Editing Weiss for theSämtliche Werke: The composer's contribution to the London and Dresden manuscripts

Tim Crawford


S.L. Weiss Congress, Freiburg, September 1992


This version is still under construction and lacks illustrations - please forgive!


Back to Tim Crawford's research
Historically, the publication of a scholarly `Complete Works' edition of a composer's music is the initial goal of musical scholarship. Musicians and scholars alike need a standard text on which to base, on the one hand, their selection of music for performance, and, on the other, their assessment of common features and differences within the corpus which will enable them to discuss the composer's style and its development.

If this Congress has achieved nothing else, it has shown -- especially through performances of his music -- that Silvius Leopold Weiss was a sufficiently important figure in his own time, with a significant corpus of good music to his name surviving to ours (around 600 pieces at the latest count), amply to justify its publication in a complete edition of this sort.

In the case of a lutenist-composer like Weiss, there is another motive for publication. Before this project began, with the exception of selections published by Hans Neeman, Ruggiero Chiesa, R. Manabe and others,[1] Weiss's music had not appeared in a form that is intelligible to non-lutenists. Quite apart from the arcane nature of lute tablature, by which the notation only really means something when `realised' in sound, this form of notation is extremely inconvenient for comparative study, especially with music not written in tablature (for example, the autographs of Bach's lute music). Although many players, and possibly even a few non-lutenists with much experience of tablature, can recognise melodies or harmonic progressions at a fairly primitive level in tablature, the large-scale investigation of relationships between pieces in different keys or between Weiss's music and that of other composers is made very much harder when dealing with two completely different notation systems. When the Weiss edition is complete, that, at least, will be made much easier, and it will allow Weiss truly to enter the world of musical scholarship for the first time.

A complete Weiss edition was Hans Neeman's dream before World War II; sadly, it was put to an end by Neeman's death in military service in 1942 or 1943.[2] Neeman's work on Weiss sources was continued by Hans Radke and Josef Klima.[3] Everyone interested in Weiss's music owes these pioneers an enormous debt. In 1979 appeared a reduced facsimile of the solo music volumes of the Dresden source with an important introductory study by Wolfgang Reich.[4] Douglas Alton Smith started the present Weiss edition some twelve years ago, and laid the basic ground-rules for the current work,[5] which I took over in 1989. Four volumes edited by Dr Smith and published by Peters Edition, Frankfurt, comprising the facsimile and transcription of the London MS, have appeared so far,[6] and work on the next four, the Dresden MS, is far advanced, although there have been significant delays for a number of reasons. Continuing financial support has been provided by the Musikgeschichtliche Kommission e.V., and Prof. Georg von Dadelsen and Dr Thomas Kohlhase of Das Erbe deutscher Musik have supplied essential advice and practical help throughout the project. Future volumes will appear in the main series of Das Erbe deutscher Musik, to be published by Bärenreiter, Kassel.[7]

There will inevitably be objections from some quarters to the methodology of the Weiss edition. It does take a particular line on two aspects which are bound to be controversial. The first is the perenially contentious matter of the style of transcription, which Dr Smith has discussed elsewhere.[8] With very few reservations I find myself in complete agreement with Dr Smith, and have attempted to follow his method in my own transcriptions of the Dresden manuscript. I do not wish to pursue this here except to stress yet again the point made by Smith and others that it is essentially impossible to be objectively `correct' in a transcription, which without exception can only represent an individual transcriber's interpretation of the tablature.

Secondly, this edition makes no attempt to establish a canon of works based on chronology. Other typical canonical arrangements for scholarly collected-works editions, by genre (masses, motets, symphonies, sonatas, etc.), by instrumentation (works for one instrument, duos, trios, etc) or by function (church cantatas arranged by liturgical cycle, etc) are not applicable. Weiss's music is all in a single genre: music for solo lute. (There is much chamber music, but its incomplete state means that it must always remain a mere appendix to the solo music until significant new discoveries are made.) This has come down to us in the main in the form of suites or partitas (Weiss's term was Suonata) or as individual dance movements almost certainly `orphaned' from these.[9] There is also a fair number of isolated abstract preludes, fantasias and fugues which may or may not have been conceived as members of Suonaten.

Only a very small proportion of Weiss's music can be dated with certainty, so the decision was taken at an early stage to reproduce the contents of the various sources, beginning with the large manuscript in the British Library (London), continuing with the source of similar size in the Sächsiches Landesbibliothek (Dresden) (omitting from the transcription those pieces that had already appeared in London), and then to present the works from lesser miscellaneous sources (to appear in a pair of volumes entitled `Varia'), including the substantial quantity of dubiously-ascribed music and the pieces probably by Weiss but without his name in the source.[10]

When the edition is complete, or perhaps in parallel with the production of the `Varia' volumes, it is essential that work begins on a formal catalogue of Weiss's music, in which matters of sources, chronology and authenticity can be addressed in detail.

* * * * *

The remainder of this paper focusses on the two sources that are represented in the current state of the Weiss edition: London, which has now appeared complete in volumes 1-4 of the Sämtliche Werke, and Dresden, on which work is currently in progress.

There are several ways in which these two central sources differ. Although both are clearly intended as monumental collections of the composer's works, and Weiss himself to some extent participated in the compilation of each, it becomes clear that they were actually assembled in very different ways, and Weiss's role was essentially different in the two cases. The most crucial difference between the sources is in fact of the simplest and most obvious external nature: London consists of a single volume, while Dresden is a group of six uniformly-bound volumes.

London

The London manuscript, British Library, Department of Manuscripts Additional MS 30387,[11] was acquired by the British Museum Library (now the British Library) at an auction in London in 1877.[12] Until 1957 it was bound in its original white vellum boards, decorated with a coat of arms. Fortunately, when it was rebound, the coat of arms itself was preserved, although somewhat faded, inside the front cover of the new binding. Obviously, the arms could offer a significant heraldic clue to the original ownership of the manuscript; I shall touch on this matter again below.

Many pieces in London are annotated at the end of the music with the composer's name[13] and frequently with a date. These rubrics have often been added to music already copied into the MS. None of them with one exception is in Weiss's own hand, although in some cases they have been added to autograph pieces. The exception is the note `Veritable original S.L. Weis', which the composer himself added to the margin of f. 78v, against the allemande of Sonata 17 in C major, which begins in the composer's autograph. The strong implication is that the dates that do not actually form part of a piece's title are dates of composition supplied by the composer to the owner of the manuscript some time after copying, presumably at the request of the owner. Lacking any contrary evidence, therefore, we can assume that these dates are reliable.

There are sufficient references to Prague in the MS to suggest that it was entirely compiled in that city. This conjecture is further supported by the presence of the Tombeaux for Count Losy (1721), Prague's foremost aristocratic lutenist of the generation before Weiss, (ff. 150v-151) and `M. Cajetan Baron d'Hartig' (d. 25 March 1719), presumably a member (as yet unidentified) of a powerful family which was prominent in Prague's musical life throughout the 18th century (ff. 88v-89). `Freiherr von Hartig' was the first patron and director of the Prague Musical Academy, which gave regular concerts every Thursday. The Academy was modelled on an earlier similar foundation in Breslau (Wroclaw) and commenced activities soon after a petition from four Prague musicians was lodged with the city authorities in 1713. Hartig acted as the Academy's patron and protector at least during 1715-7.[14]

André Burguete has made the suggestion that London was compiled by or for a member of the Prague Academy, and this seems entirely plausible, although definite proof is still lacking.

One promising line of investigation that at present remains somewhat conjectural could perhaps settle the matter. Although, as far as I am aware, the coat of arms from the original binding of London has not been examined by an expert in German and Bohemian heraldry of the 18th century,[15] my own researches suggest that one feature, an eagle-like bird with raised wings, somewhat different from the Imperial Eagle, may be what one description of the heraldry of the Hartig family calls ein schwarzer Scherbvogel.[16] If this is so, the the manuscript certainly belonged to a member of the Hartig family, which would be a most satisfactory conclusion. But further expert research is needed here.

An important feature of the copying history of the source is that the manuscript as it exists today was not copied in one straightforward sequence from beginning to end. It must have been compiled in two main separate phases, with a few additions made at a later date. In the first phase, a copyist or the composer entered complete sonatas; later, in the second phase, `extra' movements were added in spaces left in the original copying.[17] Other additional movements, sometimes in hands not associated with the two principal phases, were added later in a third phase, and these pieces (with the exception of the Menuet on f. 121v, confirmed as authentic from concordances in Dresden and WRu) seem to me to be in a style inconsistent with that of Weiss, although they have been accepted as his by previous commentators and editors.[18]

Frequently the `extra' pieces that were added in phase 2 are in fact preludes to the sonatas:[19] in some cases, spaces seem to have been deliberately left for preludes in the first phase of copying; elsewhere the preludes are squeezed in wherever they will fit, even after the first proper movement of a sonata in one case (f. 142r). At other times, the preludes were copied in the first phase. Assuming that if a prelude was copied in the this first phase it can be regarded as truly `integral' to the sonata, only 8 out of 28 solo sonatas have preludes thus conceived by Weiss as part of the complete design (about 29%).

Weiss's contribution to the London MS

Several pieces were actually copied by the composer himself -- his hand can easily be recognised from the few surviving autograph letters,[20] from the Stammbuch to which he contributed in 1742,[21] and from the fact that he signed his autograph copy of Sonata 17 as a `veritable original'. As well as copying complete pieces, he also made many corrections, additions and improvements to music already copied by others. Copious fingerings and explications indicate that the manuscript was, like so many other lute books, used as teaching material. It is important to understand that these interventions by Weiss took place in both of the principal copying phases. A very unusual and bizarre feature of Weiss's participation is that often a work begun in the composer's hand continues on the next page in another hand, and vice versa. Closer examination of the MS reveals that in fact these changes in hands, which often occur within the progress of a piece,[22] are always associated with a change in paper, and in the size of the rastrum used for ruling the tablature staves. It seems likely that at some time after March 1719 (the date of the Hartig Tombeau copied on the earlier paper) some catastrophe occurred to the manuscript which meant that several complete Bogen or bifolia were lost -- I presume that it was still in an unbound state at that time. In a later visit to Prague, according to this theory, the composer made good this damage by supplying the lost music, and took the opportunity of amending some of the music already entered in the first copying phase.

Examples 1a-f, showing folios 8 to 10 of the manuscript, neatly illustrate this situation. The change of hand between Weiss (1a and 1b: f. 8, recto and verso), another copyist (1c and 1d: f. 9, recto and verso) and Weiss again (1e and 1f: f. 10, recto and verso), is obvious; not so clear in these reductions is the change of rastral size (corresponding to paper type) from 14.5 mm to 15 mm. Note that the name `Weis' at the bottom of the autograph page f. 8r is not written by the composer, nor are the (somewhat haphazard) numberings of the movements or the page numbers. At the end of f. 9v, our anonymous copyist has only completed two of the three beats of bar 18 of the Sarabande, and Weiss continues the piece with the third beat at the top of f. 10r.

Final proof that f. 9 was a substitute for the original leaf copied by Weiss comes in the single bar of tablature written by Weiss on the line below the end of the Menuet on f. 10r and marked with a cross (not the two-bar alternative ending for the Menuet's first strain which he marked with `NB'). This bar is identical with bar 19 of the Bourée on the facing page, f. 9v, which Weiss must have inadvertently omitted in his original holograph copy. This also further supports the view that the manuscript was damaged with the loss of several pages and repaired by the recopying of leaves such as f. 9 rather than having been written by a team of copyists, including the composer, working to a complex plan of predetermined page-breaks.[23]

The Giga (beginning on f. 10v, see Example 1f) offers a good example of another significant type of alteration characteristic of this manuscript. Weiss originally wrote this piece for 11-course lute: at bar 40 (bottom line of f. 10v) he uses `e' on the 7th course to obtain a B natural in the bass; five times elsewhere in the piece, three of them on this page (bars 8, 9 and 23), this note has been erased and a `5' indicating the use of the 12th course has been added, and not by Weiss himself. Such alterations are extremely common in the MS, and there is also good reason to suggest that many other pieces apparently for 13-course lute were in fact composed for the 11-course instrument. The matter, however, cannot be pursued further here, except to remind players that the existence of a work in a version for 13-course lute does not necessarily mean that it was composed in this form. More work needs to be done on such matters.

That the `repairs' to the catastrophically-damaged manuscript were done while Weiss was on a visit to Prague is suggested by the fact that the only piece in the manuscript that remains incomplete is from an ensemble work. The final Allegro (f. 29v) of the lute part of the B flat Concert d'un Luth et d'une Flute traversiere (beginning on f. 25v) breaks off at the the end of the page after 43 bars.[24] On the present f. 30 begins a sonata in C minor. Weiss himself annotated the copy of this piece in Dresden as having been composed at Düsseldorf in 1706; the London copy has some amended passages in his hand (ff. 31r and 32r). I would like to suggest that if the `repairs' to the damaged MS had been done in Dresden, Weiss could have used or supplied his own copy of the chamber work for the recopying. No doubt, when Weiss travelled to Prague, he took with him a portfolio of his solo compositions, but it is reasonable to suppose that he would have left the much bulkier chamber and ensemble works at home in the Saxon capital. Hence, the tablature of the concerto for flute and lute remains incomplete in the MS.

Dresden

The six volumes Sächsisches Landesbibliothek, MS. Mus. 2841-V-1 (Dresden) were purchased by the library in 1929 at the sale of the enormous music collection assembled by Werner von Wolffheim. [25] A number of other lute manuscripts written out by the principal copyist of Dresden was dispersed to other libraries at the same sale. Nothing is known of the history of the manuscripts before they came into Wolffheim's possession at some time in the late 19th or early 20th century, but it is almost certain that he acquired the complete series of lute manuscripts from a single source. In contrast with London, Dresden is in fact a collection of sonata-fascicles, ordered systematically into five volumes of solo lute music by key and a sixth by genre (volume VI consists of ensemble-music parts only). With a single exception (discussed below) each sonata is written on a separate fascicle from its neighbours; the copying was almost certainly completed before binding. The sonatas are further sorted within each resulting volume by size or complexity, and hence -- as it turns out -- more-or-less by date, though whether this chronological order is either intentional or even entirely reliable is not certain.

As Wolfgang Reich pointed out in his study for the 1979 facsimile edition,[26] there are three scribes at work in volumes I-V: the composer himself; the compiler and principal copyist of the collection (whose style of writing seems to have evolved over an extended period); and a third copyist, less reliable and possibly not so familiar with tablature notation. The accompanying parts for two violins, violoncello and `Contra Bass' to the only complete ensemble work in the collection, the lute concerto in C minor by Johann Sigismund Weiss, inserted loosely in volume VI, are written by a fourth copyist.

Weiss's contribution to the Dresden MS

There are six sonatas in Dresden in Weiss's autograph. Two of these (D5 in D minor and D9 in C major) are clearly intended as straightforward sonatas for inexperienced players. Indeed, according to Hans Neeman's study of Dresden, published in 1927 before the Wolffheim collection had been dispersed at auction,[27] the D minor sonata (D5) formerly bore a note in red ink: `Diese Partie habe ich zu allererst bey Ms. Weisen gelernet', confirming its use as teaching material by the composer.[28]

The second of the three autograph sonatas in C major (D9, D10 and D12) is somewhat more substantial, but D12 is of entirely another character: this is a large-scale and virtuosic sonata carrying the dedication: `P[our] S[on] A[ltesse] S[érénissime] M[onsieur]. L[e]. D[uc]. De Lobkowitz' (or, alternatively, `... L[a] D[uchesse] De Lobkowitz'),[29] among Weiss's finest pieces, although not in the style of his very latest works.

The fully mature style is represented most strikingly by the autograph of the superb D minor sonata (D7, p.41)[30] which must have been composed as late as any other works we have; the handwriting seems consistent with that in the Schmid Stammbuch (1742),[31] and may even betray the signs of a slight hand-tremor, although this is by no means conclusive.

The remaining autograph sonata, D30 on p. 245, is different from the others. The size and type of the paper, the way it was originally folded, and the fact that the three folios are numbered with Roman numerals in the middle of the bottom of each page suggest that it originally came from a quite different source than the rest of these autographs. It seems to have been copied rather late, although the music must have been composed around 1725, and copious fingerings in the copy suggest that it was prepared by the composer for a pupil.

As well as copying some pieces that found their way (perhaps at different times) into the collection, Weiss also authenticated four sonatas. Three of these are definitely early: Düsseldorf 1706 (D31, p.261: Weiss's note reads `Von anno 6, In Düsseldorf. ergo Nostra giuventù comparisce'.); Rome 1708-14 (D18, p.144: headed in red ink, not in the composer's hand, `Suonata del Sig. Weiss composta a Roma' and signed in pencil `[in Weiss's hand:] Weiss [in another, later hand:] Sil: Leop:'); and another from the same period (D17, p.135; it carries the title `Suonata del Sigre Sigism. Weiss', but the name `Sigism.' is crossed out, and `S.L. Weiss' written by the composer in pencil. Lute compositions by Weiss's brother, Johann Sigismund Weiss (c.1690-1737) are sometimes ascribed to Sylvius Leopold in other sources.) The fourth authenticated sonata (D19, p.151: signed in pencil `S.L. Weiss') probably dates from before about 1725, since it also appears in London, though the London version may in fact represent a later revision, and the Dresden copy may be of a somewhat earlier version.

Authentification of a negative sort has been carried out in a sonata written out by the third copyist. A single piece, the Menuet in a sonata in B flat (D24, p. 208) has been cancelled with a note probably written by the compiler: `ist nicht von Mr Weis'; presumably at the same time the compiler copied a substitute Menuet, notably shorter than the rejected one, onto the spare staves of the page. It seems likely that this adjustment was made under the supervision of the composer.

In contrast to his contribution to London, Weiss did not make corrections, additions or improvements to the music, even to the eight sonatas written out by the unreliable third copyist, which are full of careless errors, often rendering the pieces hard to interpret. Fortunately, most of these pieces also exist in other copies. Presumably Weiss did not have the collection in his hands for the extended period necessary to `proof-read' the music thoroughly.

From the appearance of the signatures and other marginal comments, it would seem apparent that the annotations were carried out late in the composer's life. Some bear a marked resemblance with the pencilled annotations Weiss made in several theorbo continuo parts for Hasse operas performed at Dresden during the lutenist's final years (the last as late as October 1749).[32]

The compiler's contribution to Dresden

Perhaps the most teasing problem in Weiss research is that of establishing an identity for the compiler of Dresden. Neeman's chain of conjecture leading to the conclusion that a Saxon councillor-of-war appointed in 1739, Friedrich Wilhelm Raschke, was a pupil of Weiss who compiled Dresden under the composer's supervision, and later worked as a quasi-professional lutenist in the Dresden opera orchestra,[33] has been convincingly disposed of by Wolfgang Reich in his introduction to the 1979 facsimile edition.[34] Dr Reich clearly establishes that the compiler's contribution was carried out over a long period, from as early as c.1731 to as late as the early 1760s. For details, I refer the interested reader to Dr Reich's 1979 introduction, rather than restating his arguments here. Incidentally, no more evidence has yet come to light to support his intriguing suggestion of a possible Munich origin for the source.

Further to Dr Reich's chronology of the copying of Dresden, I should like to add a few tentative and somewhat conjectural suggestions, which owe much to Andre Burguete's long and deep acquaintance with the manuscript.

From a close study of the compiler's handwriting, the paper-types and the rastra used to rule the tablature staves, too laborious to report here, a few general observations can be made. The most significant actually concerns what is probably the last phase of the compilation.

The compiler's latest action seems to have been to `tidy up' the somewhat diverse collection for posterity, possibly even to form the basis of a `Complete Works'. The manuscript's pages were reordered, and the various fascicles rebound in sections by key. Title-pages were provided for the separate sections at the same time.[35] Some of the fascicles (notably most of those in the hand of the unreliable third copyist) show signs of having been originally bound into a different series of volumes whose fore-edges were sprinkled with red ink, a common form of decoration for books in the 18th century. Some even have an earlier written foliation (all three hands are represented among these).

At the time of this final reordering of the collection, a pair of eight-page fascicles was added at the beginning of the first volume, devoted to music in F major and D minor, on which no fewer than three sonatas in F (D1, p.1; D2, p.6; D3, p.11), numbered by the scribe, were copied. (Uniquely in this collection, D2 begins on one fascicle and ends on the second.) The lateness of the handwriting of the scribe is betrayed by clear signs of tremor.

At about the same time, preludes were added at the beginnings of sections (not necessarily specific to the sonatas that immediately follow the preludes, but possibly intended to introduce the relevant key) that lacked a prelude to the opening sonata. (I will make some further comments about these preludes below.) Also, some corrections and position-changes (made by moving a note to another string, an unusual and significant type of alteration in a tablature source) were made in previously-copied sonatas. It is impossible to say with certainty whether these technical changes were made from comparison with a better source or as an `editorial improvement' on the part of the compiler. A change of instrument might even have prompted such action. Certainly more research on this question, like that of the adaption of 11-course music to a 13-course instrument mentioned above, needs to be done.

Possibly a little earlier, it seems probably that the compiler gained access to Weiss's personal archive of music and acquired: i) the late d minor autograph (D7, p.41); ii) the C major `Lobkowitz' sonata (D12, p.91); iii) the other big, mature sonatas unique to this source as exemplars for copying (D8, 14, 16, 20, 21, 27, 30), and presumably autograph exemplars for the chamber music in volume 6;[36] iv) a good source for numerous corrections (e.g. to the E flat sonata D34, p. 288) which he considered more authoritative than his own copy - this must have been an autograph, too.

This points to a person with some connection to Weiss's widow in Dresden, and with some influence, perhaps even a high-ranking court official. Although the standard of copying and musical integrity of the copies made by the compiler is on a very high level, it is probable that we are dealing with an `amateur' or `dilettante' (in the best sense of both words) musician rather than a professional lutenist. In some cases, this person seems to have lacked the confidence to make necessary corrections in the music without an exemplar; there are almost no alterations to the corrupt versions copied by scribe 3, for example. And indeed, the quasi-obsessional nature of the 30-year quest for Weiss's music, which rapidly fell out of favour after the great lutenist's death, is an essentially amateur trait.

Finally, I should like to suggest that a professional lutenist would hardly have needed to add preludes in tablature to sonatas, except when Weiss had composed these as `integral' movements. The improvisation of preludes and fantasias was the very foundation of the Baroque lute-player's art, as is attested by the frequent allusions to Weiss's preeminence in this practice.

This raises the problem of the preludes in Dresden rejected by my colleague and friend Douglas Smith as `spurious'.[37] He spurns these on the grounds that they are too ill-composed, brief and inconsequential to be works by Weiss, the great lutenist and composer. While not wishing to disagree with the value-judgements, I hope to provide one possible explanation which could retain them in the Weiss canon, albeit by a somewhat devious argument.

Preludes are apparently `integral' in only six out of 34 sonatas (17.6%) in Dresden; they are added by the compiler to a further five, and in these cases, they appear at the beginning of key-sections in the collection. If, as is perhaps implicit in Dr Smith's rejection of them, the compiler composed the `spurious' preludes, why did he not add preludes to all the sections not beginning with a genuine Weiss example? (E.g.: p.102, before the A minor section; p.261, before the C minor section; or even on a leaf he could have inserted before p.235, beginning a single-piece F minor section.) It is clear that this amateur lutenist was in awe of Weiss, and would hardly be likely to feel competent to add his own music to that of the great man - he certainly seems to have been concerned to root out spurious music by others. Furthermore, if the `spurious' preludes are not by Weiss, why is there no explicit indication of that fact?

My impression is that these are very like fragments from a longer prelude or preludes by Weiss. They seem perfectly plausible in terms of style and idiom, but they are very `summary' in nature. As Dr Smith says, they are `weak compositionally'; but is it an absolute sine qua non for a prelude to have `compositional strength' in order to be by Weiss? The very function and raison d'être of the lute prelude is somewhat counter to notions of high art; I would suggest that Weiss's excellent composed examples, either those `integral' to sonatas or added later, for example, to sonatas in London, are the exception rather than the rule in the genre. These `spurious' preludes are surely substitutes for an improvised prelude, written in for the benefit of someone without that particular professional skill.

As to the source from which the compiler acquired the `spurious' preludes, it could be argued that they are in fact torsos from a `dismembered' large-scale piece. Perhaps this was the lost C minor `Preludio ne quale sono contenuti tutti i Tuoni musicali' advertised by Breitkopf as opening (or perhaps comprising the whole of) Partita 18 in the 1769 thematic catalogue of Weiss's lute music for sale in manuscript copies.[38] A prelude with an identical title by Adam Falckenhagen, probably the one also advertised by Breitkopf together with Weiss's a few years earlier,[39] survives in an Augsburg manuscript. Many sections of this very extended piece, taken out of context, are in similar vein to the `spurious' Dresden preludes.[40] A complete performance of the Falckenhagen prelude takes about 25 minutes; if Weiss's Preludio was of similar dimensions, it would not be so surprising if it were plundered in this way.

Summary

The most important fact that emerges from a first close critical examination of London and Dresden is that we cannot simply describe one of these sources in toto as more or less `authentic' than the other, in that it more accurately represents the composer's wishes. Both manuscripts provide authentic sources for some works, and authentic evidence for the composer's revisions in some cases, but not all works in all cases.

However, both also contain traps for the unwary. In London there seem to be some pieces that, in my view, are not by Weiss; in Dresden, some music that has been rejected by some commentators may indeed be authentic Weiss, although transmitted in unsatisfactory condition. Again, there are many complicating factors in using these sources as a basis for establishing a chronology, at best a risky enterprise, since the `original' state, or the chronological sequence of the copying `layers' is much more complex than it at first appears.

In both sources, the number of preludes added retrospectively to `complete' sonatas gives the impression that in general, Weiss, being a master of the art of improvisation, attached less importance to written preludes than did his disciples, who needed a model from which to perform. Clearly, by implication, the sonatas need preludes in performance; whether they are those written out by Weiss, or are improvised, depended then, and depends today, above all on the skill and experience of the performer.

I have barely touched on the many questions of lute technique, performance practice and musical style that these manuscripts open up and - much more rarely - answer. But I hope I have shown that the forthcoming publication of Dresden in the Sämtliche Werke is, like this Congress, a small step in the right direction, towards a new area of musicology, involving players and scholars in close cooperation, perhaps one day meriting its own properly-funded institute and journal: Weiss Studies.