfectly genuine. True, its three-flat key signature for C minor is anomalous for Bitti, who continued to use a two-flat signature for that key until at least the time of the ‘Cambridge’ sonatas – and, as it happens, also for Haym, whose two C minor sonatas in the same collection use only two flats – but this fact alone need not arouse suspicion: after all, Pisendel's copy in Dresden of the B flat sonata SV13 has a two-flat key signature, whereas Bitti himself seemingly always employed only one flat for this key.

Since we already know of Haym's links to Alessandro Bitti and, by implication, also to Martino, it is at least a possibility to be considered that the importation of a work by the latter into this collection was at the cellist's suggestion, even though Roger, as the title page makes clear, financed the publication. Haym had by this time already published two sonata collections with Roger (Op. 1, 1703, and Op. 2, 1704), and his recommendation would certainly have carried weight with the publisher. On the other hand, the failure to name the composer of one work, more typical of a publisher-sponsored collection than of one compiled by a musician, speaks against Haym's direct involvement, as does the puzzling extra flat in the key signature of the fifth sonata.

The open-ended description of the treble instrument as ‘flauto traversa, haubois o violino solo’ is, of course, a commercial tactic aimed at maximising sales. The rationale behind this flexibility is that any single-line music confined to the register d^′–c^′′′ will be equally playable on all three instruments. This does not, of course, mean that it will be equally effective on them, and we need to try to ascertain for which particular instrument the piece was originally conceived. The violin is on balance the least probable of the three, since, as the instrument with the fewest limitations, it would also be the one most likely to exceed the capability of the two others. C minor is far from being an altogether unfavourable key for the transverse flute (one recalls Vivaldi's elaborate obbligato part for the instrument in his dark-hued C minor aria ‘Sol da te, mio dolce amore’ from his opera Orlando (RV 728) of 1727), but because of its remoteness from the basic scale of the instrument (in D major) would be a rather exceptional choice. In support of that conclusion, Lowell Lindgren has mentioned, in his edition of Sonatas I–IV, that in certain later advertisements Roger omitted mention of the flute or even (in 1712) of both the flute and the violin.Footnote ⁶⁸ This leaves oboe as the strongest candidate, making the sonata a potential counterpart of Vivaldi's weighty oboe sonata in the same key (RV 53). Assuming this inference is correct, an intriguing idea suggests itself. In June 1709 the court of Tuscany was joined by Italy's premier oboist, the Prussian-born Ludwig Erdmann (1683–1759).Footnote ⁶⁹ Could this sonata have been conceived, in fact, as a ‘welcoming piece’ for the newly arrived star performer?Footnote ⁷⁰

An interesting feature of this sonata is that its bass part goes down on two occasions to B^b₁. This rules out the normally tuned cello but is compatible with the basse de violon, the seven-stringed bass viol and the bassoon, the last-named instrument seeming the most probable option. Of course, this low note could equally well be a by-product of the mechanical downward transposition of, say, a sonata in D minor for transverse flute or even of a sonata in F minor for treble recorder, providing a possible explanation for the appearance of a three-flat key signature in place of the two flats used elsewhere by Bitti. But this takes us into the realm of inconclusive speculation.

At all events, the reason why flute preceded oboe in the list of alternative instruments was probably social or commercial rather than musical. The transverse flute, like the so-called common flute (recorder) and the violin, was a ‘gentleman's’ instrument enjoying a vogue among amateurs, whereas playing the oboe was more the business of professionals. This fact alone could account for Roger's decision to give the first instrument priority of listing. I should add here that the present catalogue respects this priority by using an ‘F’ (for flute) rather than an ‘O’ (for oboe) in the prefix to the code for this sonata – not because the former is necessarily the likelier ‘intended’ instrument, but simply in order to make a more immediate connection to the title of Roger's publication.

In the Post Man for 11–13 December 1711 Walsh and Hare advertised for the first time a set of ‘New Solo's for a Violin and Harpsichord’; Compos'd by Signior Martino Bitty’. These are the eight engraved sonatas with a title-page opening:Footnote ⁷¹

One sees immediately that this title page has been put together by someone who knows Italian (though not necessarily the composer). The omission of a comma between ‘due’ and ‘Violino’ in the first line has had odd, if predictable, consequences. It has been widely assumed that ‘Violino’ is a misprint of ‘Violini’: hence the persistence in the lexicographical tradition (starting with J. G. Walther in 1732, who wrote that Bitti had composed ‘XII Sonaten auf zwo Violinen und Baß’), of a claim that he produced a set of trio sonatas.Footnote ⁷² Odd, too, is the double-barrelled nature of the title. We start with the specification of a violin-bass combination, only to learn immediately afterwards that the instrumentation of the upper part is for flute (i.e. common flute, or recorder) or for violin. Puzzlement among his customers may have been the reason why Walsh republished the set in February 1712 with a new title page opening: ‘Solo's for a Flute, with a th[o]rough Bass for the Harpsicord or Bass Violin’.Footnote ⁷³ This change confirms what one would have suspected from the most cursory inspection of the sonatas: this is highly idiomatic recorder music, and mention of the violin is simple commercial opportunism.

How did Walsh obtain the music? It is clearly a purpose-written set, not a compilation of works obtained from different sources. Both the structure and the style of the sonatas are extremely homogeneous, and the complete avoidance of key duplication, with a perfect balance between major and minor tonalities, would not have come about without some degree of planning.Footnote ⁷⁴ The music is not noticeably more mature than the ‘London’ sonatas from the previous decade – if anything, the need to take account of the recorder's lesser agility in comparison with the violin has resulted in a plainer, almost regressive style. So this could well be music left behind by the composer and made available to Walsh by an intermediary such as Haym. We must not, however, overlook the possibility that Bitti sent over the music from Florence, using a similar person ‘on the spot’ to act as his London agent. In that case, one would need to investigate whether the initiative was his own, arising from his knowledge of the British musical scene, or whether the sonatas were in fact a commission, either by Walsh or by a patron or colleague living in Britain.

There are signs that the sonatas enjoyed some favour, for three movements are found in anthologies. The Gavotta from the first sonata turns up as an anonymous voluntary in a manuscript collection of organ music assembled by Samuel Webbe (1740–1800) much later in the century.Footnote ⁷⁵ The Minuetto ending the second sonata becomes an anonymous Minuè for mandolin, and the Gavotta ending the fourth sonata an anonymous piece for mandolin with unchanged title, in a vast anthology of English provenance containing pieces for that instrument preserved in Austin, Texas.Footnote ⁷⁶ A less direct testimony to the success of the set is the collection of eight (a coincidence?) very similarly conceived recorder sonatas, entitled Divertimenti da camera, which Giovanni Bononcini brought out in London in 1722.

To find the next published collection, we have to jump to 1723, when Roger's successor, Michel-Charles Le Cène brought out a set of 12 violin sonatas by Bitti described as ‘XII Sonate a violino solo e basso continuo’. The date can be established with some accuracy from the plate number of the edition (serving also as a catalogue number for mail orders from Le Cène's customers). Unfortunately, this edition is lost, depriving us of the opportunity to find out whether all or any of its content overlapped with that of the ‘Cambridge’ sonatas, and whether it had a dedicatee of interest.

Immediately after Le Cène's death in 1743 an inventory was taken of his stock. Among the manuscripts listed, as item 41, was a set of ‘6 Sonates a Violino Solo de Martino Bitti’.Footnote ⁷⁷ This may well have been printer's copy for a projected publication that never materialized, as we know was the case for item 9, ‘12 Trios de Albinoni pour le Violons [sic] è Basse’, which were intended to become that composer's Op. 11.Footnote ⁷⁸ Item 37 is described as: ‘6 Concerts de plusieurs Auteurs, Marchitelli, e Bitti, &c.’. This may or may not have been an anthology intended for publication one day. We must remember that music publishers of the time were equally active as music sellers and sometimes employed copyists to supply customers with handwritten copies of unpublished music. Thus even music that escaped the engraver's burin could become an object of profit to them.

There are two further compositions to consider in this section, both potentially, but far from definitely, compositions by Martino Bitti. The first is a possible contribution to the anthology of six concertos published by Estienne Roger in 1714 as:Footnote ⁷⁹

The dedicatee, Leon d'Urbino, was a Jewish banker in Amsterdam who appears to have hosted a kind of collegium musicum, from the repertory of which the six concertos in Roger's collection are possibly taken. As had happened in an earlier Roger anthology of multivoice sonatas, the 14 Sonates ou concerts à 4, 5 et 6 parties issued in two sets of partbooks in 1711–12, the composers, or some of them, are named on the title page, but do not reappear at the head of the appropriate concerto. So the identification of composers becomes a guessing game aided, of course, by the possible existence of concordant sources. In the present anthology, to which Roger late gave the catalogue number 188, the first concerto is definitely by Vivaldi (RV 276), and the last, scored for trumpet, four-part strings and continuo, most probably by Torelli. The authorship of the four central concertos is more problematic. Since we have only one authenticated Bitti concerto (CV1 in our catalogue) of uncertain date from which to gauge the composer's style, it is hard to know which features especially to look for, but contrapuntal ambition (especially in relation to inner parts) and a propensity for elaborate melodic decoration in slow movements are two. On those criteria, only the second concerto qualifies for further examination. In B flat, with a one-flat key signature, this concerto has four violin parts – a characteristically Roman feature found in concertos by Valentini and Mossi but occasionally imitated by non-Roman composers such as Torelli, Vivaldi and Leo – and four movements. The first and third movements are both composite in character, with alternating quick and slow sections, while the second movement is a very well wrought fugue and the finale a catchy, dance-like movement in binary form. This is by far the strongest composition in the collection, but while one would like it to be by Bitti, there is no clinching evidence for his authorship. Nevertheless, there is some value in placing its incipits on record, just in case an attributed concordance one day turns up. Therefore, an appendix has been specially created for it in the present catalogue, with the distinctive prefix ‘a’ attached to its code (aC4V1).

A second item to ponder is a Menuet by Sig.^r Bitti published in the twelfth (1727) and thirteenth (1728) books of Walsh and Hare's keyboard anthology The Harpsichord Master and also reprinted from the original plate in the fifth book (1735) of Walsh's anthology The Lady's Banquet.Footnote ⁸⁰ The idiom of this simple piece, which may well be a keyboard transcription of a movement from a violin sonata, is considerably more advanced than that of Martino Bitti (a series of repeated diminished sevenths during its second section suggests the rhetoric of the galant age), but could well point to Alessandro Bitti.Footnote ⁸¹ For that reason, this piece has been excluded from the catalogue.

4. The music

It is easiest to begin this brief survey of Bitti's instrumental music with a look at the more external, particularly the structural, features of his music before homing in on its expressive and technical qualities. We turn first to the two works lying outside the ‘solo’ medium: the concerto (CV1) and the trio sonata (S2V1).

In four movements arranged in the Slow-Fast-Slow-Fast configuration inherited from the late seventeenth-century sonata, the concerto conforms in its general aspects to the north Italian, or (Bolognese-Venetian) model that arose during the two last decades of that century, as exemplified in the Op. 6 (1698) of Torelli. The score, which has no headings giving the instrumentation, is written on five staves, the upper two being for a principal violin with intermittent solos and a ripieno first violin part (often called violino di rinforzo at the time) that sometimes doubles the principal instrument, sometimes ‘picks out’ its salient notes and sometimes simply drops out. The first fast movement is in binary form, the second in the ‘motto’ form much used in the early concerto, where successive periods are launched with the same head-motive presented in whatever key is appropriate at the time.Footnote ⁸² Unusual, however, for the north Italian concerto tradition is the elegance of the inner-part writing, including that for the viola, which reminds one more of the luxuriant counterpoint of P. P. Bencini and Corelli in Rome. The stringhe and incrociature of which Pistocchi complained when they were transferred to Bitti's vocal writing here find a more appropriate place. Bars 4–7 of the slow movement in the relative key of B minor, shown in Figure 1, illustrate both his delight in complex fioritura and his deft part-writing.Footnote ⁸³

Figure 1. Third movement of Bitti's Concerto in D major for violin and strings (CV1.3), bars 4–7.

The trio sonata in C major is overall the most Corelli-like work of any by Bitti, but its opening movement has at least one individual feature: its bipartite structure. Bars 1–4, Adagio, which close in the dominant, seem to herald a classic opening movement: but then Bitti abruptly changes the tempo to Allegro and uses the remaining 33 bars for a non-stop imitative dialogue between the violins. We will encounter similar ‘split’ movements again. The fugal second movement (Canzona), the short, transitional slow third movement in the relative minor key and the more homophonic, binary-form fourth movement are all excellent of their type but perhaps lack the individuality of the solo sonatas.

The structure and concept of the 36 solo sonatas is most clearly outlined if we consider each set or group in turn.

The eight recorder sonatas conform with minimum deviation to what one may call the standard model of the post-Corellian chamber sonata (sonata da camera). The first of the four movements is invariably a binary-form prelude (preludio), which with one exception (SR1) is in slow tempo. The second movement is invariably a dance: generally an allemanda in common time opening with the characteristic short upbeat, although a corrente or even a giga (SR6) may take its place. The favoured option for the third movement is another fast dance (corrente, giga or gavotta), although SR5 opts for a sarabanda in slow tempo and 3/2 metre that comes close, except for the insistence on binary form with the two sections repeated, to the kind of internal slow movement favoured in Bitti's church sonatas.Footnote ⁸⁴ The final movement is a minuet, giga or gavotta. As usual in the chamber sonata tradition (except, sometimes, where an ‘abstract’ movement takes the place of the second dance), Bitti maintains strict homotonality. This preference for an unchanged tonal centre, although not absolute in his other collections, is very marked in Bitti. It is a product of his early date of birth (and, one presumes, also of his musical education) only up to a certain point: he seems by temperament to have an urge to intensify the tonal character rather than to diversify it (in which respect Vivaldi later proves himself a kindred spirit). A study of the incipits reveals that Bitti continues the seventeenth-century habit, displayed most evidently in the variation suite, of basing all or most of the principal themes of the successive movements on a common melodic shape. In this, he is not unusual for his time, and he follows the general trend in making this feature less ubiquitous and overt in his later sonatas.

The eight ‘London’ violin sonatas are superficially similar in structure but reveal interesting differences of detail. The opening movement is not given the title ‘Preludio’, is invariably in slow tempo and with only one exception (SV4.1) is in unitary (i.e. continuously running) rather than binary form. Two of the second movements (SV6.2 and SV8.2) depart from strict da camera norms by being abstract in character, although binary form is retained. The third movements, mostly in slow sarabanda rhythm (although that dance title never actually appears), favour 3/2 metre.Footnote ⁸⁵ SV3, exceptionally, has no internal slow movement and passes directly to the giga, which is Bitti's unvaried choice for the final movement; SV4.3 and SV8.3 feature unitary in place of binary form. Homotonality once again predominates, although the third movement of SV8 moves to the relative minor key.Footnote ⁸⁶ One little touch advertises Bitti's close familiarity with vocal music and willingness to ‘take over a few tricks’ from it. The concluding Giga of SV5 opens with the so-called ‘double Devise’: a short motto (in an aria, this would be delivered by the singer) first broken off and then restated after a short interval with a normal continuation. Figure 2 shows this witty opening.Footnote ⁸⁷

Figure 2. Opening of the Giga from Bitti's ‘London’ sonata no. 5 (SV5.4).

The British Library manuscript transmitting seven of the ‘London’ sonatas more often than not omits tempo markings and/or indications of dance type. Such omissions are not uncommon for the place, period and repertory, and may even go back to the autograph originals. These were works conceived, after all, for concert and recreational use rather than for the demanding eyes of patrons or collectors.

The lone flute (or oboe) sonata SF1 exemplifies even more clearly the growing interpenetration of church and chamber elements in the solo sonata, a development actually prefigured in Corelli's own supposedly ‘ideal’ sonatas representing the two types. The first tradition is to the fore in its opening two movements, which are entirely abstract in character, and it even conditions the last two, which are respectively sarabanda-like and giga-like in their musical physiognomy, but do not advertise the fact through use of a dance title. However, the second tradition simultaneously exerts a parallel and opposing influence through the choice of binary form for each movement. The result is a sonata of mixed character, a type that will only grow stronger in the closing decades of the Baroque era. If one may generalize, the da camera elements in such a hybrid sonata tend to be strongest at the rear, the customary site of the lighter movements, while the da chiesa elements congregate at the front, where the main musical heft normally lies.

The Dresden group illustrates in its seven violin sonatas the structural variety that could be obtained within the chiesa-camera duality. At one extreme, we have the B flat sonata (SV13), which carries on the da camera design of the London sonatas, its only novelty being the choice of a movement styled as a pastorale (in the style of Valentini or of Corelli's ‘Christmas’ Concerto) to occupy second place. The other six sonatas all incline weakly or strongly towards the da chiesa model. The number of movements they contain ranges from three to six. SV9, in C major and with three movements, begins with a unitary movement in ‘motto’ form,Footnote ⁸⁸ continues with a slower, binary-form movement effectively in 9/8 metre (though notated in 3/4) and finishes with a gavotta harbouring some sly digs at the routines of academic counterpoint.Footnote ⁸⁹ The four-movement SV11, in G minor, opens with another act of homage to vocal music. As the incipit for SV11.1 shows, its first notes are those of a short ritornello for continuo, exactly as in a typical cantata aria; moreover, this ritornello is treated as a free basso ostinato, a very common procedure in cantata arias around 1700.Footnote ⁹⁰ SV11.2 is a dashing unitary movement reminiscent of the concerto genre.Footnote ⁹¹ The last two movements, a binary-form adagio and a beautifully worked ‘stretto’ fugue,Footnote ⁹² both make copious use of double stopping.

SV10, in D major, expands the da chiesa format to five movements, the ‘extra’ movement being a centrally placed bipartite movement of unusual character. Its first 22 bars are an adagio in which, over a walking quaver bass, the violin plays slow, almost inane-sounding, scales in minims and crotchets as if practising;Footnote ⁹³ then, in bar 23, the tempo suddenly changes to Allegro, and we have moto perpetuo semiquavers in perfidia style over a bass in steady crotchets. The first movement of this sonata is a composite adagio-allegro structure following the familiar pattern of the opening of Corelli's first violin sonata in Op. 6, which may indeed have served Bitti directly as a model.Footnote ⁹⁴ As its second movement, the sonata has a vigorous ‘double-stopped’ fugue (Canzona) featuring some very adventurous episodic writing.Footnote ⁹⁵ The final two movements are a more conventional, but no less cogent, Adagio in 3/4 and an Allegro in 12/8.

In the A major sonata SV 12 the movements expand to six: Bitti has effectively inserted an extra Adagio-Allegro pair in the centre of the sonata. The first Allegro is another ‘double-stopped’ fugue;Footnote ⁹⁶ this follows Corellian precedent in turning one episode into a ‘parenthesis’ in which the thematic material is deliberately kept neutral in character, unrelated to the main musical argument, thereby allowing the player to engage in uninhibited technical display. The central Adagio, in F sharp minor, follows Corellian precedent, as so many slow movements by Bitti do, in closing its first phrase with a rhetorical Phrygian cadence, the semantic equivalent of a question, but immediately repeating the same phrase in a different key (here, the relative major), with the effect of sidestepping rather than answering the question. Figure 3 shows this opening.

Figure 3. Opening of the third movement of Bitti's fourth ‘Dresden’ sonata (SV12.3).

The following Allegro is, so to speak, a homophonic counterpart, dominated by rapid passage-work, of the fugal second movement. Once again, the two final movements are in a more standard mould.

SV14, again in A major, and SV15, in C minor, can conveniently be considered together, since they share many features in addition to the fact of being preserved anonymously. They are alike in being ‘pure’, four-movement church sonatas, with no movements in binary form. Similarly, their second movements run along parallel lines by being bipartite: opening with a strenuous ‘double-stopped’ fugue and continuing half-way through the movement with the by now familiar perfidia-like appendix in continuous semiquavers for the violin – marked ‘Adagio’ in the case of SV14.2, but ‘Presto’ in that of SV15.2. The internal slow movements of the two sonatas, which as usual revert to single-line writing for the violin, are Bitti's supreme essays in the ‘bel canto’ style. The opening of SV14.3 is quoted as Figure 4.

Figure 4. Opening of the third movement of Bitti's sixth ‘Dresden’ sonata (SV14.3).

Only in their final movement do the two sonatas part ways. Here, SV14 has another bipartite movement, opening with a short section, with abundant chordal writing, in the rhythm of a French courante (and making much use of hemiola); this is succeeded a longer section featuring intensive contrapuntal treatment (as in a Bach two-part invention) of a motive introduced as a countersubject on the violin to the opening notes of the ‘courante’ in the bass. This section includes a short passage where the continuo divides, the figurational writing presumably passing to the cello (or perhaps lute) and the slower-moving bass notes to the keyboard instrument. SV15 simply has a second ‘double-stopped’ fugue in a tripla-like rhythm (6/8); this is a ‘stretto’ fugue very similar in character to SV11.4.Footnote ⁹⁷

The 12 ‘Cambridge’ sonatas were certainly written in the knowledge of Ottoboni's musical taste with regard to sonatas, which from Corelli (1694) to Mossi (1733) unswervingly favoured the chamber variety. However, Bitti has moved on from the ‘classic’, Corellian model followed faithfully by the recorder sonatas and only a little less so by the ‘London’ violin sonatas. The number of movements is more variable; the incidence of ‘abstract’ movements is higher, particularly as regards those in quicker tempo; ‘short’ metre (2/4) appears in three second movements (SV16.2, 19.2 and 26.3), something unthinkable two decades previously; the number of movements varies between four and five; slightly more internal movements move to a different tonality (the choice of the parallel key of C minor for the siciliana-like second movement of SV24 is very up-to-date); the violin line is chock-full of appoggiaturas and written-out embellishments; as already mentioned, there are three variation movements.

Despite all this novelty, the fact is that several movements and even one complete sonata are refurbished and updated versions of music already present in the ‘London’ or ‘Dresden’ violin sonatas. The modifications are made sensitively, and there is no reason to assume that Bitti reached back into his old stock merely to save time or to avoid dependence on a failing imagination. The third ‘Cambridge’ sonata (SV18) is an elaboration in all four movements of the first ‘London’ one (SV1), even though this is not immediately evident from the incipits of the two gighe. The first three movements of the second ‘Cambridge’ sonata (SV17) are related in identical manner to those of the seventh ‘London’ sonata (SV7); the first and fourth movements of the twelfth ‘Cambridge’ sonata (SV27) to the second and fourth movements of the fourth ‘London’ sonata (SV4); the second movement of the fifth ‘Cambridge’ sonata (SV20) to that of the sixth ‘London’ sonata (SV6), with modification of the original binary form to ‘motto’ form. Finally, the opening theme of the first movement of the fourth ‘Cambridge’ sonata (SV19) is a major-mode paraphrase of that opening the third movement of the third ‘Dresden’ sonata (SV11). Like Vivaldi, Bach and Handel, Bitti was evidently in the habit of periodically revisiting his older works and movements in order to improve, customize or modernize them.

To illustrate in a comprehensive manner the still virtually unknown music of the ‘Cambridge’ sonatas is an impossibility, but, as a sample, the first five bars of the Minuet and its four variations closing the eleventh sonata (SV26) are given as Figure 5. Since the bass is identical for each variation – very often, in contemporary sources of variation movements (though not here), it is written out only once – the example ‘stacks’ the theme and each variation in turn over a common bass.

Figure 5. Opening of the Minuet and its four variations in the eleventh ‘Cambridge’ sonata (SV26).

Turning now to more general aspects of Bitti's musical style, the first thing to strike the listener or student after the deliciously long-breathed, sculpted melodic lines is the skilful interchange between parallel motion and contrary motion in the writing of the outer parts (which are, of course, generally the only notated parts in the solo sonatas). Bitti is as fond of the euphony of long chains of parallel tenths as he is of taking the bass – contrarily in more senses than one – in an opposite direction to sequential motion in the upper part or parts. Except where this is required by the procedures of fugue or ‘motto’ form, the literal restatement of material (and therefore thematic ‘rounding’) occurs remarkably sparingly in his music, even in pieces as late as the ‘Cambridge’ sonatas of the 1720s: not for him the blatant ‘cut-and-paste’ practices of Valentini or the more moderate recourse to tonic reprises and/or matching section openings and closes in the later Corelli. This quality of continuous evolution, which even stretches unimpeded over the repeat sign separating the two sections of a binary-form movement, is a trait inherited from the seventeenth century, and one evidently too precious to relinquish. The phrase structure varies with delightful unpredictability between symmetry and asymmetry, as does also the ratio between the lengths of the two sections in binary form.

Bitti's music is predominantly diatonic: his only known composition in which chromatic colouring plays a very important part is the C minor ‘Dresden’ sonata (SV15). On the other hand, his music often displays quasi-modal colouring, either ‘Lydian’ (with raised fourth scale degree) or ‘Mixolydian’ (with lowered seventh scale degree).Footnote ⁹⁸ He is restricted for his time in his choice of base tonalities. One is not surprise to see his recorder sonatas confined within the range C minor – G major, since this arises from the limitations of the wind instrument itself, but the narrowness of the range encountered in his violin sonatas, which runs from B flat major and C minor on the flat side to A major and F sharp minor on the sharp side, is a surprise: Corelli certainly ventured further in both directions, and Valentini further still.

There is nothing of a technical nature that surprises in Bitti's wind sonatas, which mostly seem to be designed as much for amateur as for professional performance. His writing for violin, however, contains much of interest. In comparison with Corelli, Albinoni or Vivaldi, he employs much less cross-string, arpeggiated writing, even in the ‘Cambridge’ sonatas: his natural comparators here are composers of an earlier generation such as Giovanni Battista Vitali, for whom smooth, linear movement is much preferred for the faster passage-work. However, Bitti is very advanced for his time in cultivating cantabile writing on the violin in altissimo (see the openings of the first and third variations in Figure 5). The complexity of his polyphonic writing for the violin in not merely two but even three parts, often with additional, percussive notes on the fourth string, is scarcely matched in Italian violin literature until we reach Tartini and Zuccari.Footnote ⁹⁹ An interesting and rather original technique seen in Bitti's violin writing is one combining self-imitation and leapfrogging. To explain: ‘self-imitation’, a useful term introduced by Eleanor Selfridge-Field, is the simulation by a single instrument, via abrupt changes of register, of the imitative interplay of two or more instruments; ‘leapfrogging’ is where two instruments (or voices) continually exchange relative position as they rise or fall together.Footnote ¹⁰⁰ This combination of devices is illustrated by Figure 6.Footnote ¹⁰¹

Figure 6. Bars 21–25 of the violin part in the fourth movement of the seventh ‘Dresden’ sonata (SV15).

The final feature of Bitti's violin writing to be mentioned here is startlingly modern. On the evidence of both the concerto score and the volume containing the ‘Cambridge’ sonatas, both of them autograph sources and therefore authoritative for notational details, it is safe to conclude that Bitti was firmly, even fanatically, wedded to the so-called rule of down-bow: the mandatory selection of a down-bow for accented notes, particularly at the start of a bar. Not only that: he seems to have disdained the common contemporary practice of using the craquer stroke (a repeated and independently articulated down-bow, as described by Georg Muffat) as a useful expedient to obtain a down-bow for the desired note. Instead, he makes sure that within practically every bar there is an even number of bow-strokes, even when this sometimes creates unusual or slightly inconsistent slurring patterns (the pervasiveness of this procedure can be confirmed in an instant by studying the incipits of the relevant works in the present catalogue).

5. The catalogue: I – Principles and procedures

There can be few thematic catalogues of a composer's musical works in existence that do not cause irritation or puzzlement among at least some users. The present catalogue does not claim to have escaped altogether from this unfortunate condition, but by following a few simple principles, enumerated below, it hopes to have avoided the worst failings:

1. There is no universal formula for a successful catalogue. Each catalogue has to take account of the exact, sometimes unique, nature of the materials it organizes and orders.

2. A catalogue has to prepare in advance for the addition of works omitted by accident or discovered subsequently. As far as possible, such additions should be assimilated smoothly into the general plan of the catalogue.

3. A catalogue should be a handy tool for statistical analysis and any other kind of overview, which implies that works of a similar kind should be grouped together as a single block rather than assigned to different places in the catalogue.

4. A catalogue should respect as far as possible the grouping of compositions in the most authoritative primary sources or repertories.

5. Ordering compositions in the first instance by date of composition or by key is rarely recommendable, since it is incompatible with desiderata 2–4. (In a few instances, however, either of these criteria may be found useful in second-level ordering.)

6. The catalogue numbers (represented in the present instance by a prefix defining the genre, followed by a serial number defining the individual work and lastly, if needed, a serial number defining the movement) should be simple and intuitive enough to memorize easily and quote in either literal or expanded form.

7. The binary division into ‘definitely genuine’ and ‘misattributed and definitely spurious’ works needs to be maintained, but an additional, distinct category is needed for works of uncertain authorship – candidates, in other words, for either status. The assignment of a work to this third, intermediate category can be particularly useful, since it avoids giving an unnecessary hostage to fortune and may in fact encourage research leading to a firm conclusion.

For the cataloguer, Bitti is a dream of simplicity. Nearly all the extant works are preserved in unique sources, and where they are not, the priority of one source is evident. The only possible exception to this is the fourth ‘London’ sonata (SV4), which exists both as a print and as a member of the manuscript London group. But in practice, this ‘dual preservation’ causes no methodological problem whatever. The genres and subgenres represented are remarkably few: we have one violin concerto (plus a four-violin concerto of uncertain authorship), one trio sonata, one flute (or oboe) sonata, eight recorder sonatas and 27 violin sonatas. The recorder sonatas form a unified series of eight works ready-made to become a single block in the catalogue. The violin sonatas form three such natural blocks (‘London’; ‘Dresden’; ‘Cambridge’), each distinct in chronology, history and present-day location.

In the present catalogue, each subgenre has its distinct prefix, as follows:

CV = Concertos for violin with strings and continuo.

SF = Sonatas for flute (transverse) and continuo.

SR = Sonatas for recorder (treble) and continuo.

SV = Sonatas for violin and continuo.

S2 V = Sonatas for two violins and continuo.

aC4 V = Concertos of uncertain authenticity for four violins with strings and continuo.

Additional works known to have once existed from catalogues, inventories or written report are not listed in the catalogue, although they are discussed in earlier sections. The scope of the catalogue for receiving additions is practically limitless. Should, for example, the 12 sonatas published by Le Cène in 1723 suddenly come to light, it will be easy to add them (or whichever sonatas they contain that are not already known) as a continuation beyond the number SV27. In other words, the sequence of additions will from now on follow the sequence of discovery, reproducing where appropriate and possible the order in which the works occur in the newly discovered sources. Were these Le Cène sonatas to be discovered and added, the neat chronological arc described by the triad London–Dresden–Cambridge would quite possibly be compromised, but that would be a very small price to pay: the coherence of the ‘SV’ series would at all events remain intact. There is of course a continued usefulness in certain situations in referring to individual violin sonatas by the number defining them within their individual set or group (which also happens to correspond to the numberings employed in the critical edition in progress from Edition HH) rather than by their overall serial number (where different), and for that reason ‘London’, ‘Dresden’ and ‘Cambridge’ numbers will continue to be quoted in the catalogue entries.

There is also a possibility that works in new, hitherto unsuspected genres will be discovered. For these, it will be easy to invent genre-specific prefixes and serial numbers by analogy. Cello sonatas, for instance, could have the prefix ‘SC’, and oboe concertos the prefix ‘CO’.

As the reader will already be aware, individual movements within each work are identified by inserting a serial number after a dot. In theory, there could be difficulty in defining what is a movement and what is a section within a longer, composite movement (eighteenth-century notation does not recognize any distinction between thin-thin and thin-thick double barlines), but in practice this problem has not yet occurred for Bitti.

6. The catalogue: II—Entries for the individual works

Since this is a catalogue designed primarily for the purpose of identifying and locating works rather than recording bibliographical minutiae, which are in almost every instance easily ascertainable via open access online catalogues, RISM or published reference works, priority is given in it to the extra data most relevant to the musical evaluation of the works.

Each entry is based on a standard template:

Any field or element thereof that does not apply to the given work or source is simply omitted. Where there are multiple original sources of a work, information on the principal source appears under (a)–(d), and that for other sources is given in as full detail as needed under (f). In (a), the short title is given in normalized form, while the individual title or heading in (b) appears in a semi-diplomatic transcription respecting the original orthography but not showing line-breaks or different lettering styles. ‘Source type’ describes the nature of the source (as score or parts, etc.). Where a work is preserved in separate parts, its title is given in the most complete form found there. ‘Normalized description’ is a phrase such as ‘Recorder sonata no. 5’ or ‘“Cambridge” sonata no. 2’ placing a work within its set or group. Under (c), library locations are given via RISM sigla, which if not recognized immediately can be interpreted with the help of the prefatory pages of the New Grove (either edition) or via the website https://opac.rism.

In (d), keys are given in the usual way, with upper case for major and lower case for minor, italicized in order to avoid possible confusion with the common time (‘C’) signature. ‘Title’ is an expression such as ‘Canzona’ or ‘Sarabanda’, for which italics are used. Tempo directions are given in normal spelling, with abbreviations resolved. Where tempo changes in the course of a movement (ignoring the occasional tempo change at the very end of a movement signifying ‘ritenuto’ or ‘rallentando’), the tempo sequence is shown with horizontal arrows (e.g., Allegro → Adagio → Allegro). The metre is given exactly as it appears in the time signature, and any complicating factors such as double-length bars are explained under (f). Common time is represented as ‘C’, and cut time as ‘C*’. It should be pointed out that the cut time signatures appearing in several second movements of the manuscript ‘London’ sonatas are probably in replacement of original common time signatures: many northern Europeans were in the habit of using a ‘barred C’ for common and cut time alike. No modification has been made here, however, since the anomaly is innocuous. The initial bar count applies to the whole movement, but in cases where the movement is in binary form with two repeated sections, the respective section lengths are shown also in the form ‘(10:20:)’; in bipartite or composite movements, the lengths of the separate sections are shown in similar fashion by using arrows corresponding to those of the entry for tempo ‘(5 → 20 → 15)’.

In (e), the modern editions listed are limited to critical editions, facsimile editions and selected performing editions of complete works. Section (f) is a catch-all field for information on additional sources or concordances and further explanations or comments of every kind.