A Matter of Hono[u]r: Editing and Performing Beethoven’s Late Quartets in 1840s London

Abstract

The importance to reception history of the first complete cycle of Beethoven’s string quartets, given in London in 1845 by the Beethoven Quartett Society, is securely established. Less well recognized is the significance of the complete edition of the quartets prepared by the cellist Scipion Rousselot and published in London in 1846. This article offers the first close examination of Rousselot’s edition and posits that while its claims for unparalleled correctness were unsustainable, its inclusion, uniquely in the case of the late quartets, of two uncommon features – rehearsal letters and instrumental cues – constitutes a trace of the rehearsal practices of the four players responsible for these historically outstanding performances.

The reciprocal relationship of text and act (to borrow an established formulation) is acutely demonstrated in the compositional history of Beethoven’s late string quartets. ¹ As is well known, Beethoven was able to draw on the resources of his ‘Leibquartett’ – Ignaz Schuppanzigh, Karl Holz, Franz Weiss, and Joseph Linke – in shaping and refining many details of the scores. ² The conversation books for 1825–26 preserve precious record of exchanges, above all with Holz and Schuppanzigh, concerning notation and execution as well as the first performances of the three ‘Galitsin’ Quartets, opp. 127, 132, and 130. ³ More recently, Nancy November, drawing a parallel between Beethoven’s interactions with his personal performing ensemble and her own work with the Fitzwilliam String Quartet, has argued that performance can inform not just composition but also editing. ⁴ And contemporary performance scholarship benefits more generally not just from the increasing availability of recordings but also from the testimony of performers themselves. ⁵ Accessing the performing mores of earlier generations is, of course, another matter. Writing of performances of the late quartets prior to the mid-nineteenth century, Christina Bashford makes the obvious point that ‘the nature and quality of the actual performances – lost to thin air, unlike ones of the gramophone era – can of course never be recovered’. ⁶ Reviews and personal accounts, inasmuch as these survive, assist in the imaginative reconstruction of events; ⁷ and projects such as CHASE and TCHIP have sought to harness the potential of performing editions, some supplemented with performance markings, in advancing the study of performance practice, as well as bringing to bear the importance of ‘pre-performance’ issues. ⁸

The present study is the first to examine closely an edition of the Beethoven string quartets that intersects vitally with a particular performing milieu. The historical stature of the 1845 cycle given in London by the Beethoven Quartett Society is not in doubt; on the other hand, the complete edition of the quartets prepared by the group’s cellist, Scipion Rousselot, and published the following year has received scant acknowledgement. As will be argued later, two seemingly unique features incorporated into the text of the late quartets alone in this edition are best understood not simply as objective aids to future performances but as a trace of the interactions – the performative acts – of a specific ensemble in a specific cultural setting.

I

‘HONOUR TO BEETHOVEN.’ Turning to page 263 of The Musical World for 29 May 1845, the reader would have encountered this eye-catching heading to an advertisement in the top left-hand corner. The notice, dated 8 May, was placed by the publisher R. Cocks and Co. to

inform Amateurs and Professors of Music that they are about to publish an entire edition of BEETHOVEN’S SEVENTEEN QUARTETS For two Violins, Tenor and Bass, in such a correct and elegant form as will surpass all other editions. To assist this object R. Cocks and Co. acknowledge with lively gratitude their obligations to the Beethoven Society of London for the loan of their corrected copies of those stupendous works. The whole will be edited and carefully corrected by Scipion Rousselot. No exertion or expense will be spared by R. Cocks and Co. in rendering this edition entirely free from the errors of which amateurs and professors have had such just reason to complain in former editions which have formed an obstacle to the perfect knowledge of Beethoven’s exalted genius. ⁹

By ‘the Beethoven Society of London’ was meant the Beethoven Quartett Society, founded in 1845 by Thomas Massa Alsager, although one of its chief purposes, to introduce the almost unknown and unappreciated late quartets to a discerning audience, was already in his mind some three years earlier. ¹⁰ Each of the five late quartets – the Grosse Fuge, op. 133 went unplayed – was programmed along with an early and a middle-period work in the series of five fortnightly concerts in 1845 that constituted the first complete Beethoven quartet cycle ever to be performed:

• 21 April: op. 18 no. 1; op. 59 no. 3; op. 127

• 5 May: op. 18 no. 3; op. 59 no. 1; op. 131

• 19 May: op. 18 no. 4; op. 74; op. 132

• 2 June: op. 18 nos. 2 and 6; op. 95; op. 135

• 16 June: op. 18 no. 5; op. 59 no. 2; op. 130

An important element in Alsager’s quest for performances that would be truly worthy of the greatness of these works, and the late quartets in particular, was consistency of personnel; as he had expressed it in a letter to Mendelssohn of 14 June 1842, ‘The obvious & direct mode is that of inducing four players to make a careful study of those wonderful quartetts & then to collect an audience of professional musicians to hear them performed.’ ¹¹ Thus The Musical World could announce already on 1 May 1845 that ‘Mr Alsager has secured the services of Messrs. [Henri] Vieuxtemps, [Camillo] Sivori, [Henry] Hill, and [Scipion] Rousselot, for a series of five seances; the performance to consist of the quartets of Beethoven’, noting in addition that Prosper Sainton had substituted for Sivori, who was yet to arrive in London, for the already concluded first concert. ¹² And implicit in the project to raise awareness and appreciation of the little-known late quartets was their performance from an accurate musical text; thus, given the short period of time between the first of the five ‘seances’ and the 8 May advertisement placed by Cocks and Co., it may be assumed that the plan to produce a new, authoritative edition of the quartets was part of Alsager’s plan for the Society from the outset.

That Beethoven’s late quartets were regarded not merely as works to be performed – and performed accurately – but as objects of study and intellectual enquiry was recognized by their earliest reviewers, and marked by their unconventional (near-)simultaneous publication in both parts and score. ‘We must be aware of great gratitude toward the Schott music firm for making the works of our Beethoven available to us in score’, runs an anonymous 1828 notice in the Allgemeine Musikzeitung zur Beförderung der theoretischen und praktischen Tonkunst for the newly published score of the Quartet in C♯ minor, op. 131, ‘for such works must be studied most urgently by prospective artists. But such a score is likewise very welcome for works that present such great difficulties in performance’. Even earlier, and allowing for commercial interest in boosting sales, Schott emphasized in a notice in Cäcilia the benefit they were providing in issuing the Quartet in E♭, op. 127 in score in 1826: purchase was recommended ‘not just to students of harmony but to all those […] who wish to study the quartet itself and perform it genuinely well’. Writing in the same journal that same year, Gottfried Weber agreed, praising the fact that the quartet ‘is also given to us in score, in which form it is both accessible for study and also a most rewarding aid and facilitation for performance, particularly in view of the significant difficulty of execution’. ¹³ For Beethoven Quartett Society audiences and others, study could be combined with the simultaneous experience of performance thanks to the pocket-sized miniature scores of the quartets published by the Mannheim publisher Karl Ferdinand Heckel during the 1840s and 1850s, and distributed in London through firms such as J. J. Ewer and G. A. Augener. ¹⁴ Indeed, a fortnight before the announcement of Rousselot’s complete quartet edition, The Musical World on 15 May 1845 carried an advertisement for Ewer’s scores of ‘Beethoven’s, Mozart’s, Mendelssohn’s, and Onslow’s Quintetts and Quartetts in Score’, at prices beginning at 2 shillings and sixpence. ¹⁵

The timing of the Ewer advertisement may have been influenced by the enthusiastic reception of the first two Beethoven Quartett Society concerts. Its main purpose, however, was to promote a new edition of Haydn’s eighty-three string quartets in score, complete with ‘Thematic Catalogue’. In describing this edition as not only ‘NEW’ and ‘COMPLETE’ but also ‘CORRECT’, the advertisement heading pre-empted the most prominent element in the 29 May 1845 announcement by Cocks and Co., with its promise of a Beethoven quartet edition of surpassing correctness and elegance at the hands of the cellist-editor Rousselot. ¹⁶ In the wake of the unprecedented complete series of concerts, the Cocks announcement reappeared in a more extended and elaborate form in The Musical World for 17 July. ‘Amateurs and Professors of Music’ were now invited to subscribe at a cost of three guineas, ‘to be paid when the whole is delivered’; moreover, ‘a short Biographical Notice will be given of Beethoven with the first Quartet; and, at the close of the Series, a copy of each programme of the five performances of the Beethoven Quartet Society will be given.’ The publishers hoped to deliver op. 18 ‘by the end of July, and the remainder as soon as practicable’. ¹⁷ In the event, the process would take a further year to complete. A notice ‘to the subscribers to M. Rousselot’s edition of Beethoven’s Quartetts’ issued by Cocks and Co. on 7 March 1846’ announced

that this Edition is now in a state of great forwardness, the Twelfth Quartet [op. 127] of the Series having received the last corrections of the Editor’ […]

The whole of them, after final correction, will be carefully played over by Messrs. SIVORI, SAINTON, HILL, and ROUSSELOT, the Artistes engaged for the performances of the Beethoven Society, to discover whether any errors yet remain; and thus Messrs. R. COCKS and Co. hope to redeem the pledge given in their first announcement, in presenting to their Subscribers a publication of Classical Music, unrivalled in point of style and correctness. ¹⁸

By 21 March it was announced that the final quartet, no. 17 [op. 135] ‘is already engraved, and the whole, ere long, will be ready for delivery’. ¹⁹ Thus, the second performance cycle for the Society, scheduled weekly rather than fortnightly between 4 May and 1 June 1846, might itself have been expected to furnish some of the careful playing over projected by the publishers.

As the 17 July advertisement suggests, it was not the intention to release all seventeen quartets together in a single publication. Rather, each quartet was issued, individually priced, with a separate plate number, each part provided with its own title page headed ‘Rousselot’s Edition’, as a fascicle of unbound parts formed of a combination of gathered bifolia and single leaves. Each part was separately paginated at the top left- and right-hand corners; however, an additional pagination in brackets was added centrally, to run continuously through the seventeen quartets, so that each of the four parts might in due course be bound into a separate, fully paginated volume. Additionally, Cocks and Co. produced a wrapper for each fascicle, ornately styled (Figures 1–2):

HONOR [sic] TO BEETHOVEN. / Complete Edition / OF THE / QUARTETTS, / OF / L. VON [sic] BEETHOVEN, / Carefully Edited from the Copies prepared & corrected / FOR THE /Beethoven Quartett Society of London, / BY / SCIPION ROUSSELOT, / and Dedicated to / THE NOBLE PRESIDENT AND TO THE COUNCIL, / of that distinguished Society.

Figure 1. Beethoven, String Quartet in C♯ minor, op. 131 (Violin I), Rousselot’s edition, title page. The Bodleian Libraries, University of Oxford, Mus. 177 c. 137 (15).

Figure 2. Beethoven, String Quartet in C♯ minor, op. 131 (Violin I), Rousselot’s edition, wrapper. The Bodleian Libraries, University of Oxford, Mus. 177 c. 137 (15).

Given that the total cost of the edition if each quartet were purchased separately would have amounted to £8 15s 6d, the subscription price of three guineas (£3 3s) was something of a bargain. Moreover, the prefatory matter supplied with op. 18 no. 1 included not just a reproduction of the 1824 Steinmüller portrait of Beethoven, complete with signature, and a four-page ‘Brief Memoir of Beethoven’, but a tribute (in French) from Rousselot to Alsager; a ‘Notice’ by Cocks and Co. (this and Rousselot’s tribute bear the date 1 August 1846); a list of subscribers extending from Prince Albert and the Duke of Cambridge through London conservatories, provincial music sellers and individuals as far away as Bermuda and India; and a thematic catalogue (‘Catalogue thématique’) of the seventeen quartets. ²⁰ The question of accuracy still loomed large in the publisher’s mind: in describing the publication as ‘an undertaking far surpassing in difficulty any of a similar kind which have been attempted in this country’, Cocks’s ‘Notice’ offered thanks to the Council of the Beethoven Society and above all to Alsager ‘for the loan of his corrected copies of the scores and separate parts of those unrivalled compositions, which had previously been prepared, at great expense and trouble, for the performances of that Society’. Despite the unfortunate delay that had occurred in bringing out the edition, ‘it is hoped that its correctness (owing to the great care and unwearied attention of M. ROUSSELOT) will in some measure compensate for the time which has been devoted to it’. For his part, Rousselot assured Alsager of his efforts to render the edition ‘as perfect as possible for the execution of the composer’s intentions’; ²¹ and referring specifically to the ‘posthumous Quartetts’, the author of the ‘brief memoir’ argued that

we can easily conceive how they have been pronounced unintelligible by persons who have judged them according to their confused execution, which has been the necessary result of the wretched incorrectness of all the editions that have appeared to the present day. If the performers take the requisite pains to produce all the effects, and all the nuances, following a good edition, these Quartetts will appear as intelligible as the rest, and the beauties with which they are filled will be easily appreciated. ²²

Reviewing the edition on 5 September 1846, The Times – no doubt in the person of J W. Davison, concurrently owner and editor of The Musical World – was convinced:

The present edition, by Messrs. Cocks, the first complete one in the English language [!], is worthy of the zeal by which the publishers appear to have been incited to spare no pains or expense upon a publication which is calculated to elevate among foreign artists the character of this country in this department of art. Everything which genius and judgement could effect seems to have been accomplished for the purpose of rendering the text immaculate; and the mechanical parts of the publication, including the plates, the printing, and even the paper, are of the highest excellence. The work is got up in a style worthy of the great genius whom it is intended to honour, and is indispensable to the collection of every musician who is capable of enjoying the numberless beauties which it contains. ²³

II

Even allowing for a degree of hyperbole in the interests of commercial sales, the repeated emphasis on the superiority of this new edition over all previous ones demands attention: exactly how justifiable were these claims, and what kind of editorial scrutiny had Rousselot, and Alsager before him, applied to their task? Cocks’s ‘notice’ refers to Alsager’s ‘corrected copies of the scores and separate parts’; and Henry Hill’s prefatory ‘short account’ of the Beethoven Quartett Society, published in 1846 together with a new edition of the 1845 programmes, recalls that

Great obstacles were to be overcome in the first instance, in the rescue of Beethoven’s Quartetts from obscurity, in consequence of the numerous errors with which all the editions of them abounded. The Queen Square Society, and that which we are at present describing, were under great obligations in this respect to the great musical knowledge, and research, and industry of M. Scipio Rousselot; who succeeded with infinite labour in such a correction, first of the scores and then of the separate parts, that probably not a single fault remained when these copies were employed for the performances of the Beethoven Quartet Society. ²⁴

It is natural to assume from Hill’s text that the abounding errors in existing editions that had so long marred performances and comprehension, especially of the late quartets, were principally pitch-based, although it will be suggested below that this claim was exaggerated. Indeed, Hill’s ‘short account’ reprises the then common narrative ‘relative to the later works of Beethoven, which were pronounced crude, and wild, and discordant; as indications, in fact, that, with the loss of the faculty of hearing, the monarch of modern musicians had lost the power of producing just harmonies, and works at once truly great and original’. ²⁵ However, it is notable that the ‘brief memoir’ prefacing the new Cocks edition adds to the avoidance of ‘wretched incorrectness’ the need also to observe ‘all the effects, and all the nuances, following a good edition’. ²⁶ This implies attention to matters of phrasing, dynamics, and other non-pitch-based elements in the text.

Neither Alsager’s nor Rousselot’s own scores and parts are known to survive. In the case of the late quartets it is highly unlikely that they could have had access to Beethoven’s autographs or the several corrected manuscript copies available to modern editors. ²⁷ It is reasonable to assume that they would have had access to the first editions published variously by Schott, Artaria, and Schlesinger. But their most immediate source is likely to have been the Collection complète DES Trios, Quatuors et Quintetti … PAR L. VAN BEETHOVEN published in parts by Maurice Schlesinger in Paris in 1827 and succeeded two years later, in May 1829, by a ‘Nouvelle Édition très soigneusement corrigée’. As Emil Platen has pointed out, Schlesinger’s announcement that this second edition incorporated corrections carefully made by those in Vienna who had performed the quartets ‘under the direction of Beethoven himself’ can refer only to the Schuppanzigh Quartet, and above all to Karl Holz, who had been deeply involved in the negotiations for and publication of the late quartets. Platen accordingly considers Schlesinger’s 1829 ‘Nouvelle Édition’ to be ‘the most authoritative of all early prints’ of these six works. ²⁸

That Schlesinger’s edition was well known to Alsager and to the Beethoven Quartett Society is evident on at least three fronts. In his 14 June 1842 letter to Mendelssohn, lamenting the widespread ignorance of ‘some of the greatest works the art of music has ever produced’, Alsager specified ‘the quartets of Beethoven N^o. 12 to 17, Schlesinger’s Edition, published after his death’. ²⁹ And in the 1845 Quartett Society programmes, not only are the late quartets numbered as they appear in Schlesinger’s edition (Table 1) but opp. 95, 130, 132, and 135 are specifically listed as being played in ‘Schlesinger’s edition’, which implies that the other works were played from some other text. ³⁰ It is reasonable to assume that Schlesinger’s edition – presumably the later, 1829 ‘Nouvelle Édition’ – was the ‘base’ text from which Rousselot, and perhaps Alsager before him, worked; although no part of Rousselot’s edition had been issued by Cocks before the first season of concerts ended on 16 June 1845 (recall that Cocks and Co. had hoped to issue op. 18 by the end of July that year), the performers might reasonably have played from copies of Schlesinger’s edition already edited by Alsager and/or Rousselot in preparation for their subsequent publication. That no. 17 (op. 135) was reported as being near readiness for publication only in late March 1846 (see earlier) is also consonant with its documented performance in ‘Schlesinger’s edition’ almost a year prior to that date.

Table 1 Numbering of late quartets in Schlesinger, Collection complète (1827; ‘Nouvelle Édition’, 1829), Beethoven Quartett Society Programmes (1845) and Rousselot’s edition (1846)

Op. no.	Schlesinger Collection complète (1827; ‘Nouvelle Édition’, 1829); Beethoven Quartett Society Programmes (1845)	Rousselot’s edition (1846)
Op. 127	No. 15	No. 12
Op. 130 a	No. 13	No. 13
op. 133	No. 14 (not performed by BQS)	No. 14
Op. 131	No. 16	No. 15
Op. 132	No. 12	No. 16
Op. 135 b	No. 17	No. 17

^a The 16 June 1845 programme incorrectly gives op. 132.

^b Schlesinger’s edition gives op. 134; the 1845 programme correctly gives op. 135.

Other factors strongly suggest the influence of the Schlesinger edition upon that of Rousselot. In their front matter both include the 1824 Steinmüller portrait of Beethoven, complete with signature, and a ‘Catalogue thématique’ with the seventeen quartets individually numbered. ³¹ Contrary to their (Schlesinger) numbering in the 1845 Quartett Society programmes, however, the late quartets now appear in a new numbering in Rousselot’s edition (Table 1).

As is evident, Rousselot’s numbering corresponds to the sequence of opus numbers, with the exception that op. 133 is placed next to its ‘parent’ opus; moreover, a footnote to the ‘Catalogue’ not found in Schlesinger clarifies that ‘This [op. 133] was Composed as a Finale for op. 130 and was published afterwards as a separate Quartett.’ With this exception – which might be considered a kind of ‘correction’ – Rousselot’s numbering corresponds to that of Haslinger’s 1832 ‘Systematische Verzeichniß der […] Originalwerke von Ludwig van Beethoven.’ ³²

Another respect in which Rousselot’s edition reflects that of Schlesinger is to be seen in the presentation of the ‘mottos’ at the head of the finale of op. 135. In both editions the principal language employed in the underlaid text is French, rather than German: ‘Le faut il? il le faut! il le faut!’. Beethoven’s German, ‘Muss es sein? es muss sein! es muss sein!’, appears in smaller print underneath. Schlesinger gives ‘seyn’ rather than ‘sein’, and even places a French heading, ‘Un effort d’inspiration’, in relatively large print above the German original, ‘Der schwer gefasste Entschluss’. In this respect, the two editions do differ slightly, in that the French heading is not present in Rousselot’s edition.

Finally, the presence of Schlesinger’s Collection complète DES Trios, Quatuors et Quintetti behind the aspirations of Cocks and Co. is made clear in the publishers’ ‘Notice’ to Rousselot’s edition, in which they

also beg respectfully to announce, that, in consequence of many applications from the Subscribers to the Quartetts for an uniform edition of the Quintetts and Trios, they intend publishing by subscription Beethoven’s Three Quintetts and Four Trios for stringed instruments. They have been so fortunate as again to secure the valuable services of M. ROUSSELOT, as editor; and they earnestly request an early intimation from those gentlemen who kindly intend to assist this undertaking.

This edition was being advertised by Cocks and Co. by December 1848. ³³

Writing of the performances of late quartets led by Henry Blagrove and Nicolas Mori in London in the late 1830s, Bashford writes that ‘both sets of musicians played from published parts (Schlesinger’s Paris edition seems to have been used), which, in spite of Scipion Rousselot’s later claims to the contrary, were probably reasonably accurate texts’. And while noting that ‘the validity of the claims [of Rousselot] has never been systematically examined’, she notes that ‘spot checking reveals minor differences of articulation and dynamic placing rather than of pitch or rhythm’; ³⁴ a more thorough cross-checking of the entire quartets in the two editions shows this assertion to be broadly correct. Bashford was writing well over a decade before Emil Platen’s exhaustive textual apparatus supporting the new Henle edition of the late quartets became available, and although Platen does not consider Rousselot’s edition, his tables of principal variants (Leitvarianten) between the manuscript and early printed sources for each quartet on which his edition is based make for a useful comparison with the fruits of Rousselot’s ‘great musical knowledge, and research, and industry’, as Henry Hill had described it. ³⁵

The results are mixed. In the case of op. 127, for example, Platen cites five variants in the second and third movements that are unique to the original March 1826 Mainz and Paris editions of the parts (his sources C and D); of these five, three are found in Rousselot’s edition, while the other two readings are in agreement with Schlesinger’s edition (Table 2).

Table 2 Variant readings in op. 127, second and third movements, after Platen

Op. 127	Platen sources C, D (Mainz and Paris, March 1826, parts)	Rousselot (1846)	Platen sources F1–4 [= Schlesinger 1827, 1829]
Second movement, Viola, bar 43, penultimate semiquaver	c ′ /e ′♭	c ′ /e ′♭	a ♭/e′♭
Second movement, Violin 1, bar 71, 3rd note	d′′′♯	b ′′	b ′′
Second movement, Violin II, bar 89, 5th note	g ′	g ′	f′
Second movement, Cello, bar 93, final note	b ♭	b ♭	g
Third movement, Violin I, bar 376, 1st note	ff	sf	sf

Of six further variants, Rousselot and Schlesinger share their readings in four cases, while in two they differ: in the first movement, Violin II, bar 123, 4th note, Rousselot’s edition lacks the natural sign before a′ present in Schlesinger, while in the third movement, Violin I, bar 53, 5th note, Rousselot’s edition again lacks the necessary flat sign before the printed d′.

Similar examples can be found in the other quartets. Particularly interesting, in view of the arguments made earlier for the likelihood that the Schlesinger edition was Rousselot’s ‘base’ text, are those cases in which Schlesinger’s 1827 edition differs from that, ‘très soigneusement corrigée’, of 1829 (Table 3).

Table 3 Variant readings in op. 130, 132, and 135, after Platen

	Schlesinger (1827)	Schlesinger (1829)	Rousselot (1846)
Op. 130, sixth movement, Cello, bar 292, 1st note	f	F	F
Op. 132, first movement, Violin I, bar 130, 2nd note	d ′♯	e′	d ′′♯
Op. 132, third movement, Cello, bar 56, 4th note	A	d	A
Op. 133, Violin I, bar 360, 1st note	d ′♭	d ′	d ′♮
Op. 133, Violin I, bar 362, 2nd note	d′′♭	d ′′♮	d ′′♮
Op. 135, second movement, Viola, bar 120, 5th note	d′♯	d ′	d ′
Op. 135, fourth movement, Cello, bar 146, 2nd note	F♭	F	F

As can be seen, Rousselot’s readings tend to favour the 1829 Schlesinger edition, but not consistently. In other cases where both Schlesinger editions agree, Rousselot’s may offer a different reading, as in the case of op. 132, second movement, Violin I, bar 177, the fourth note: a′ appears in all early sources with the exception of the original Berlin edition of the score (Adolph Martin Schlesinger, ?October 1827), which gives b′; and b′ proves to be the reading in Rousselot’s edition.

Needless to say, the fact that one or more editions adopt a particular reading does not of itself argue for the ‘correctness’ or otherwise of that reading. There is at least one case, though, in which it is clear that Rousselot succeeds where Schlesinger erred. Platen points out that the 1827 and 1829 Schlesinger editions, uniquely among the early sources, give the tempo marking Allegro ma non troppo at the head of the first movement of op. 130, in the Violin II part only. This was presumably a straightforward printing error; it is not found in Rousselot’s edition, which correctly, one can be confident in this instance, gives Adagio ma non troppo in all four parts.

All in all, then, the evidence is not unequivocally in favour of claiming that Rousselot’s edition provided a text that was notably closer to ‘the composer’s intentions’, as he had put it, than those already available to quartet players at this time; certainly, the claims of the ‘wretched incorrectness’ of previous editions made in the edition’s ‘Brief Memoir of Beethoven’ seem exaggerated. Moreover, just three instances are sufficient to counter Hill’s surmise that ‘probably not a single fault remained’ when the Beethoven Quartett Society performed from these scores. The first can probably again be ascribed to a simple printer’s error. In op. 130, first movement, the change of key signature to six flats is missing before bar 71 in the Violin I part. Bars 71–7 2 are the last two bars on the first page of this part, and the correct new signature appears at the top of the following page, before bar 73. The error is unique to this part; if played as written, the result would indeed have been, in Hill’s words, ‘crude, and wild, and discordant’.

The other two instances, both from op. 135, are more significant. In the first movement, Viola, bar 95, the second triplet group reads a–b–c′. As Platen shows, this reading is common to all editions consulted other than Beethoven’s autograph, which reads c′–b–a. Considering the Viola part alone, one can see that a–b–c′ would appear consistent with the figure on the first and third triplet groups in the bar; but as Platen notes, it produces parallel octaves with Violin II and doubtless must be incorrect for that reason, so that Beethoven’s autograph reading should stand. ³⁶ It might be thought that such a contrapuntal solecism would have been discovered in the ‘careful playing over’ that Sivori, Sainton, Hill, and Rousselot were due to give the latter’s final corrected versions. Even more is this the case with the third instance, from the final movement of op. 135. Rousselot’s edition perpetuates here an error found in the Berlin parts and score editions as well as the 1827 Schlesinger edition, namely the absence in the Violin I part of bars 133–34, which are compensated for by two extra bars (bars 146a/b in Platen’s numbering) between bars 146 and 147 entered by Beethoven in his autograph but then cancelled; the error is corrected, though not without a ‘compensating’ textual corruption, in the 1829 Schlesinger edition. ³⁷ Again, the barely disguised parallel octaves in Violin I and Cello resulting from this incorrect alignment of parts (Example 1) ought alone to have raised suspicions.

Example 1 Op. 135, fourth movement, bars 124–50, Rousselot’s edition.

A different kind of absence from Rousselot’s edition, although not from Schlesinger’s 1829 one and others, are the ritmo di due, tre, quattro battute indications found in the third movement of op. 127 and nos. 5 and 7 of op. 131. And suspicions might also be aroused in the mind of Rousselot’s modern reader or player on turning to the Cavatina from op. 130 and finding the beklemmt direction missing at the beginning of the central C♭ major section at bar 42. But this is in fact an omission that is shared by all early sources, including the first editions and Schlesinger’s Edition complète, other than the autograph score, where it was a later addition rather than part of the initial writing out. ³⁸ Thus, prior to some time after the mid-nineteenth century, this highly suggestive marking that has been so influential in the reception of this particular movement was simply unavailable to inform performers’ approaches.

III

The last six, the posthumous Quartetts, directly belong to the third epoch […]. Here there is nothing but free, bold fancy. The composer no longer restricts himself to conventional limits; all is independent, unexpected, original […]. Hence these Quartetts require great attention, and a very careful execution, that all their beauties may be comprehended. ³⁹

Alsager’s insistence, in writing to Mendelssohn on 14 June 1842, on the need for quartet players ‘to make a careful study’ of the late quartets has been noted earlier. An emphasis on preparedness for performance, in precisely Alsager’s words, is also communicated in the account of the Beethoven Quartett Society in The Times for 6 June 1845, where the ‘undertaking’ of the ‘London Beethovenists’ is held to be ‘the performance of the best music by the best artists diligently trained for the purpose, in the presence of an audience […] able to appreciate it’. Thus,

The zeal of the performers has corresponded with the claim thus made upon their best exertions; each has engaged, we understand, in a careful study of his part, and no piece has been played at these meetings without having previously undergone seven or eight rehearsals. ⁴⁰

Hill’s account of goals of the Society likewise pointed to the reciprocal relationship between audience and performers when he noted that ‘under such circumstances alone can the art of the accomplished player, employed in rendering the inspirations of a genius like that of Beethoven, be truly brought out and developed’. ⁴¹ But while the individual performer is emphasized in these cases, Hill’s earlier attention was on ‘careful study’ of a somewhat different kind:

Though the formation of the Society is but recent, the system of which it is the result has been for many years in operation. A small party of amateurs, well known in musical circles as the Queen Square Select Society, had gradually associated with them many leading professors, foreign as well as native, who were induced to a careful study of the principles upon which success in Quartett-playing depends; and these being applied to the later works of Beethoven in that class, his posthumous Quartetts, their true character soon became manifest. ⁴²

Here it is not individual competence and artistry that is the target, but rather the matter of accomplished ensemble playing. Successful performance of the ‘posthumous Quartetts’ required not merely four players, each addressing his individual part, but rather the combination of all four playing as one. ⁴³

This aesthetic of string quartet playing had been vividly articulated in 1828 by Adolph Bernhard Marx, in a review of the Schlesinger scores of op. 132 and op. 135. For Marx, Beethoven’s late quartet writing effectively reconfigured the genre. Silently referencing the conversational metaphor favoured by Goethe and others, Marx understood the individual instrumental parts of op. 132 as ‘no longer four happy brothers in art who make music for us for their, and our, enjoyment; they are four deeply affective creative spirits who soar up into magnificent freedom and wondrous sympathy in a brotherly embrace intertwined fourfold’. Likewise,

If the performers do not constitute an equal union of noble, equal, free, brotherly spirits, no perfect manifestation of the work of art is conceivable, nor is full satisfaction for the players to be hoped for. No small amount of training is required of every player in order to master his part technically and give it suitable tone, power, delicacy, and facility. In addition to these outward conditions, a deep sensibility is required to grasp it inwardly, with deepest soul in their own innermost soul. True artistic training, and long practice for the most highly trained and gifted, are all required before one voice accommodates itself freely and flexibly to the others, and does not appear to give up anything of its own content if it risks everything so as never to disturb the free progress of the others. — Heightened artistic training will make the conditions of such playing easy, as the conditions for declaiming Haydn’s quartets no longer strike our generation as difficult. ⁴⁴

It is against this background that the repeated praise for the ‘perfection’ of the ensemble in the Beethoven Quartett Society performances is to be understood. ⁴⁵ And it is likewise against this background that we may consider two features of Rousselot’s edition of the late quartets that clearly set it apart from Schlesinger’s and other early editions. Both can be traced to Karl Holz’s close involvement in the publication process of the quartets, particularly that of op. 132 and op. 133; and both are directed towards cultivating that ‘great attention’ and ‘very careful execution’ for which Rousselot and his quartet partners were so highly praised.

Instrumental cues

The first of these features is the inclusion of instrumental cues in the individual parts of Rousselot’s edition. These may be said to originate in the set of manuscript parts of op. 132 prepared in July–August 1825 by Joseph Linke and Holz, used in the first private performances in the presence of Maurice Schlesinger in September 1825, then purchased by him and used as the Stichvorlage for his first edition of the parts. As Holz told Beethoven in a conversation book, the labour of copying had given Linke headaches, whereupon he himself took over the task, completing each part from the Più allegro following the fourth movement onwards. ⁴⁶ In copying the Violin II, Viola, and Cello parts of the Più allegro, Holz resorted to the unusual step of providing two conjoined staves, with the complete Violin I part notated above the individual part being copied. This feature was retained in the Paris and Berlin first editions of the parts (the former being that of the Einzelausgabe printed in advance from the plates of Schlesinger’s Collection complète (1827)) and in Schlesinger’s ‘Nouvelle Édition’ of 1829. ⁴⁷

The only other examples of this feature to be found in the printed editions up to and including Schlesinger’s 1829 edition occur in the Violin I part of op. 135: at the beginning of the first, third, and final movements, a cue from the Viola part prefaces the Violin I entry in each case. Beethoven himself wrote out the set of parts which served as the Stichvorlage for the first edition; as with op. 132, that edition is the Einzelausgabe printed in advance from the plates of Schlesinger’s Collection complète by the end of August 1827. ⁴⁸ But unlike Holz’s parts for op. 132, Beethoven’s parts for this quartet do not include the cues found in the printed edition; these must presumably have arisen in-house in Schlesinger’s Paris premises.

Table 4 lists all the cues found in Rousselot’s edition of the late quartets. As will immediately be clear, these include the examples from opp. 132 and 135 already discussed but go well beyond those two cases, extending across all quartets and all instruments. In some cases the role of the intervening notation is not strictly to ‘cue in’ the main part but rather to show how that part coordinates with another: an example of this is op. 132, second movement, Viola, bars 221–22, where the Violin II part is notated against the Viola’s held a.

Table 4 Instrumental cues in Rousselot’s edition (1846) of the late quartets

	Op. 127	Op. 130	Op. 133	Op. 131	Op. 132	Op. 135
Vln I	ii: 0–1; 38–39; 76 iii: 75/346; 81/352; 144–47/415–18; 430–31			iii: 6–7; 9–10 a v: 36–37/202–3/368–69/416–17	ii: 218–21 iii: 84–86; 168–70 v (recit.): 1–2; (finale): 1–2	i: 1–2 ii: 0 iii: 1 iv: 1; 10–12; 248–49 b
Vln II	ii: 0–1; 38	i: 14–15 ii: 62–63 iii: 65	17–18; 30; 159; 233–34/533–34; 660	iii: 2; 6–9 iv: 98–99 c; 162–63; 241–42; 264 d v: 33/199/365/413; 36–37/202–3/368–69/416–17	i: 1–2 iii: 0; v (recit.): 1; 3–22	i: 1–3 iii: 1 iv: 1–3; 10–12; 247–48 b
Vla	ii: 0–1	i: 14 ii: 62–63 iii: 66	17–18; 30–31; 159; 233–34; 657; 660	iii: 2; 6; 8–9 d iv: 98–99; 241–42; 265 d v: 36–37/202–3/368–69/416–17	i: 1 ii: 221–22 e iii: 0–1; 84; 168–69 v (recit.): 3–22	ii: 0 iv: 1; 3–4; 12–13; 247–48 b
Cello		i: 14; 24 ii: 62–63	17–18; 30; 159; 493; 657; 660	iii: passim iv: 241–42; 265 v: 35–36/201–2/367–68/415–16	ii: 221–23 iii: 0–1; 84–85; 168–69 v (recit.): 3–22	ii: 0 iii: 1–2; 32 iv: 1–2; 3–4; 12–13

^a Op. 131/iii, Vln I: not really a cue, but showing how cello and Vln I parts coordinate.

^b Op. 135/iv, Vln I: not really a cue, but showing how Vln II and Vln I parts coordinate.

^c Op. 131/iv, Vln II: cue followed by seven bars’ rest.

^d Op. 131/iv, Vln II/Vla: not really a cue, but shows Vln I part above Vln II/Vla held notes.

^e Op. 132/ii, Vla: not really a cue, but shows Vln II entry above Vla held a.

It goes without saying that the purpose of these cues is, at its simplest, to aid ensemble performance by assisting the players to enter at the rhythmically and metrically correct point. In the case of op. 127, the opening of the slow movement may have proved problematic from the perspective of establishing a correct sense of metre at a slow tempo and in a context where syncopations obscure the onset of beats in a ‘long’ four-beat bar. (Similar considerations probably suggested the insertion of cues at the beginning of the second and third chorale sections of the ‘Heiliger Dankgesang’ in op. 132.) Likewise, the several cues for Violin I in the third movement of op. 127, with the exception of that at bars 430–31, occur at points where a change of tempo is involved. The exception might yet have been influenced by rhythmic or metric concerns: in corresponding previous passages (bars 137ff. and 408ff.), Violin I has remained silent; its entry at bar 431 is anomalous not only in this respect but also in that its characteristic dotted figure is shifted a beat earlier than has been the case throughout almost the entire long movement.

It may come as no surprise that the most densely cued section of music in all the quartets, with the exception of the special case of the op. 132 Più allegro, is the 11-bar no. 3 from op. 131, which is of course kin to the op. 132 passage in its improvisatory, recitative-like gestures requiring accurate coordination between instruments. Less easy to understand is the need for the Violin I and II cues in bars 10–12 of the op. 135 finale, where the rhythmic and metrical situation seems uncomplicated; and stranger still is the provision for Violin II of a cue from Cello and Viola at the start of Variation 3 in op. 131 no. 4, bars 98–99, since this is merely followed by a seven-bar rest. The facilitation of the player’s sense of metre at the outset of a new, contrapuntal texture may again be the justification here; nevertheless, one might query why similar assistance was not thought necessary for Violin I.

Rehearsal letters I

The Große Fuge – more properly, the Grande Fugue – op. 133 is celebrated, or notorious, in many respects; one of the lesser known of these is that of its being apparently the first musical composition to be accorded rehearsal letters in its published forms. The impetus for this feature of the first engraved score and parts came from the publisher Mathias Artaria following private performances in the first half of August 1826. Karl Holz was again centre stage: not only had he taken part in the performances, but he also reported on them to Beethoven and conveyed Artaria’s suggestion that rehearsal letters be provided. This necessitated explaining the concept of the letters, after which Beethoven probably also left the task of providing them to Holz. Thus the first edition of op. 133 in score and parts appeared in May 1827, each bearing a sequence of thirteen letters, A–N (I was excluded); these were carried forward to later editions, including those of Schlesinger’s 1827 Collection complète and the ‘Nouvelle Édition’ of 1829. ⁴⁹

Rousselot’s edition not only preserves the original rehearsal letters for op. 133 but also provides them for almost all movements of all the other late quartets (see Appendix). This feature distinguishes his edition from all previous ones, just as does the more extensive provision of instrumental cues discussed earlier. Moreover, these two features – cues and rehearsal letters – are found only from op. 127 onwards; they do not appear in Rousselot’s Quartets Nos. 1–11 (opp. 18, 59, 74, and 95).

In considering the rationale for Holz’s task in op. 133, Barry Cooper notes that Holz would have realized ‘that there was no sense in adding rehearsal letters at changes of key signature or time signature, since these could already function as reference points’. ⁵⁰ Although the point is well made in general, and does hold for op. 133, it is not the case that changes of key signature would infallibly fulfil this function, since they do not always occur at the same point in each part. In the case of the first movement of op. 127, for example, the change of signature from three to two flats that coincides with the beginning of the exposition second group is found consistently before bar 41 in Beethoven’s autograph, the überprüfte Abschrift of that score made by Wenzel Rampl, and in the first edition of the score published by Schott; and it is found at the same place in the first edition of the Violin II, Viola and Cello parts, while in the Violin I part the change occurs one bar later, before bar 42. This arrangement is replicated in Schlesinger’s and Rousselot’s editions. ⁵¹ A much more complex example of this phenomenon may be studied in the parts of the finale of op. 135.

In Rousselot’s edition there are four movements or sections of music that bear no rehearsal letters: these are the second movement of op. 130, nos. 3 and 6 of op. 131, and the fourth movement and linking Più allegro of op. 132. ⁵² In all cases but the first, it is easy to suggest that their brevity would have made the provision of rehearsal letters unnecessary. In the case of the op. 132 Alla marcia, moreover, there were the repeat signs by which to orientate; and Rousselot’s Violin I cue in the succeeding Più allegro made for ease of reference among the other three instruments. The eleven-bar no. 3 in op. 131 was well supplied with cues in all parts; and the combination of frequent repeat signs and the Violin I cue to the reprise at bar 64 would presumably have compensated for the considerably greater length of the second movement of op. 130. This leaves only the twenty-eight bars of op. 131 no. 6, which lacks visual internal divisions such as repeated sections, and also did not prompt Rousselot to add any cues. It is kin to the op. 130 Presto in its strict quadratic phrasing and its largely homophonic textures, however: features which would have made any realignment of parts in rehearsal easy to coordinate, if found necessary at all.

These considerations help to put in perspective the case of the next two shortest movements: the Cavatina from op. 130 (66 bars) and the slow movement of op. 135 (54 bars). Notwithstanding the underlying variation structure, based on ten-bar units, of the latter, it shares with the Cavatina a surface ABA design made all the more visible by the change of key signature from five flats to four sharps and back, together with the concomitant più lento and Tempo primo directions. No such signposts are provided in the Cavatina; and as we have seen, the central section (from bar 42) was without its beklemmt direction at this stage in the quartet’s textual history. On the other hand, the shift to constant quaver triplets in the accompanying parts at bar 40, together with the effective change of key signature with the sudden introduction of so many flat accidentals, and the coordinated sotto voce direction at bar 50 would likely have provided reliable enough markers by which to navigate.

Given all this, it is striking that in both these relatively short and formally clear slow movements Rousselot should have thought it worthwhile to provide two additional points of coordination by means of rehearsal letters (A: bars 16/17; B: bars 37/38) in the Cavatina, and as many as three (A: bars 15/16; B: bars 26/27; C: bars 39/40) in the shorter op. 135 Assai lento. ⁵³ Writing of the rationale for Holz’s decisions in op. 133, Barry Cooper suggests that rather than simply adding rehearsal letters at random points, or spacing them evenly, ‘bars that were structurally important, reflecting a change of texture or new theme, made better reference points. This principle has been widely adopted in more modern editions’. ⁵⁴ While this might be true of letter A in the op. 130 Cavatina – bar 17 marks the beginning of a new, third melodic strain in Violin I following two beats’ rest, even if the remaining instruments are simultaneously cadencing on V/vi at that point – the same cannot be said of the location of letter B, which falls one bar before the perfect cadence which closes the A section of the movement. Likewise in op. 135, letter A is placed between the third and fourth bars of the first variation (beginning at bar 13); letter B at least stands at the onset of a new phrase in the central più lento, following the 2+2 structure of bars 23–26, while letter C marks the eighth bar of the major-mode thematic reprise carried by Rousselot’s cello.

Turning to more complex movements, it is again noticeable that Rousselot’s rehearsal letters repeatedly confound any obvious ‘formal’ readings; this is all the more obvious when, as is often the case, they fail to coincide precisely with parallel passages of music, for example in the exposition and recapitulation of a sonata form. ⁵⁵ The formally challenging and much-analysed first movement of op. 132 may serve as an example. Various tempo instructions already provide points of orientation for the four players, in addition to which Rousselot provided rehearsal letters A–F. That Letter A (bars 37/38) falls so close to a change of key signature may seem perverse; but the change is not congruent from part to part, occurring at bar 37 in Violin II, Viola, and Cello, but only at bar 39 in Violin I. Instead, bar 38 offers a point at which all four instruments converge on a root-position triad on the downbeat, following four bars of syncopated chordal accompaniment beneath the gradually diminuted note values in Violin I. Letter D, at bars 149/150, is the correlate to A (no key signature change is involved here) but in fact is placed one bar later than is ‘correct’ (bar 38 corresponds to bar 149, not bar 150); in this case, Rousselot may have been misled by the fact that the Violin I part in bar 150 is almost identical to that in bar 38. Letters B (bars 66/67) and E (bars 182/183) are likewise correlates of one another, although E is again placed one bar beyond where it should be to correspond with the root-position C-major harmonies that mark the end of what may be understood as Exposition 2 or Recapitulation 1, depending on one’s point of view of this formally notorious movement. ⁵⁶ Letter C (bars 113/114) again falls near a non-congruent key signature change (before bar 111 for Violin II, Viola, and Cello, but one bar later for Violin I), at a point where the four instruments strike a root-position triad on a downbeat. Finally, Letter F (bars 222/223) does correspond to a formal articulation, this being the beginning of the tonic-major recapitulation of the second group; but aside from the fact that bar 223 begins with silence (no instrument strikes the downbeat), the placement of a rehearsal letter here only draws attention to the absence of a corresponding cue at either of the previous instances of this music: bars 48 (F major) and 159 (C major). ⁵⁷

The second movement of op. 127 offers an intriguing example of Rousselot’s choices. The inherently sectional nature of a variation movement, visually emphasized in the edition by thick double bars separating the variations from one another, might seem to offer built-in opportunities for an ensemble to regroup where necessary. In addition, this movement contains three changes of time signature and four of key signature, all of which are in this instance entirely congruent across all four parts (these changes also correspond mostly to the beginnings of variations). That Rousselot should have chosen to place letters A (bars 19/20) and E (bars 93/94) before the final bar of the theme and the corresponding bar in Variation 4 might therefore seem perverse. Letters B (bars 36/37) and D (bars 74/75), meanwhile, correspond to the penultimate bar of the theme, while C (bars 46/47) marks the beginning of the second half of Variation 2 and F (bars 111/112), finally, marks the fourth bar of the final variation that begins in bar 109. Something similar can be observed in the other large-scale variation movement, op. 131 no. 4. The rehearsal letters for this quartet will be discussed from a different perspective later; but in the present context it is relevant that of six letters supplied for no. 4, which is just as sectionally laid out as the slow movement of op. 127, one is placed before the penultimate bar of the theme (bar 31) and another before the corresponding point in Variation 3 (bar 128), while a third marks the midpoint of Variation 6 (bar 203). The others occur at mid-phrase in Variations 2 and 4 (bars 77/78 and 150/151, respectively) and at the onset of the cadenza-like variation beginning in bar 220.

To the extent that one can draw conclusions from this relatively small and admittedly not entirely consistent body of evidence, it seems possible that in identifying regrouping opportunities for the ensemble just before the end of the theme and certain variations, Rousselot was concerned with questions of continuity from one section of the movement to the next. These movements are replete with ‘bars that [are] structurally important, reflecting a change of texture or new theme’, as Cooper puts it; but rather than needing to mark these already obvious moments in the players’ parts, the rehearsal letters offered points from which the ensemble could work precisely at negotiating those changes successfully. Again, turning to the third movement of op. 130, one might wonder why bars 35/36 was chosen for Letter B rather than the tonic reprise of the main theme at bar 38, or bars 78/79 for Letter D when the following bar 80 brings all four instruments together on a root-position B♭ major triad. ⁵⁸ If, however, the very articulation of that triad (dramatized by one of Beethoven’s characteristic series of dynamic inflections) or the earlier tonic reprise were the important issue from the performers’ point of view, it made sense to identify a somewhat earlier gathering point from which the moment in question could be approached: Marx’s admonition that ‘true artistic training, and long practice for the most highly trained and gifted, are all required before one voice accommodates itself freely and flexibly to the others’ comes to mind. Put simply, Rousselot’s unique and individual rehearsal numbers and instrumental cues are perhaps best understood not merely as guides for future users of his edition but more fundamentally as the trace of a particular rehearsal and performance tradition for the late Beethoven quartets associated with him and his fellow Beethoven Quartett Society colleagues in London in 1845 and 1846.

Rehearsal Letters II: The Case of Op. 131

The idea that Rousselot’s edition captures a trace of a specific performance tradition – or better, perhaps, the repeated musical interactions of a specific group of performers in a particular context – finds especial resonance in the case of op. 131. Once again, though, we begin with Karl Holz. In a conversation with Beethoven in late August 1826, Holz set down the following concerning op. 131:

Must it be played right through without stopping?

But then we won’t be able to repeat anything!

When are we supposed to tune?

Before the Presto.

We must play it on a cool evening.

Some decades later Holz corroborated this exchange in material from his diaries supplied to Wilhelm von Lenz. It had been Beethoven’s intention, he wrote, that the seven Stücke of op. 131 were to be played without a break, irrespective of the need for the players to retune or of potential audience fatigue. All that Beethoven would concede was the possibility of a short break following (and certainly not preceding) the Presto, no. 5, in which case the two transitional bars (497–98) connecting this to the following Adagio quasi un poco andante, were to be left out. ⁵⁹ ‘Before the Presto’ in the 1826 conversation might seem to contradict Holz’s later account, although there is no way of knowing the context of the earlier remark, or Beethoven’s reaction to what may have been a suggestion by Holz, whose later insistence that the break must be after and not before no. 5 is perhaps telling. However, he went on to lament that Viennese audiences were not yet up to experiencing the whole quartet without the usual breaks between movements, so that the Hellmesberger Quartet divided it into four sections (nos. 1–2; nos. 3–4; no. 5; nos. 6–7). ⁶⁰ To this, von Lenz added that in St Petersburg Ferdinand Laub’s quartet played op. 131 with a break both before and after the Presto; the former of these was particularly disruptive, he felt, and he was insistent that the only proper performance of the work was one with no breaks whatsoever. ⁶¹

Von Lenz’s remarks, along with Holz’s contribution, appeared in 1860 and thus have no direct bearing on Rousselot’s edition of op. 131. Nonetheless, Rousselot’s rehearsal letters tell their own story. As in the Schlesinger 1829 ‘Nouvelle Édition’ and earlier editions, the seven Stücke are not numbered, something which can only have emphasized the unconventional continuous appearance of the quartet for its earliest performers. ⁶² Uniquely, however, the sequence of rehearsal numbers does not always commence from A at the beginning of each Stück. Instead, the sequence A–L (I is omitted, as in Holz’s sequence for op. 133) spans nos. [1]–[4], after which no. [5] is lettered A–D, no. [6] is clean, and no. [7], finally, is lettered A–G. The two new beginnings at A suggest the arrangement later favoured by Laub, as reported by von Lenz. But that suggestion is contradicted by Ivan Mahaim’s claim that at the Beethoven Quartett Society performances op. 131 was ‘played without interruption from beginning to end’. Mahaim gives no evidence for this claim, which he makes in connection with an observation about the Beethoven Quartett Society programme: ‘for Op. 131, the excerpts from each movement are extremely lavish, with a total of twelve illustrations’. ⁶³ In fact the original lithographed programme (Figure 3) includes thirteen excerpts, separated from one another by double bars and laid out on five systems. System 1 presents a seven-bar incipit for no. [1], followed by the closing bar (121), with crotchet upbeat, and then an eight-bar incipit for no. [2]. System 2 gives incipits for no. [3] and the theme and Variation 2 of no. [4]; representation of this number continues with incipits for Variations 3, 4, and 5 on system 3, followed on system 4 by incipits for Variation 6 and the C-major Allegretto (bb. 231ff.), and then the final bar (277). Attacca, written vertically across the system, follows the double bar; after a brief space, a nine-bar incipit of no. [5] completes system 4. System 5 then presents the opening and concluding bars of no. [6], and lastly a five-bar incipit of no. [7]. Systems 1, 2, and 3 are all supplied with a concluding marginal ‘Segue’ direction.

Figure 3. The Beethoven Quartett Society [Five programmes] ([London]: 1845). London, British Library, M. d. 54.

It is difficult to know exactly how to interpret all this. The layout of systems 1–4 clearly points to the continuous performance of nos. [1–4], as suggested by Rousselot’s sequence of rehearsal letters; meanwhile, the attacca (of course not present in Rousselot’s or other editions) following that latter number clearly speaks for immediate continuity with the Presto, no. [5], in line with Mahaim’s claim but contradicting the new letter sequence in the edition. Moreover, one begins to wonder why it was felt necessary to include the attacca following no. [4] at all rather than allowing the Presto to follow on immediately, as for example in the case of nos [1–2]. There is even a visual contradiction of sorts between the attacca and an ensuing gap before the Presto, which must have contributed to the necessity of extending this fourth system a little beyond the length of the first three: the sense of a new beginning with the Presto is hard to avoid. As for the Presto itself, it is notable that whereas care was taken to indicate the opening and closing bars of nos. [1], [4], and [6], in order to direct the listener from one number to the next, the same opportunity was not taken here, so that Beethoven’s own explicit attacca direction in the score goes unrepresented. The lack of a ‘segue’ at the end of the fourth system further gives the impression of a break between this and no. [6]. Reading the programme ‘précis’ of the quartet together with Rousselot’s rehearsal numbers, the evidence seems to point reasonably towards a performance that allowed a break either side of the Presto, even if an ideal of continuous performance was conveyed by the performers’ indicated attacca on one side and Beethoven’s identical instruction on the other. ⁶⁴

IV

In his prefatory tribute to Alsager in the Violin I part of his complete edition, Rousselot had emphasized his efforts to render the edition ‘as perfect as possible for the execution of the composer’s intentions’. And as we have seen, Cocks and Co. lost no opportunity to trumpet their own intention to publish an edition of the Beethoven quartets ‘in such a correct and elegant form as will surpass all other editions’, so that performers and audiences might have untrammelled access ‘to the perfect knowledge of Beethoven’s exalted genius’. ⁶⁵ The contested nature of such claims from the perspective of twenty-first-century performance aesthetics (at least Rousselot and Cocks avoided the fraught term ‘authenticity’) need not concern us here; ⁶⁶ but we might nevertheless note that Rousselot’s project was effectively the creation of something like an Urtext edition in all but name, even if, despite his ‘great musical knowledge, and research, and industry’, ⁶⁷ he could not in 1846 have been in a position to carry out such a task.

Over a century and a half later, in the Preface to the Henle study score of op. 132, Emil Platen could write in terms almost identical to Rousselot’s: his edition follows the ‘stated precept’ of the new Beethoven Gesamtausgabe in ‘producing a musical text that “reflects Beethoven’s intentions as accurately as possible.” The degree to which a text can approach the composer’s intentions depends on the number and the value of the surviving musical documents.’ ⁶⁸ If one were to choose a single detail of Platen’s edition as a litmus test for accuracy, it might be bar 246 of the first movement in the Violin I part, which reads c′′–e′′′–e′′′–f′′′. This reading corresponds to the corrected state of Beethoven’s autograph: having at first written c′′–e′′′–g′′′–a′′′, Beethoven corrected the third and fourth pitches to e′′′–f′′′ in pencil, supplied the letter names beneath the notes, and made a marginal note, ‘X Berlin’, indicating that the correction should be incorporated into the edition being prepared there by Adolph Martin Schlesinger. But the Berlin Schlesinger edition somehow mangled Beethoven’s instruction to give the reading c′′–e′′′–g′′′–f′′′, which was then adopted in virtually all published editions (including Maurice Schlesinger’s 1827 Collection complète, 1829 ‘Nouvelle Édition’ and the ‘old’ Breitkopf & Härtel Gesamtausgabe of 1862–65), performances, and recordings down to Platen’s edition of 2002: all this despite the fact that Holz had drawn attention to the error as early as 1857. ⁶⁹

It is hardly surprising that in this instance, as in the others discussed earlier, Rousselot and Cocks and Co. failed: their edition, dependent as it must have been on ‘the number and the value of the surviving musical documents’ to which Rousselot could have had access, is not error-free. In the case of op. 132 only the autograph score (or Holz himself) could have revealed the correct reading in bar 246 of the Violin I part in the first movement. But honour where honour is due: what Rousselot actually achieved was not a proto-Urtext edition but something else. His edition of the late quartets may be regarded as a particular kind of performers’ edition, not in the later nineteenth-century sense of the ‘great pianist’ edition of the Beethoven sonatas, for example, but rather in the sense that, as alluded to earlier, its unique provision of instrumental cues and rehearsal letters preserves a trace of the interactions in extensive rehearsal and performance of himself, Henry Hill, Prosper Sainton, and Camillo Sivori.

‘The nature and quality of the actual performances – lost to thin air, unlike ones of the gramophone era – can of course never be recovered’; ⁷⁰ but in Rousselot’s edition is preserved something of the unrelenting work that made those performances not only possible but convincing to their audiences, and in so doing entrenched for ever the ‘hono[u]r’ due to Beethoven’s late quartets.