Aug 29, 2013

The Continuing Mutilation of Schubert's "Der Leiermann"

In November 1999, Naxos released a recording of Schubert's "Winterreise" D. 911, performed by baritone Roman Trekel and pianist Ulrich Eisenlohr. This CD, which was the first volume of Naxos's "Deutsche Schubert-Lied-Edition" series, generally received positive reviews. The critics deservedly praised Trekel's singing and only had nice things to say about Eisenlohr's playing. The only person known to me, who – contrary to all the supposedly expert critics – immediately realized that something was wrong with this recording and publicly spoke out about it, was retired college professor of Criminal Justice and avid Fischer-Dieskau aficionado Dr. Celia A. Sgroi, who, on 19 January 2000, posted the following on the "Lieder, Melodies, Art Songs in any language" mailing list, maintained by the University of Houston Listserv:
The pianist, Ulrich Eisenlohr, who delivers competent, and sometimes quite imaginative, playing throughout the cycle, goes completely off the rails in "Der Leiermann." His hurdy-gurdy is more loudly and jarring discordant than any I have ever heard. It gives the song a strangely "modern" sound that is interesting for a moment, but it palls very quickly and the disadvantages are enormous: the listener ends up completely distracted from the singing, and the dischords in the piano make it sound as if the singer is off-pitch. If this is the judgment of one of the initiators of this complete Schubert edition, Naxos may be in considerable trouble. At any rate, I feel sorry for Trekel, who is singing his heart out at that point and all in vain.
Eisenlohr's interpretation of the song "Der Leiermann" is very strange indeed. He apparently thought to be the first pianist who realized how Schubert really meant the piano accompaniment of this song to be performed. Eisenlohr not only plays the short appoggiatura in the left hand as an acciaccatura together with the principal note (he plays the dissonant note a tick earlier than the half notes only in the first two bars), he is also convinced that Schubert wanted the dissonant note to be repeated on the beat of every bass chord throughout the whole song in the way of a simile or segue. The first two chords in Schubert's score have an e# as appoggiatura, but the following ones have none. There is no "simile" or "segue" written in the score.

The first page of Schubert's "Der Leyermann" in the autograph b minor version (US-NYpm, Cary 215, Part II, fol. 15r)

The result of Eisenlohr's bizarre idea is fatal. At the beginning of each measure, all the way through the lied, we now hear a dissonant chord in the bass which reaches a peak of tasteless cruelty at that moment when we realize that, for the first time ever, Schubert's song cycle now ends on a dissonant chord. In a personal e-mail, sent to me on 1 May 2000, via Roman Trekel, Ulrich Eisenlohr defended his interpretation as follows:
As you might perhaps know, the "Leier" is always played with bordun-chords. If some of these bordun-strings are not welltuned,-which is quite likely in winter-time and outdoors,-they will sound a little bit "dirty" ; that is why I play the appogiatura on the beat. There is nothing less meant by Schubert than a correct appogiatura, done by a well-aducated musician. Can we presume that the Leiermann will stop his playing after 2 bars to tune his bordun-strings again? Probably not: he will continue as he has begun: "dreht, und seine Leier steht ihm nimmer still." So, why change the playing of the left hand in the third bar? It is just meant like "simile" or "segue", which is a familiar instruction in classical and romantic music, often not even written by the composers, because they trusted in intelligent interpretors who would know how to read and understand the music. Although I don't think that my kind of playing distracts the listener from the voice - even if it did: would it not be the pendant to the fact, that the singer is distracted from himself, his sorrous and pains, by the playing of the Leiermann? Hope I could calm you just a little bit with these explanations. Kind regards, Ulrich Eisenlohr
Eisenlohr firmly rejected the suggestion that by his curious interpretation he may have branded his illustrious predecessors as unintelligent musicians, who, for over 100 years, have been unable to understand Schubert's score. Some of these pianists, who (owing to obvious lack of musical intelligence) failed to grasp the whole "never standing still hurdy-gurdy" idea, are Gerald Moore:

Sviatoslav Richter:

Alfred Brendel:

Murray Perahia:

Daniel Barenboim:

and Leonard Borwick:

Making musical pieces sound "like they have never sounded before" is one of the most effective artistic gimmicks in the current classical music business. In their evaluation of the quality of recordings, some critics nowadays seem to apply the following simple rule: "If I've never heard it played like this before, it must be brilliant!" It seems that the commercial success of some flawed "historically informed" performances is mainly based on this crazy paradigm, and the quite obvious fact that most positive reviews of CDs in glossy music magazines are nothing but paid ads of the recording industry. A perfect recent example is René Jacobs's recording of Mozart's Symphony No. 38 "Prague", which was released in 2007, by Harmonia Mundi. The blurb of this recording promises a revelatory listening experience: "The guiding principle of this interpretation is clarity of texture [...] It forsakes the 19th-century symphonic tradition for a quite different style of rhetoric." Such an announcement cannot bode well. And accordingly, Jacobs forsakes what he considers "the 19th-century symphonic tradition" by ruthlessly botching the long appoggiaturas in bars 17-28 of the symphony's slow introduction.

Bars 16-19 of the strings in the slow introduction of Mozart's symphony K. 504

The musicologist Theodor Kroyer considered the interpretation of the aforesaid appogiaturas so obvious that he had them written out in both of his Eulenburg pocket score editions (Edition Eulenburg No. 446).

Bars 20-25 of the slow introduction of K. 504 in Theodor Kroyer's edition for Eulenburg

Instead of having them played as regular 1/32nds (as intended by Mozart), Jacobs turns the appoggiaturas into short grace notes, because he obviously considers himself more knowledgable in matters of eighteenth-century performance practice than the composer himself. He also misses the point that the five-note motif he is messing up in the introduction is a brilliantly applied thematic inversion of the recurring motif which, from bar 55 on, features prominently in the following Allegro. And yet, this kind of breathtaking musical ignorance goes completely unnoticed, while the critics are raving. Some of them fall for the advertisement ("a quite different style of rhetoric") and the others cannot even read music and do not know the piece to begin with. Jacobs's recording of K. 504 was awarded the "Diapason d'or Arte" from the French magazine Diapason, as well as the highest possible ratings from Classica Répertoire, and This method of fooling gullible (and totally unqualified) critics works in every genre of classical music, including Schubert songs.

In September 2009, Harmonia Mundi released an eagerly awaited recording of Schubert's "Winterreise", performed by tenor Mark Padmore and pianist Paul Lewis. The British music critics were giddy with excitement. John Steane, critic of the magazine Gramophone ("The world's authority on classical music since 1923") raved as follows:
Ah, this journey! How many have made it, sincerely and imaginatively, two setting out as nearly as possible as one! So many on records too, following the elusive track as with torchlight concentrated upon it. Yet, of all, I cannot think of one (not even Fischer-Dieskau in his 1965 DG recording with Jörg Demus) that leads more faithfully to the cold comfort of its end. And when we get there in this performance, what an end it is! [...] On we go, lulled and tormented by the magic music-box of "Frühlingstraum", till the tragic chord before "so elend nicht" in "Einsamkeit" brings a dreadful reality into focus. The deceptive sweetness of "Die Krähe", the giddy disorientation of "Letzte Hoffnung", the subdued feverish excitements of "Täuschung" find an almost holy stability in "Das Wirtshaus", but still the external world exists, felt as almost an intrusion in "Mut". And soon we meet the organ-grinder. And his secrets must on no account be revealed by reviewer or arts-gossip. And the listener must wait, out of respect to this marvellous partnership of Mark Padmore and Paul Lewis, until time can be taken for it, alone and uninterrupted, to accompany them on the journey through to its unearthly end.
On ClassicalNet the critic Mark Sealey found the following words of praise:
This excellent new Winterreise is one for the twenty-first century. Understated yet passionate; reflective yet not self-indulgent; spare, yet rich in the wonderful melodies in which the cycle abounds, it succeeds in meeting many quite disparate expectations, yet makes no compromises. To sing about recollection, lost love, death and resignation is actually harder than merely to sing mournfully, slowly and wistfully. The two performers here (Mark Padmore, tenor, Paul Lewis, piano) have produced an excellent embodiment of the songs' varying (for variety is key) moods and outlooks.
The London Evening Standard's Nick Kimberley (who also thinks that "Winterreise is about the voice, not the piano") stated: "Schubert couldn't be better served […] Padmore's great gift, apart from his prodigious technical ability,whether to float a line with perfect legato or to enter pianissimo at the top of his range, is to sing from the soul." The recording was awarded the highest honors. It got an "ffff" rating from Télérama, received Gramophone's "Editor's Choice", and "Recording of the Month" laurel, an "IRR Outstanding" from the International Record Review, and eventually won the 2010 Gramophone Award in the category "Solo Vocal". And yet, Padmore's and Lewis's recording is marred by grave musical flaws for which only the pianist bears responsibility. I will not even delve deeper into the embarrassing fact that Lewis – being obviously unaware of the scholarly literature on this topic – still mistakenly aligns the dotted figure in the left hand with the triplet in the right in the songs "Wasserflut" and "Irrlicht". This is not a minor musical detail that can be performed ad libitum. In the Bärenreiter complete edition, Walther Dürr claimed that "The notational style in the sources suggests that triplets and dotted eighths should be treated as rhythmically identical throughout the lied." Dürr was wrong. The whole issue has convincingly been settled once and for all by the American pianist David Montgomery, not only in his article "Triplet Assimilation in the Music of Schubert: Challenging the Ideal" in Historical Performance vol. 6/2 (1993), pp. 79–97, but also in his excellent book Franz Schubert's Music in Performance (2003, Hillsdale: Pendragon Press). In his 1971(!) recording of "Winterreise" with Hermann Prey, the late Wolfgang Sawallisch already showed how the piano introduction of "Wasserflut" is played correctly

As far as the piano part of "Der Leiermann" is concerned, Lewis takes Eisenlohr's mistaken concept to a more extreme and even crueler level. While Eisenlohr, in the first two bars, concedes the dissonance at least a little grace note quality, Lewis plays the dissonant grace note together with the fundamental and the fifth and then releases the grace – but much too late to make it the 1/16 note that is notated in the first two bars. This is the method of playing acciaccature that Artur Schnabel used to recommend: two notes together and then release one of them quickly. But this is not what is written in the score and it does not produce the hurdy-gurdy sound that Schubert was trying to convey. Lewis shows no mercy at all. He ruthlessly repeats the dissonant bass chord up to the final bar of the "Winterreise" which now – in the well-known Eisenlohr tradition – has to end on a dissonant chord (warning to listeners: extremely ugly sound bite!).

Let me explain what Schubert had in mind and why Eisenlohr's and Lewis's interpretations are completely untenable from a a musical point of view. Of course, Schubert relied on the "intelligent pianist", but in no way does this mean that Schubert himself was not intelligent enough to unambiguously write down what he wanted the pianist to play. Hurdy-gurdies have multiple drone strings which give a constant pitch accompaniment to the melody, resulting in a sound similar to that of bagpipes. These drone strings are tuned in fifths and fourths, but never in diminished fifths (as played by Eisenlohr and Lewis). Schubert achieves the bagpipe effect of the drone strings with long perfect fifths in the left hand.

The 1/16 grace note, a semitone below the fifth, imitates the short slur caused by the transient state of the string between the first motion of the hurdy-gurdy's wheel and the moment the string reaches its maximum (i.e. continuous) frequency. The grace note in the second bar suggests that the wheel has come to a standstill after the first bar and then is being cranked again. From the third bar on, the short slur (the appoggiatura) is no longer audible, because the wheel must be moved steadily to produce the melody on the hurdy-gurdy's melody strings (i.e. the pianist's right hand). Eisenlohr's argument that his continuing dissonant note shows the "never standing still" of the hurdy-gurdy is the exact opposite of what is going on in Schubert's score: fact of the matter is that the absence of the grace note after two bars proves the steady movement of the hurdy-gurdy's wheel. The dissonant grace note does not represent a third drone string or "a hurdy-gurdy out of tune" (as proposed by Eisenlohr). It cleverly imitates a very short and purely physical event in the course of "warming up" the instrument. There are also esthetic reasons as to why there are only two appoggiaturas in the song. Schubert doubtlessly realized that the repetition of the grace note throughout the whole song would heavily distract from the singing voice and would result in exactly the boring unmusical overkill that Eisenlohr and Lewis achieve with their merciless dissonant redundancy. Schubert knew very well how to repeat a figure if he wanted it. Contrary to a hypothesis, put forward by Nigel Nettheim in The Schubertian (No. 31, January 2001, p. 2f.), there is no "implied simile" in "Der Leiermann". There is no need to even touch the issue of ending the song on a dissonant chord. Not only is it unthinkable from a historical point of view, it is the kind of musical tomfoolery that actually is evidence that Schubert did not want the appoggiatura to be applied in more than two bars: if he had really implied a simile, he would have indicated to end it before the final chord.

None of the above problems in Paul Lewis's performance were ever noticed and addressed by the critics who awarded this particular recording the 2010 Gramophone Award. They seem to have followed Eisenlohr's principle that "there is not just one proper way to read and interprete a work of art" and they were obviously under the impression that Lewis had found some kind of Philosopher's Stone of Schubert performance. Of course, there is a grain of truth in Eisenlohr's point of view. But we must never forget that the number of improper ways to read a work of art is infinite. I have been told that the pianist Wolfram Rieger has been following Eisenlohr's and Lewis's example in his live performances of "Winterreise", together with the singer Thomas Hampson, by droning out the poor "Leiermann" with sixty-one consecutive dissonant bass chords. He will certainly not be the last pianist to do so.

© Dr. Michael Lorenz 2013. All rights reserved.

Updated: 26 September 2023

Update (20 January 2018)

In his next recording of Winterreise, released in January 2018 by Harmonia Mundi, Mark Padmore found yet another willing accomplice in Kristian Bezuidenhout who (performing on a period instrument) ruthlessly mishandles the piano part of "Der Leiermann" in the fashion described above. It seems that Padmore is the driving force behind this mistaken idea that makes his accompanists suddenly lose all their musical common sense.

Update (13 April 2018)

The next proof that stupidity can go viral among musicians was presented by the English pianist Christopher Glynn who, on his recording of "Winterreise" with the baritone Roderick Williams (in an English version titled Winter Journey), mercilessly demonstrates that Schubert's acoustic depiction of a hurdy-gurdy is much too sophisticated to be understood and performed correctly by today's mortals.

Update (16 December 2021)

One should not believe that the reservoir of Drehleier-nonsense will ever be exhausted. More recent recordings of Schubert's "Winterreise" also offer curious abuse of the piano part of "Der Leiermann". On Markus Schäfer's recording of "Winterreise", which was published in June 2021, the pianist Tobias Koch presents a bizarre combination of Eisenlohr's approach and his own arbitrariness: he brutally plays the dissonance through the entire song, including on the final chord, but demonstrates a bizarre inconsistency in the introduction where he sometimes holds the grace note into the chord and sometimes plays it in unison with the chord. Note the final chord in this recording where human stupidity again comes to audible fruition.

In 2015, the tenor Daniel Behle published a double CD on the Sony Classics label, entitled "Winterreisen" where he combined his own arrangement of the cycle for voice and piano trio with the "original version" for voice and piano. Unfortunately, Behle's pianist Oliver Schnyder could not untangle his mind from Behle's "arrangement" (recomposition) of "Der Leiermann" where, at the beginning of the song, a grossly unmusical "grace note" from a third above is introduced by the ruthless arranger (composer). Accordingly (and unfortunately), Schnyder also plays this grotesque note in the supposed original version of "Der Leiermann" – which does not come from Schubert but from Behle. One has to hear it to believe it.

This reminds me of a question I always ask on such occasions: where are the Schubert police when you really need them?

Update (12 February 2022)

Fortunately, only now, it came to my attention that in a more recent recording of "Winterreise", Roman Trekel found yet another accompanist in Oliver Pohl who proudly demonstrates that he has no idea how a hurdy-gurdy works and how to correctly play a Schubert score. Pohl's imitation (parody?) of Ulrich Eisenlohr's misinterpretation of "Der Leiermann" is availablable in a complete recording of "Winterreise" on YouTube.

On 11 February 2022, Alpha Classics released Bernhard Appl's recording of "Winterreise" which had been long awaited by servile journalists who in the past had heaped praise on the German baritone's recordings. Appl's accompanist on this new CD is the South African pianist James Baillieu. One could write a detailed review of this CD, addressing all of Appl's vocal problems, which reveal themselves in the interpretation of Schubert's lieder. Appl has a quite pleasant voice which sometimes sounds very nice in slow tempi. However, the moment he has to sing fast melodic material, his intonation slips completely and he produces (usually slightly sharp) notes that leave the listener in the dark as to which key he is actually singing in. So far, this has been particularly unpleasant in his interpretations of up-tempo arias by Bach, such as "Großer Herr, o starker König" from BWV 248. Not surprisingly, Appl runs into similar problems in the fast lieder of "Winterreise" and it is certainly no coincidence that a promotional video on YouTube only features the lied "Gute Nacht". In "Der stürmische Morgen" he is close to tipping over into an off-pitch Sprechgesang that merely outlines the melody. But that is a side issue here. As he has already shown in 2021, in a live performance with Mark Padmore, the pianist James Baillieu is completely infected with the Eisenlohr virus and plays the well-known dreaded dissonance in the bass in every bar of "Der Leiermann". And of course, he also applies this unmusical tomfoolery in the final chord. What an unrecognized avant-gardist and revolutionary Neutöner this Schubert was! It can be safely presumed that the paid reviews in glossy music magazines will be ecstatic in their praise of this new release.

Aug 24, 2013

Not Mozart, Not Zoffany. So . . . What?

The following is a guest post by Dexter Edge, an internationally known Mozart scholar who currently lives in Ypsilanti, Michigan.  He can be contacted at dexedge(at) A pdf of a revised version of this essay (2016) can be downloaded at


On Tuesday, 18 June 2013, Librairie Henri Godts in Brussels auctioned a large ink drawing of an eighteenth-century orchestra, ranged in a semicircle around a boy seated at a harpsichord, with another boy standing just to the left of the keyboard. The drawing is described as lot 24 in the auction catalog, where it is given the title “Le Concert de musique (anciennement: Le jeune Mozart en concert)” (The Music Concert, formerly: The Young Mozart in Concert), and said to have been “anciennement attribué à Zoffany” (formerly attributed to Zoffany). The auction house advertised the item with the following image, from an online slideshow, consisting of a collage of five segments from the drawing, with a caption that still attempts to trade on the Mozart association.

The pre-auction estimate for the drawing was €8,000–10,000, and the hammer price was €11,000 (around $14,600 at the current exchange rate). Had the drawing been firmly attributed to Zoffany, the price would undoubtedly have been much higher. Had both attributions been firmly accepted, the price would likely have been in six figures.

About a week before the auction, I was asked by two friends—a Mozart collector and an antiquarian dealer—for my evaluation of the drawing. Using just the secondary literature that I had at home, and with the help of relatively high-resolution images of the drawing, I quickly concluded that it had nothing to do with Zoffany, and therefore almost certainly nothing to do with Mozart—for as we shall see, the case for the latter depends almost entirely upon the case for the former. Moreover, closer inspection of the drawing revealed obvious flaws in draftsmanship that further undermined the case for Zoffany and cast doubt on the drawing’s artistic merit. What remained, then, was an anonymous drawing of indifferent quality that might, indeed, be of some interest to students of eighteenth-century performance practice....but this potential interest did not, in my estimation, justify the estimate of €8,000–10,000. My friends did not bid on it. 

As it turns out, the drawing was purchased by the Stiftung Mozarteum Salzburg, which announced the acquisition in a press release dated 8 July 2013. The press release makes clear that the Stiftung does not accept the attribution to Zoffany, but it nevertheless describes the drawing as showing “an orchestra, perhaps with Mozart.” It also calls the drawing the “work of a high quality illustrator,” and claims that it is of “major musicological significance” as evidence for eighteenth-century performance practice. As we shall see, there is ample reason to question all of these claims.

This drawing first came to public attention in 1991 as part of the exhibition Mozart. Bilder und Klänge in Salzburg. A color facsimile of the drawing is included in the exhibition catalog (item 146, p. 150), with the unequivocal caption “Konzert mit Wolfgang Amadeus Mozart.  Zeichnung von Johann Joseph Zoffany” (Concert with Wolfgang Amadeus Mozart. Drawing by Johann Joseph Zoffany); however, the catalog text for the item, by Geneviève Geffray, seems to take a more agnostic position, making no explicit argument for either the artist or the purported subject. Geffray does, however, note that the Mozarts met Zoffany in London during their stay there in 1764–65: Leopold’s travel notes (about which more later) contain the entry “Mr: Zoffani Mahler” among the long list of people they met during their stay in the city.

In spite of this hint of equivocation, the drawing soon attracted scholarly attention. Already in November 1991 it was used to illustrate an article in Early Music (Lockwood, 1991), even though that article made no reference to the drawing. The drawing has subsequently been discussed several times in the Mozart literature, most often accompanied by a facsimile. Although the Mozarteum press release states that “the reference to Mozart has been intensely questioned...because the leaf is neither signed nor dated,” the published literature on the drawing largely fails to question either attribution, to Zoffany or to Mozart. In fact, rather than questioning the drawing, there has been a tendency in the literature toward ever more elaborate confabulation about it.

When I learned that the drawing had been purchased by the Mozarteum, it seemed worth fleshing out my earlier research and presenting it here. To my surprise, this subsequent research led me into a tiny wonderland of scholarly fantasy, where evidence, observation, and logic played little if any role; an echo chamber in which most subsequent writers accepted uncritically what had been written by previous ones, while adding embellishments of their own. 

Consequently, this post is not merely an examination of the drawing and a critique of the attributions: the attributions are laughably weak, and the drawing is, in the final analysis, of minor importance. But I hope that the wider story may serve as an object lesson in the perils of musicological fantasy. Most of the prior literature on this drawing consists of unsubstantiated claims and flights of fancy, abetted by a failure of careful observation, a failure to question, a failure to approach earlier scholarship with a critical eye, and a failure to take into account directly relevant work in related fields.

In what follows, I will unravel the mythology that has grown up around the drawing. Then, based on a careful examination of its content, I will evaluate the drawing as a source of evidence for eighteenth-century orchestral performance practice. Along the way, I will have the opportunity to touch on a minor Mozart mystery.

The Case for the Attributions

We start with the recognition that this is a case of dual attributions, with two separate claims that should (if one is proceeding carefully) be argued on their own merits: a claim that the drawing is by Johan Zoffany (to use the spelling preferred by art historians), and a further claim that Mozart is one of the figures depicted in the drawing. While most writers seem to assume or to hope that both claims are true, it is entirely possible for the drawing to depict Mozart, but not be by Zoffany, or for it to be by Zoffany, and not depict Mozart.  Or, of course, it may not depict Mozart and not be by Zoffany. In fact, this last should be the default assumption in cases like this one.

Astonishingly, no previous writer in the past twenty-two years has mentioned the source of the attribution to Zoffany, which was, to my knowledge, first discussed in print in the Godts auction catalog published earlier this year:
Mention ms. signée au verso: «M. George Granau directeur de la Galerie Cassel croit ce dessin de Zoffgny [sic] de Ratisbonne. Il y a peut être à Parme (Paris?) oeuvres de lui analogues à celle-ci. M. Laurence Bonyon [sic pour Binyon] conservateur de la section des estampes de dessins du British Museum dit que ce dessin est certainement allemand quelque peu dans le genre de Chodowiecki. de cet artiste plus que de tout autre. Sujet Mozart enfant (?) (voirs leurs lettres juillet 1914)».
Manuscript annotation on the verso: “M. George Granau, director of the Galerie Cassel, believes the drawing is by Zoffany of Regensburg. There are perhaps in Parma (Paris?) works analogous to this one. M. Laurence Binyon, conservator in the prints section of the British Museum, says that this drawing is certainly German, somewhat in the style of Chodowiecki. Of this artist, more than any other. Subject, Mozart as a child (?) (See their letters of July 1914).”
Thus in 1914 (if this description is accurate), there were two conflicting opinions, perhaps solicited by the owner of the drawing at that time (or perhaps a dealer or a prospective owner), one attributing the drawing to Zoffany, the other finding it to be in a style more reminiscent of Chodowiecki. Art historian Georg Gronau, a specialist in Italian Renaissance art and the source of the Zoffany attribution, is unlikely to have been a Zoffany expert, simply because there were no real Zoffany experts in 1914. Knowledge of Zoffany’s oeuvre was very poor at that time, and many paintings were attributed to Zoffany that have subsequently been shown not to be by him.

So Zoffany’s style was still poorly understood in 1914, and no one was in a position to make reliable attributions to Zoffany on the basis of style alone. And, as we shall see, very few drawings by Zoffany survive, something Gronau clearly did not know. As William L. Pressly explains in his foreword to Penelope Treadwell’s 2009 biography of Zoffany:
….the artist’s papers and many of his drawings were burnt in 1832. Zoffany’s much younger second wife and one of his daughters died in the cholera epidemic in London that year, and because the origins of the illness were not understood, the contents of the family home were incinerated in a misguided attempt to prevent the spread of the disease. (vii-viii)
Most likely, then, Gronau was making a wild guess. But on this flimsy basis, the Zoffany attribution has embedded itself like a tick in the Mozart literature. The notion that the drawing depicts Mozart also derives from this 1914 annotation, but again this seems to have been merely a guess—or perhaps wishful thinking.

The catalog of the Salzburg Mozart exhibition is silent about Gronau’s attribution, and no subsequent writer on the drawing has mentioned it. Most later writers have glossed over the question, implicitly accepting Zoffany as the artist, albeit with dutiful question marks. Rampe (1995) simply assumes the attribution to be correct, and it remains unquestioned in Edge (1996), Brunner (1998), Irving (2003), Basso (2006), the Mozart-Handbuch (2007), and Gétreau (2010). (In their defense, the identity of the artist was largely irrelevant to Edge and Irving, who were concerned with the drawing’s potential relevance to concerto performance practice.)

Daniel Heartz (1995), one of the first scholars to discuss and reproduce this drawing after the Salzburg exhibition, is also the only one so far to argue for the Zoffany attribution on stylistic grounds:
The care with which the artist has depicted the instruments and the way they are played, also the physiognomies of the players and observers (surely meant to portray real people) do indeed point to Zoffany, who was famous for his lifelike portraits of individuals within larger groups. (617)
This last point is certainly true. Among Zoffany’s most famous paintings is The Tribuna of the Uffizi (1772–7), now in The Royal Collection of Her Majesty Queen Elizabeth II.

Every person (and indeed every painting and sculpture) in this painting is precisely identifiable (see Johan Zoffany RA: Society Observed, 2011, catalog 53, pp. 230–32, which includes a key to the painting). In fact, one of the figures depicted in the painting is Zoffany himself (fourth standing figure from the left, behind the Niccolini-Cowper Madonna of Raphael), who inserted himself, Hitchcock like, into several of his group paintings. Similarly, every figure in Zoffany’s The Portraits of the Academicians of the Royal Academy (1771–72) can be identified, and again the painting includes a cameo appearance by the artist, in the lower left corner (op. cit., catalog 44, pp. 218–19, again with a key; see also the splendid larger reproduction across a two-page spread, pp. 26–27). Even a casual comparison with the promotional image from the Godts website shows that the most strongly characterized faces in the drawing are mere cartoons compared to Zoffany’s carefully-wrought likenesses.

To find a plausible point of comparison between the orchestra drawing and Zoffany’s large group paintings, we need to turn to those in which some or all of the figures are anonymous—in other words, those in which some or all of human figures are not evidently intended to represent actual people. An example is Colonel Mordaunt’s Cock Match, c. 1784–88, painted during Zoffany’s sojourn in India (op. cit., catalog 86, pp. 270–71; see also the larger image on pp. 136–37). Here, the Europeans are identifiable (and again Zoffany inserts himself, as the first seated European from the right in the back row). A few of the prominent Indians in the painting are also identifiable, but most seem not to be intended to represent particular people, and their features are more generic or even slightly cartoonish. Even so, the craftsmanship in the depiction of these generic figures is of a much higher order than the haphazard renderings in the orchestra drawing.

Zoffany’s celebrated later paintings of the French Revolution seem to consist entirely of invented figures—for example, the Plundering of the King’s Cellar at Paris, 1794 (op. cit., catalog 104, pp. 294–95; see also the larger image, fig. 37, pp. 42–43).

Yet even here, where Zoffany is at his most “cartoonish,” the rendering of anatomy (for example, the hands) and the depiction of human motion and emotion is much more expertly done than in the orchestra drawing. So the case for the attributing the drawing to Zoffany on the basis of style (or even simple technical competence) does not look very promising.

Heartz was also the first to associate the orchestra drawing with Mozart’s Italian tours, albeit only in the most general terms:
The intriguing idea of Mozart performing harpsichord concertos during his Italian tours is a subject with possible links to the fine drawing attributed to Johann Zoffany, an artist … who made his mark in London, then spent most of the 1770s in Italy.
Heartz provides no evidence that the drawing actually depicts an event in Italy, but one supposes that his speculation was motivated by the apparent age of the boy at the harpsichord, who seems (although the rendering is quite sketchy and the bodily proportions are awkwardly done) to be a teenager. If it is Mozart, then Italy or Salzburg would seem to be the only reasonable possibilities. But Heartz offers no explicit justification for the notion that the boy in the drawing is Mozart. And as we shall see, the drawing does not depict the performance of a concerto.

Siegbert Rampe, the principal fantasist in our story, picks up Heartz’s Italian hint and develops it into an elaborate and detailed backstory. In his Mozarts Claviermusik, published the same year as Heartz’s book (1995), Rampe devotes over one and half pages of tiny print to the drawing (pp. 64–67), accompanied by a rather poor full-page black and white facsimile (p. 66). Rampe’s book was published by Bärenreiter, the publisher of the Neue Mozart-Ausgabe, with a muddy burgundy cover not dissimilar to the familiar color of the NMA itself, thus lending the book an air of authority and solidity that it hardly deserves—but which may have played a role in persuading subsequent writers to accept it uncritically.

Yet Rampe’s discussion of the drawing is an extraordinary hodgepodge: conclusions stated as fact without evidence; leaps of logic and circular argument; and obvious blunders, stemming from a simple failure to examine the drawing carefully or to check basic facts in the Zoffany literature. Rampe begins by stating that the drawing is “wahrscheinlich von J. J. Zoffany, ca. 1770–1775.” No grounds are given for the attribution, and no argument is made for it; it is simply assumed. After noting that Zoffany’s name appears in Leopold Mozart’s travel notes from London, Rampe makes a spectacular leap:
Diese Beziehung zwischen Zoffany und der Familie Mozart stützt zwar die Identifizierung des Clavierspielers auf der vorliegenden Zeichnung.

This relationship between Zoffany and the Mozart family indeed supports the identification of the keyboard player on the drawing in question.
This is extraordinary. Even a moment’s casual thought makes clear that the acquaintance between Zoffany and the Mozarts (no “relationship” has been shown), at best, does not rule out the possibility that Zoffany is the artist and Mozart one of the subjects of the drawing. More than this we cannot say without additional evidence.
The mere fact of their acquaintance provides no positive evidence for either attribution whatsoever.
This is a key point: for the meeting between Zoffany and the Mozarts is the only solid fact in the mythology of the drawing, and it is repeatedly cited as if it were evidence for the attributions. Yet the meeting plays no role at all in the argument for either attribution. It does not even make the attribution to Zoffany (or Mozart) more or less plausible. In the absence of additional evidence to connect the dots (and there is none), it is simply irrelevant. Since the only solid fact is irrelevant, the case has already essentially disintegrated.

The remainder of Rampe’s argument is fundamentally circular. From this point on, he simply assumes that the boy at the harpsichord is Mozart, and this assumption is used in support of his other key point: that the work being performed is somehow related to K. 120 (111a), a D-major Presto in 3/8 that is said to have been written as a finale for a symphony derived from the overture to Mozart’s serenata Ascanio in Alba, K. 111. Rampe relies here on what he calls the “fotographische Genauigkeit” (photographic exactitude) of the drawing, implying that it is the representation of an actual event, and he goes on to describe the scoring of the orchestra, in which he discerns two transverse flutes, two oboes, two horns, a string ensemble, and a harpsichord, with two bassoons reinforcing the bass line. According to Rampe:
Die einzige Instrumentalkomposition Mozarts mit einer Besetzung von je zwei Flöten, Oboen, und Hörnern sowie mit Fagotten (als nicht-obligate Generalbaßinstrumente!) ist die einsätzige Sinfonia D-Dur KV 120 (111a)…

The only instrumental composition by Mozart scored for pairs of flutes, oboes, and horns as well as bassoons (as non-obligato continuo instruments) is the single-movement Sinfonia in D Major, K. 120 (111a)...
On this basis, Rampe suggests (or seems to suggest) that the drawing depicts a performance of K. 120 or Ascanio in Alba in Milan in the fall of 1771—or perhaps in Salzburg after Mozart returned from Italy (a hedge that Rampe fails to explain).

To recapitulate the argument thus far, with implicit assumptions in brackets:
  • The boy at the keyboard is Mozart
  • [Therefore the orchestra is performing a piece by Mozart]
  • The only orchestral piece by Mozart [by the time he was this age] with the instrumentation depicted in the drawing is K. 120
Therefore, the orchestra is performing K. 120 (or the serenata from which it is derived), and Mozart is the boy at the keyboard.
This could serve as a classic example of petitio principii in a textbook on fallacies. But Rampe is not finished.

He recognizes that some of the music in the drawing is legible, mentioning the viola part and the part for first horn. Rampe admits that neither the viola nor the first horn actually match K. 120 as it has come down to us. K. 120 is in 3/8, but the viola part in the drawing is in cut time (Rampe erroneously says common time). But he is not disheartened:
Weder die Viola-Stimme noch die Partie des ersten Horns auf angeführter Zeichnung—beide gut lesbar—stimmen jedoch mit einem der bislang unter KV 111a angeführten Sätze überein, so daß—Mozarts Autorschaft an der aufgeführten Komposition vorausgesetzt—immerhin vorstellbar ist, besagte “Serenata” hätte noch wenigstens einen weitern (offenbar Eröffnungs-) Satz eingeschlossen.
However, neither the viola part nor the part for first horn in the drawing under discussion—both quite legible—match any of the movements previously associated with KV 111a, so that—assuming Mozart’s authorship of the composition being performed—it is at least conceivable that the aforesaid “Serenata” [i.e. Ascanio in Alba] had at least one other (obviously opening) movement added to it.
In other words, the music on the stand doesn’t match a known movement of K. 120, so the work being performed may well be an unknown movement by Mozart. Problem solved.

Let’s look more closely at the music on the stands in the drawing. In fact, at least some music is visible on four orchestral parts, not just the two that Rampe mentions: a full two-page opening of the viola part, an entire page of the first horn part, a small portion of the second horn part, and perhaps most crucially, the melodic opening of the “Flauto primo.” The top of the oboe part is also visible, but no music can be clearly made out on it (apart, perhaps, from one note). Also visible on the music desk of the harpsichord is the far left of a verso page, showing two keyboard systems, with braces, clefs, and a G-major key signature, but no music.

Rampe writes:
[Die Violastimme] trägt die üblichen Bezeichnung alto auf der linken und Viola auf der rechten Notenseite. Im ersten System der linken Notenseite steht Sinfonia und ein weiteres unlesbares Wort...

[The viola part] carries the usual designation alto on the left page of music and Viola on the right page. In the first system of the left-hand page is written Sinfonia and a further illegible word.
This is correct so far as it goes; but the second “word” is not illegible: it is clearly “del,” normally used at the beginning of composer attributions on title pages and headings on manuscript music in the eighteenth century, when these were written in Italian. The word “del” is followed by the beginning of another word or name. Unfortunately, the artist has fudged at this point: one or possibly two letters after “del” are imprecisely hinted at, but there is nothing more. The first two letters may perhaps be the beginning of the abbreviation “Si[g]” (for Signore), which would normally have been expected after the “del.”  Or, the first letter may also be read as an “F” (perhaps “Fi”), possibly the name of the composer.  At any rate, it is not “Mozart.”

The artist has made no attempt to give a clear indication of the composer’s name. Yet, this would have been a most peculiar omission if the drawing had actually been intended to depict Mozart in a performance that had anything to do with Ascanio in Alba in Milan in 1771.

Twenty-two bars of the Viola mp3 part are legible, spread across two pages. The artist has been forced by the scale of his drawing (and the resulting small size of the orchestral parts on the music stands) to lay out the music far more expansively than one finds in actual eighteenth-century parts: in the drawing, there are between 1 and 3 measures per line (exceptionally 4 in one line that includes a 2 measure rest), and the music paper has only 6 staves, rather than the 10 or 12 that it would almost certainly have had in reality. In several cases, measures are broken across lines. Thus the rendering of this viola part is by no means realistic, and the layout of the other visible parts is similarly unrealistic. But the music is perfectly sensible, if extremely dull. Here is the viola part for the movement in D major. (In the following transcription, notes that are partially obscured or otherwise conjectural are given in blue. The note in red in the first measure is almost certainly intended to be an a, but the artist has rendered it imprecisely, and it reads as b.)

This extract is not terribly helpful in identifying the piece, but it is obviously quite un-Mozartean, and is much more strongly reminiscent, perhaps, of mid-eighteenth-century orchestral music influenced by Vivaldi.

Three measures of the “Flauto primo” part are also visible:

This is the opening of a movement in D Major in ¾ (although the time signature is not given in the drawing), thus evidently not the same movement that appears in the viola part. That this is an opening melody gives us some hope of identifying the piece—but a search for this musical incipit in the online database of the Répertoire International des Sources Musicales fails to turn up a compelling match. (The first six notes of an anonymous minuet in the library of the Accademia dei Rozzi in Siena are identical, but the continuation differs.)  It does not correspond to any known piece by Mozart.

Forty-five measures of the first horn part are legible, although fully 22 of these are rests. A much smaller number of measures are legible in the part for second horn, but enough is visible to make clear that this part fits together with the one for first horn in a musically sensible way. The first horn part is clearly marked “Ton G” (something Rampe fails to mention), and the music is evidently for a movement in the key of G major.

For convenience, I have transcribed both horn parts together on a single staff. The musical notes given in purple in the second horn are relatively clear readings; those in blue are conjectural to some degree.

Thus these four parts belong to three different movements: one in D major, cut time (viola); one in D major, ¾ (flute); and one in G major, common time (the horns). No previous writer on the drawing has pointed out this oddity. None of the three musical extracts matches any known movement by Mozart, and no one would want to accuse Mozart of perpetrating the tedious viola part. But the fact that musicians are reading from parts for three different movements shows that the drawing is not a direct depiction of a real event.

It should be sufficiently clear by this point that the drawing has nothing to do with K. 120 or Ascanio in Alba.

In any case, contrary to Rampe’s count, there is only one flute player depicted in the drawing. There is, to be sure, a man standing just behind and to the left of the seated flutist, but he has no visible instrument, so we cannot simply assume that he is a flutist (as we shall see later on, there are other auditors scattered among the orchestra). A careful accounting would say that the orchestra contains “at least one flute, and possibly two.”

Rampe further states that the style of the harpsichord points to an origin in Salzburg, southern Germany, or Austria. He continues:
Daß es sich tatsächlich um ein Cembalo handelt, geht aus der großen Zargenhöhe und dem auf dem Resonanzboden erkennbaren 4’-Steg hervor. Das Design des Instrumentengestells, dessen Beine in Hufe münden, deutet auf den süddeutsch-österreichischen Raum; mit dieser Gestaltung korrespondiert die Bemalung des Cembalocorpus, die, wie auch das Gestell, verblüffende Ähnlichkeit mit dem in der zweiten Hälfte des 18. Jahrhunderts in Österreich dekorierten Dulcken-Cembalo (1745) des Wiener Hofs zeigt.

That it is actually a harpsichord follows from the great height of the case wall, and the 4-foot bridge that can be seen on the sound board. The design of the instrument stand, whose legs terminate in hooves, points to the South German-Austrian region; corresponding to this configuration is the painting of the harpsichord’s body, which, as with the stand, shows an amazing similarity to a Dulcken harpsichord (1745) belonging to the Viennese court and decorated in Austria in the second half of the 18th century.
A direct comparison of the Dulcken instrument with the harpsichord in the drawing suggests that “amazing similarity” may be going a bit far.

There is indeed some similarity in the painted decoration on the sides of the case (although the style is rather generic, and one wouldn’t want to make too much of this), and the legs are indeed similar—although Rampe fails to point out that the harpsichord in the drawing seems to be missing its right front leg on the keyboard end. The harpsichord in the drawing has a prominent Latin motto painted on the bent side of the case, an extremely unusual placement for a motto, something Rampe also fails to mention, presumably because it would undermine his claim of “photographic exactitude.”

However, as builder and restorer David Sutherland has pointed out to me in a private communication, the harpsichord in the drawing is simply “incoherent”; in other words, it is extremely unlikely that it is an accurate representation of an actual instrument. Sutherland writes:
Two interpretations are possible: (1) either the instrument has very high case walls, with a thickening layer applied all around the lower portion of the case; or (2) the thickened band may be supposed to represent the apron of a table stand.
The first option (with legs attached directly to the case) matches no surviving antique instrument and would have required very robust legs with thick threaded dowels screwed directly into the case to keep the instrument from collapsing (assuming it had all of its legs). The second option (which would imply a single-manual instrument rather than a double, as Rampe and others claim) is somewhat more plausible, but (as Sutherland notes):
The table stand [under this interpretation] resembles none known from surviving instruments—the apron is considerably too wide, and lacks any scalloping or other relief that one would expect in such a design. The visual impression is of an ungainly, muscle-bound mass, quite out of tune with eighteenth-century taste.
But most likely this is simply an inexpertly remembered harpsichord, rather than one drawn from life, by an artist who was far from practiced in depicting keyboard instruments. The missing leg and the incoherence of the harpsichord’s design are not the only problems. The left front leg on the keyboard end is badly misplaced toward the center of the instrument. And the perspective is badly skewed: as one would expect, the front legs are drawn roughly 75% of the length of the back legs (judging by the front leg on the left of the drawing that can be seen full length), given that they are “further” from the viewer. The case itself, on the other hand, is drawn at the same height at the keyboard end of the instrument as at the tail, thus giving the illusion that the case actually increases in height from tail to keyboard as the distance from the viewer increases. 

Rampe makes several other unsubstantiated claims. He states that the style of clothing points to a date between 1770 and 1775. That may well be true, but he cites no evidence or secondary literature. Rampe is also the source of the remarkably resilient (but utterly unsubstantiated) notion that a violinist with glasses reading from the music stand nearest the keyboard end of the harpsichord and to its right, is Leopold Mozart.

Rampe claims that the boy standing to the left of the keyboard is the page turner, even though he could not possibly reach the music from where he is standing. And his stance, dress, and demeanor are remarkably aristocratic for a page turner; indeed, this boy could be seen as a principal focal point of the drawing, much more the focus than the boy behind the harpsichord. (That the standing boy has been ignored or relegated to peripheral status is one of the many oddities in the discourse on this drawing.) In any case, evidence of the use of page turners by continuo harpsichordists in orchestral performance during this era is non-existent, so far as I know.

Lastly, and perhaps most bizarrely, Rampe suggests that the “notably small dimensioned oboes” (auffallend kurz mensurierten Oboen) indicate a high Kammerton, which Rampe claims (citing Bruce Haynes) was common in Venice and Lombardy at the time, and could reach to a’=460 Hz. However, given that the oboes are also incoherently drawn, with the bells literally occupying the same point in space, one would probably not want to put too much faith in their being accurately indicative of anything, least of all a high Kammerton.

I’ve spent so much time disentangling Rampe’s claims because nearly every subsequent writer on this drawing has accepted them uncritically, and indeed, most (up to and including the auction catalog) have recapitulated them, in whole or in part. Wolfgang Brunner uses the drawing as the cover illustration for the booklet (and also the cover) for his recording of early Mozart keyboard concertos (1998; 2006), and he devotes a page to the drawing in his notes.

Brunner acknowledges that the attribution to Zoffany and the identification of Mozart are uncertain, but nevertheless repeats the speculation that the violinist with glasses is Leopold, and also reproduces (without citation) Rampe’s claim that the harpsichord is “exactly like” the Dulcken instrument in Vienna. Alberto Basso, in his entry for Zoffany in the “Dizionario delle persone” in his lavishly produced book I Mozart in Italia (2006, pp. 678–79), simply repeats Rampe’s principal claims. Wolfgang is at the keyboard; Leopold, with glasses, fiddles next to him. Perhaps because Rampe’s claims about K. 120 and Ascanio are incoherent, Basso seems to have mistaken him to say that the visible music is definitely K. 120 (“...mentre la musica che si legge sui fogli disposti sui leggii è quella del Finale (Presto) della Sinfonia in re maggiore KV 120...”). Peter Rummenhöller, in his chapter “Zur Orchestrierung von Mozart Konzerten” in volume 1 of Das Mozart Handbuch (2007) simply recapitulates Rampe, including a relatively long quotation, before continuing on with more than a page of ultimately vacuous speculations based on the assumption that what Rampe claims is true. (Oddly, the reproduction of the drawing appears widely separated from Rummenhöller’s discussion of it on pp. 230–32; the drawing is on p. 176, in the section “Salzburger Klavierkonzerte KV 175–217.”) Florence Gétreau, in her comprehensive survey of Mozart portraits (genuine and otherwise) that contain depictions of keyboards, likewise adopts all of Rampe’s principal claims. She entitles the drawing “Concert avec W. A. Mozart, Milan ou Salzbourg, entre septembre et decembre 1771 (?), c. 1770–1775,” and like Basso, takes Rampe to have meant that the piece being performed is K. 120. She likewise claims that the violinist in glasses may be Leopold. The principal novelty of her contribution is to identify a couple of other instruments that might be seen as similar to the harpsichord in the drawing, in addition to the Dulcken instrument mentioned by Rampe. 

Gétreau also cites the meeting between the Mozarts and Zoffany as evidence for the attribution:
Quant à l’attribution à Johann-Joseph Zoffany, elle est soutenue, en dehors des questions stylistiques, par le fait que ce peintre, originaire de Francfort, séjourna durant sa carrière autant en Angleterre qu’en Italie et qu’il semble avoir réalisé un portrait de Wolfgang enfant. En tout cas dans le journal de Zoffany, à la date du 9 juillet 1765, l’artiste indique: «Visite de Leopold Mozart».

Regarding the attribution to Johann-Joseph Zoffany, it is supported by (apart from stylistic questions) the fact that the painter, born in Frankfurt, spent time during his career both in England and Italy, and that he seems to have created a portrait of the young Mozart. In any case, in Zoffany’s journal, for the date 9 July 1765, the artist indicates “Visit of Leopold Mozart.”
Her phrase “portrait of the young Mozart” likely refers to Boy with a Bird’s Nest, a painting attributed to Zoffany that is certainly not by him (I will return to this painting later on). Thus a spurious work is used in support of a dubious one. The date, 9 July 1765, has its own peculiar history in the literature on Zoffany and the Mozarts. Gétreau’s reference to “Zoffany’s journal” is curious, for there is no mention of this document anywhere in the Zoffany literature. Indeed, as we have seen, most of Zoffany’s papers and drawings that were in the possession of his family after his death are thought to have been burned in 1832. In referring to the date and the journal, Gétreau cites Geneviève Geffray’s commentary in the 1991 Salzburg exhibition catalog to yet another painting attributed to Zoffany (no. 148, p. 152), identified there as Porträt von Mozart (?) im Alter von 9 ½ Jahren (I will also have more to say about this painting later on). But Geffray correctly cites Leopold’s travel notes, explaining that the reference to Zoffany is undated. She mentions the date 9 July 1765 by way of noting that it is given without a source in Mary Webster’s Johan Zoffany, 1733–1810, the catalog of an exhibition held in the National Portrait Gallery in London in the first months of 1977. The date is indeed given there in a “Chronology of the known events of Zoffany’s Life” (p. 17), and Webster repeats the date, again without a source, in her massive 2011 biography of Zoffany (p. 96 and Appendix I, “Chronology,” p. 615). Yet Webster herself writes in the exhibition catalog that “[i]t is not known whether [Zoffany] kept an account or a diary” (p. 19), and she mentions the burning of his papers and drawings in 1832. So it seems that Gétreau has simply misread Geffray’s commentary, and transformed Webster’s phrase “Visited by Mozart” (quoted in English by Geffray) into “Visite de Mozart” attributed to a non-existent journal of Zoffany’s. The source for Webster’s date is unknown, but whatever it is, she does not cite it.

What can we say about the meeting between Zoffany and the Mozarts? Not very much with any assurance. The Mozarts arrived in London on 23 April 1764 and departed on 24 July 1765. The entries for London in Leopold’s travel notes cover six pages in the original manuscript, consisting almost entirely of names. Leopold gives dates only for their arrival, their meetings with King George III and Queen Charlotte, and the period of several weeks when they lodged with a Mr. Randal and his family in Chelsea (6 August until sometime in September 1764). The rest consists merely of a list of names, sometimes with locations or indications of a person’s title or profession, but no dates. It is tempting to think that the list is chronological, but it is not clear that this is so.

Zoffany’s name appears on the second page of the London entries:

Zoffany’s name appears to be bracketed together with the name directly above, with the marginal annotation:  Lincolns In[n] Filds / Fields”. Together, the two bracketed entries read:
Mr. Sipruntini Violoncellist und Frau.
Mr: Zoffani Mahler
These entries appear just before the one referring to the Mozarts stay with the Randals in Chelsea in the late summer of 1764. Geffray, in her discussion of Porträt von Mozart (?) im Alter von 9 ½ Jahren in the 1991 Salzburg exhibition catalog, seems to assume that the meeting with Zoffany took place shortly before or even during the stay in Chelsea, but it is not clear that this is a safe assumption; the chronology of the meetings in Leopold’s notes seems not to have been the subject of systematic investigation.

“Sipruntini” was the cellist Emmanuel Siprutini, a Dutch Jew who lived and performed in London between 1758 and at least 1764 (see Conway, 2012). When Leopold was ill in the late summer of 1764, it was apparently Siprutini who introduced him to a cousin who treated him with the rhubarb powder that (Leopold believed) cured him; and around that same time, Leopold made an effort to persuade the non-believing Siprutini of the superior merits of Christianity. (Both incidents are recorded in Leopold’s letter to Lorenz Hagenauer of 13 September 1764, Mozart. Briefe und Aufzeichungen, no. 92, lines 81–88 and 103–114.)

The reference to Lincoln’s Inn Fields has been a minor mystery. Schurig, in his semi-diplomatic transcription of Leopold’s travel notes, mistranscribes it as “Lincolns fn / Fields.”  The current standard edition of the Mozart family letters, Mozart. Briefe und Aufzeichnungen (no, 99, lines 48–49), transcribes it correctly (although, like Schurig, it silently omits Leopold’s canceled misspelling “Filds”), but ignores it in the commentary to the entry. Mozart’s sister Nannerl (Maria Anna Mozart) also mentions “Lincolsin fielsgarten” (Lincoln’s Inn Fields Garden) in her travel diary as one of the many places she visited in London, but she mentions no dates or people (Mozart. Briefe und Aufzeichnungen, no. 100, line 18). Martin Postle, in Johan Zoffany RA: Society Observed, writes:
During the course of his London sojourn between April 1764 and July 1765 the German-born composer Leopold Mozart noted down the names of individuals he had met, including Zoffany and his neighbour in Lincoln’s Inn, the cellist and composer Emanuel Siprutini. (p. 22)
Postle cites no source for the notion that Siprutini lived in Lincoln’s Inn Fields. 

But we know that Zoffany did. According to Penelope Treadwell (2009, p. 137), Zoffany moved to a three-story house on Portugal Row in Lincoln’s Inn Fields at the beginning of July 1765 (a date consistent, incidentally, with Webster’s date for his meeting with the Mozarts). Treadwell reproduces a plan of the house from 1814 (more than 40 years after Zoffany lived there) showing a large room (18 by 33 feet) that she speculates Zoffany used as his studio. But she also points to another attraction of that location for Zoffany:
Also in Lincoln’s Inn was the Roman Catholic Sardinian Chapel...which had been restored after a fire in 1759 and reopened three years later. Zoffany had been a member of its congregation when he first arrived in London and the story was later put about that having worn out his soles through his assiduous attendance of services—which accounted for his appearance barefoot in the taverns of Soho and Covent Garden—Zoffany moved from the Great Piazza to Lincoln’s Inn to save on shoe leather. Satiric humour aside, the presence of the ‘Romish Chapel’ probably was an important consideration for Zoffany, less for religious than for social reasons: the chapel served as a focal point for meeting Catholic friends and acquaintances, from whom he could be confident of attracting rich and important patrons. (139–40)
One suspects that the Mozarts might have visited the Sardinian Chapel for similar reasons. However, there seems to be no reference to it in the Mozart family letters.

Thus the date of the meeting between Zoffany and the Mozarts remains unresolved. Mary Webster’s unsourced date of 9 July 1765, just a little over two weeks before the Mozarts left London, is consistent both with the the date of Zoffany’s move to Lincoln’s Inn and with the mention of Lincoln’s Inn in connection with Zoffany in Leopold’s travel notes. But that by no means proves that the meeting took place on that date or even in July 1765. The Mozarts might have met Zoffany and Siprutini at Lincoln’s Inn on some other occasion for some other reason. The subject awaits further research.

The description of lot 24 in the Godts auction catalog makes an admirable attempt at due diligence, citing the 1991 Salzburg exhibition catalog, Rampe, and Gétreau, as well as the Mozarteum’s correspondence on the drawing around the time of the exhibition in 1991, and the correspondence of Daniel Heartz and Geneviève Geffray as the former prepared to use the drawing in his book. The auction house can’t be blamed for placing its faith in the reputedly scholarly literature, and thus they repeat its errors, including the notion that the piece being performed is K. 120, and Gétreau’s reference to Zoffany’s phantom journal.
On the same day as the Mozarteum’s much more cautious press release (which firmly rejects the attribution to Zoffany), the online edition of the magazine The Strad published an anonymous news report about the purchase, describing it as an “anonymous picture of a 24-piece Baroque chamber orchestra....[dating] from around 1770....said to include an early depiction of Wolfgang Amadeus Mozart.” Most of the remaining content of the article is based on the Mozarteum’s press release, but the article adds its own twist, stating that the “figure assumed to be Mozart is standing to the left of the harpsichord”—in other words, Mozart is said to be the rather aristocratic-looking “page turner.” So far as I know, this idea is novel to The Strad.

Case Dismissed

There is a fundamental flaw in the notion that Zoffany created this drawing in Milan in 1771.
Zoffany was not in Italy in 1771.
Zoffany did, indeed, move to Florence in the summer of 1772 with a commission from Queen Charlotte to create the painting that became The Tribuna of the Uffizi. However, he was still in London in 1771. Nor is he known ever to have visited Milan (or Salzburg, for that matter), and he is unlikely to have crossed paths with Leopold and Wolfgang when they traveled to Milan again at the end of 1772 for the production of Lucio Silla. The basic facts of Zoffany’s whereabouts are included in Webster’s catalog of the London exhibition in 1977 and in the subsequent literature on the painter. It is remarkable that none of the writers on the drawing so far consulted the Zoffany literature to check this.

To summarize: this drawing has been claimed, on the basis of a guess made in 1914 by an art historian who was not an expert on the artist, to be by Zoffany. But scarcely any Zoffany drawings survive, and none of the handful of surviving ink drawings by Zoffany are even remotely like this one. The draftsmanship in the drawing is flawed and cartoonish: there are multiple mistakes in rendering, proportions, and perspective (I will expand on this point below). Zoffany was a highly trained and meticulous craftsman whose paintings are known for a nearly photographic realism in depictions of human figures and objects (including musical instruments). No known painting or drawing by Zoffany shows any of the basic faults of draftsmanship found in the orchestra drawing. It is true that Zoffany met the Mozarts in London in 1764 or 1765, but that fact, while often adduced as evidence, is actually irrelevant to the attribution of this drawing to the artist. The drawing has been widely claimed to be a depiction of a performance of music related to Ascanio in Alba in Milan in 1771. But Zoffany was not in Milan (or Italy) in 1771, and the music on the stands has nothing to do with Ascanio in Alba or the related piece K. 120 in any case.

Thus this drawing certainly has nothing to do with Zoffany.

But if it has nothing to do with Zoffany, is there any reason still to think that it has something to do with Mozart?

The case (insofar as there is one) for the identification of Mozart in the drawing has been based entirely on the acquaintance between the Mozarts and Zoffany, and the supposed identification of the music in the drawing as having something to do with Ascanio in Alba. The acquaintance is irrelevant and the identification of the music is false. Thus there is no positive case for identifying Mozart in the drawing. All we have left, then, is a drawing of two boys in eighteenth-century garb, one at the harpsichord and one standing next to it looking self-satisfied, surrounded by an orchestra and a handful of onlookers. Clearly it would be foolish to assume that either boy is Mozart simply on the grounds that they are dressed in eighteenth-century garb and depicted in a musical setting. The boy at the harpsichord is so poorly rendered, and mostly hidden behind the instrument, with a generic cartoonish face, that an identification of the boy as Mozart is literally incredible. Mozart in his second decade would have been a star of any musical production with which he was associated, and would never have been depicted so anonymously and ineptly.

The boy at the harpsichord is bizarrely proportioned, the apparent distance between his knees and his head suggesting a freakishly long torso. The self-satisfied boy to the left of the harpsichord has much more the dress and demeanor of an aristocratic patron than of a participant in the music making. So he is also very unlikely to be Mozart. And if he were intended to be Mozart, standing nobly among the orchestra, why are the orchestral movements being performed not by him?

Thus the drawing is certainly not Zoffany, and it is virtually certain that it is not Mozart. What’s left?

The Drawing

The drawing is nevertheless of potential interest as evidence of eighteenth-century performance practices, and most writers on the drawing have focused on that aspect. Depictions of eighteenth-century orchestras that show this level of detail are rare. Depictions of such an orchestra performing a piece of instrumental music are rarer still. Every writer on the drawing has given a count of the number of players, although these counts vary. (Totals in the table below omit the harpsichordist.)

Basso and Gétreau (and Rummenhöller, who is omitted from the table) refer specifically to Rampe and basically repeat his counts, but omit the third oboe and simplify his assertions about the upper strings (Rampe counts 3 or 4 second violins, and posits that an additional second violin and a violist may have been cut off when the drawing was trimmed). Brunner roughly agrees with Rampe’s count, but omits the third cello and the third oboe. Irving agrees with Edge; both count only one flute (because only one flute is actually visible). Rampe does not give grounds for his claim that there are 6 first violins and 3 or 4 seconds, but later writers have taken him at his word.

This variation in counting has arisen because of ambiguities in the drawing.

In the reproduction above, the numbers in light blue are associated with figures who are clearly engaged in music making, yellow numbers with figures whose status is ambiguous, and red lowercase letters with onlookers.

There are 34 distinguishable individuals in the drawing in its current state. Six of these are onlookers or auditors: a boy and 5 adult men (a–f) standing along the back wall. Of the remaining 28 figures depicted in the drawing, 23 (including the harpsichordist) are clearly engaged in music making; the status of 5 others is ambiguous or uncertain. The standing boy (number 10) might conceivably be thought to be a singer, were it not for the fact that the viola part is marked “Sinfonia.” The man (3) standing behind and to the left of the seated flutist is undoubtedly the one that Rampe and his followers have identifed as the second flute, but he has no instrument and no music. The men labeled 18 and 19 are standing among the violins, but have no instruments. A bow that might first appear to belong to the latter actually belongs to number 20; in fact, on close examination, figure 19 can be seen to have his hands folded in front of him. So he is not one of the violinists.

A long straight line just to the right of the head of 18 is much too long and at too steep an angle to be a bow; it is more likely intended to be a walking stick or something of the sort propped or hanging in the back corner of the room. So we have no grounds for claiming that he is a violinist either. Number 22 certainly has no instrument, and seems simply to be meditating on the music from his vantage point among the violins. So 22 is an onlooker, and 18 and 19 probably are as well.

Thus by a conservative and careful count the makeup of the orchestra is:
1 flute, 2 oboes, 2 horns, 2 bassoons, 8 violins (nos. 15–17, 20–21, and 23–25), 2 violas, 3 cellos, and one double bass, with a third oboe (26) among the violins, giving an orchestra of 22 (or 23 counting the harpsichordist).
Rampe and his followers count 9 to 10 violins, presumably because they include the men labeled 18 and 19. But 19 is certainly not playing an instrument, and 18 has neither visible instrument nor bow. Figure 28, seated just to the right of the violist 27, is mostly obscured, but we can see make out a bit of his instrument tucked under his chin, presumably a viola as he seems to be reading from the same music as his neighbor to the left. Thus our conservative (but closely observed) count of upper strings is 10, lower than that given by every previous writer (including me).

Several writers have noted that two cellists, a bassoonist, and the double bassist appear to be reading off the keyboard score, and this is surely of some interest for eighteenth-century performance practice. It is often said that basso lines in keyboard scores at that time were sometimes written with extra large note heads precisely so that these would be visible from a distance by the several players who were reading from that part. The practice of having several bass instruments reading from the keyboard part can be seen in other depictions of eighteenth-century orchestras (albeit rarely with quite so many players reading from a single part).

There would appear to be at least seven music stands in the drawing: four for the upper strings, one for the horns, at least one for the flute and oboe, and one shared by a bassoonist and cellist. There is no clear way to distinguish precisely which violinists are reading from which stands, because of the uncertainty in the direction of their gazes. Violinist 20, for example, seems not to be looking at any music at all, and his neighbor, violinist 21, seems implausibly to be looking at a stand several feet away. At any rate, the distribution of stands and upper strings is consistent with two or three players playing from each part (8 upper strings and 4 stands), although it would appear that at least four violinists are reading from the music on the stand used by the violinist with glasses.

I have already mentioned deficiencies in the rendering of the harpsichord and the oboes, but the instruments in general are rather poorly drawn. The tubing of the first horn is eccentric, to say the least, and the instrument appears to have no bell.

The bow of the double bassist seems to be slicing through the rim of the instrument.

Some of the violins are quite crudely rendered.

And so on.  Compare these with the precise and realistic renderings of instruments in Zoffany’s well-known painting The Sharp Family (1779–81).

Several of the instruments depicted in this painting survive in the Bate Collection at Oxford.  When compared side by side with Zoffany’s painting (as seen to good effect in Webster 2011, p. 391), the nearly photographic “presence” of the instruments in Zoffany’s rendering is astonishing.

What else can be said about the performance depicted in the drawing?  It has been pointed out that the lid has been taken off the harpsichord, but this is not really a surprise. Keyboard instruments in eighteenth-century depictions of performances are often shown with the lid removed. Many of the instrumentalists are standing: just four out of the ten upper string players are seated, as are the cellists (as one would expect) and the harpsichordist. The first horn and first flute are seated, as is, perhaps, one of the bassoonists (no. 7). The rest of the players are standing. They are also very densely crowded together, especially the violinists, so much so that some players seem to be poking their neighbors with their bows. A Baroque cellist of my acquaintance has suggested for precisely this reason that the drawing may be satirical: because the string players simply could not function as they are depicted here.

The imprecision and inaccuracies in the rendering of instruments and the men playing them, the cartoonish aspects of the drawing, the impossible harpsichord with a motto on the bent side of its case, and the fact that three different movements are open on the music stands all suggest very strongly that this drawing was not made from life and that it must be approached with considerable caution as evidence of eighteenth-century orchestral performance practice. Indeed, seen in this light, it appears more of a cartoon or caricature than a realistic drawing. It is true that the artist seems to have been musically literate, given that he (or conceivably she) has accurately depicted written music on four different parts in two different clefs and two different keys, all while “editing” the music to fit his format. The drawing suggests a complement of 4-4-2-3-1 in the strings (although we cannot be entirely sure of the division of the 8 violins). This relatively small group is consistent with evidence from other eighteenth-century sources. The heavily reinforced bass line (three cellos, double bass, and two bassoons) is notable, as is the presence of an extra oboist among the violinists, who seems to be playing with them. The distribution of instruments in a semicircle around the continuo harpsichord is plausible and perhaps indicative of actual practice, as is the presence of auditors behind and among the orchestra, suggesting a more fluid boundary between performer and audience than we are accustomed to today.

What other attributes of the drawing might shed light on its subject, its occasion or date, or the artist’s intent? It is notable that all the figures in the drawing are entirely male, including all of the apparent auditors. One expects that audiences at most orchestral performances in palaces, aristocratic residences, and middle class homes would have included women; the exposed beams and rather spartan decor of the room in the drawing would seem to argue against an aristocratic residence in any case. The violist nearest to the viewer is wearing a sword, not a normal accessory for a professional orchestral musician. This may suggest that the orchestra consists at least partly of amateurs.

Several writers have mentioned the Latin motto on the harpsichord and the oddity of its placement on the bent side of the case. (David Sutherland has pointed out to me that a motto of this sort “is a seventeenth-century decorative element that was invariably placed inside the lid or elsewhere in the interior of the instrument, never on the exterior.”) Only Basso (2006) mentions that this same motto appears on the lid of a surviving double virginal by Ioannes (Jean) Ruckers, now in the Württembergisches Landesmuseum in Stuttgart. (Basso fails to cite his likely source for this information, O’Brien 1990, pp. 247–48. A second virginal by Jean Ruckers with this motto is listed in McGeary 1981, but this instrument apparently does not survive. McGeary also claims that the motto appears on a harpsichord in a painting by the seventeenth-century artist Gonzales Coques, but I can identify no such painting in the recent biography and catalog of that artist by Lisken-Pruss (2013), not even among the dubiously and spuriously attributed paintings.) The motto itself, “AUDI VIDE ET TACE SI VIS VIVERE IN PACE” (“Listen watch and be silent if you want to live in peace”), a rhyming couplet, is probably of medieval origin. Hans Walther (Proverbia Sententiaeque Latinitatis Medii Aevi) records numerous occurrences of this motto in slightly variant forms, including (in addition to the form found in the drawing) “si tu vis vivere pace” and “si vis vivere cum pace”). The prominent and unlikely placement of the motto in the drawing may perhaps shed light its intended meaning, yet another avenue for future research. The motto later became associated with the Masons (see, for example, this site), but we ought not to stampede to the conclusion that this has anything to do with the motto’s appearance in the drawing.

Lastly we may note that the drawing includes a sketchily rendered painting on the back wall.

The painting depicts a seated woman in an outdoor setting, with a book (or perhaps music) open in her lap, with a flutist (apparently a boy) to her right, a seated violinist (apparently also a boy) to her left, a standing figure of indeterminate species holding something over its head, also to her left, and yet another head in the background. A cello (or perhaps a gamba) is in the lower right corner of the painting. It is perhaps a long shot, but if this is a sketch of a real painting rather than an invention of the artist, then it could, if identified, shed light on the location, occasion, and provenance of the drawing.

Attributions to Zoffany

At least three other paintings have been claimed to have resulted from the meeting between Zoffany and the Mozarts: the so-called Boy with a Bird’s Nest (Knabe mit dem Vogelnest), in the possession of the Mozarteum since 1924/25; the so-called Porträt von Mozart (?) im Alter von 9 ½ Jahren, which was included in the 1991 exhibition in Salzburg; and a more recent attribution put forward by Daniel N. Leeson, a pastel painting on paper that he has named A Portrait of a Young Girl at Chocolate, which he claims is a portrait of Nannerl Mozart by Zoffany.

It is difficult to see how the first two of these attributions ever gained any traction. Zoffany painted many children in his career, both in family and individual portraits. Indeed, he can be said to be one of the major painters of children in the eighteenth century. At his best (and his best is very frequent), his renderings of children have an almost uncanny sense of living presence. Comparing one of Zoffany’s paintings (here, for example, Sophia Dumergue, 1780–81; see Webster 2011, p. 402) with the two reputed portraits of Wolfgang show the great discrepancy in technical skill.

The portrait of Sophia Dumergue dates from around fifteen years after Zoffany met the Mozarts, and one could argue that his style and technique might have matured in the interim. However, this detail from Three Sons of John, Earl of Bute, dating from 1763-64, shows that Zoffany’s skill in depicting children was already of a very high order, even before the Mozarts arrived in London.

The Boy with the Bird’s Nest has been discussed and reproduced frequently in the Mozart literature, but it has received little attention in the Zoffany literature. Webster ignores it, and Martin Postle relegates his terse rejection to a footnote, where he writes simply: “It has been stated, incorrectly, that a picture of a young boy holding a bird’s a portrait of Wolfgang Amadeus Mozart by Zoffany” (Johan Zoffany RA: Society Observed, p. 48, note 19). Penelope Treadwell attempts to place the painting in a somewhat better light: “The handling of the Salzburg picture is not Zoffany’s, but it could be a copy of a lost original, or perhaps an unfinished sketch taken away by the Mozarts and completed by another artist back home” (p. 154, accompanied by a good reproduction of the painting). But this seems to be merely wishful thinking. The attributions to Zoffany and Mozart have recently been firmly rejected by the Mozarteum (Mozart-Bilder Bilder Mozarts, catalog 67, p. 118).

The other portrait head is rather more accomplished technically, but has been resoundingly ignored in the Zoffany literature. The case, such as it is, is summarized in Geneviève Geffray’s commentary on the painting in the 1991 Salzburg exhibition catalog (item 148, p. 152), based on a  typescript in the library of the Mozarteum by Gianfranco Cavicchioli, “Storia di un dipinto che ritrae W. A. Mozart a nove anni e mezzo” (1987/88). However, it is unclear whether this typescript offers any concrete evidence for either attribution, to Zoffany or Mozart. Geffray’s summary notes only that:
Eine Analyse der Farbe, der Leinwand und der Malweise ließ den Schluß zu, das Porträt könnte von Zoffany gemalt worden sein.

An analysis of the paint, the canvas, and the painting technique is consistent with the conclusion that the portrait could have been painted by Zoffany.
Geffray also notes the facial resemblance between this portrait and the Boy with the Bird’s Nest. But none of this actually constitutes a case for either attribution. The fact that the Mozarts met Zoffany when Wolfgang was somewhere between the ages of 8 ½ (in mid 1764) to 9 ½ (in July 1765) is not evidence that any anonymous eighteenth-century portrait of a boy around that age is Mozart. Although unstated, the logic here seems to be that if the boy is Mozart, then the painter must be Zoffany—which is, of course, not an argument at all.

Ordinarily, I would pass over in silence Daniel N. Leeson’s self-published book The Mozart Cache (2008), although it perhaps deserves a special prize for the sheer density, industry, and creativity of its pseudo-scholarship. Simon Keefe’s devastating review in Music and Letters (2010) devotes a long paragraph to the deconstruction of Leeson’s claims for A Young Girl at Chocolate, but only scratches the surface of the problems. It would take a book much longer than Leeson’s own to unravel all of his claims and specious arguments, and this would be no more worthwhile than devoting a book to refuting the claims of Flat Earthers or those who maintain that the moon-landing never took place. Leeson has, in any case, shown himself to be stridently impervious to criticism, and has become something akin to the Florence Foster Jenkins of Mozart scholarship.

Leeson devotes six pages to his case for A Young Girl at Chocolate, a work of such mediocre quality that it is incredible that anyone would seriously put forward an attribution to Zoffany. An entire page (23) is given over to a detailed discussion of the date of Zoffany’s meeting with the Mozarts, yet the entire argument crucially depends on an assumption (that Leopold was recording a meeting with Siprutini to arrange the childrens’ first public concert) which may easily be untrue. The fact that the meeting took place is not positive evidence for the attributions in any case. Another page (24) is devoted to a comparison between A Young Girl at Chocolate and Boy with a Bird’s Nest, appealing to the expert testimony of Angus Trimble, Curator of Paintings and Sculpture at the Yale University Center for British Art, whose wry dismissal of A Young Girl seems to have been lost on Leeson, who quotes it as if it were approval. He spends another page (25) concocting an elaborate fictional history for the pastel, including dates and addresses, and even a table of the ages of the Mozart children during the period of their sojourn in England. Another page is spent on yet another appeal to expert testimony, to show that the style of clothing in the portrait is consistent with the date that Leeson has assigned to it. All of this is transparently silly and fails to make any sort of positive case for the attribution to Zoffany or the identification of Nannerl as the subject.

However, Leeson does adduce one bit of evidence that appears nowhere else in the literature on any of these Zoffany attributions: he identifies a monogram on the reverse of the portrait as being “Zoffany’s most frequently used monogram.” In support of this claim, he reproduces an image with three similar monograms, all consisting of the letter “J” intersecting the letter “Z.”  Leeson cites no source for this image, and mentions no paintings by Zoffany on which it appears.

It turns out that Leeson seems simply to have lifted the image without credit from H. H. Caplan. The Classified Directory of Artists Signatures Symbols & Monograms (1976, p. 672), editing it slightly to decrease the amount of whitespace between the monograms. It is the monogram on the left that Leeson claims most resembles the one on the reverse of the portrait.

The superfluous dot above the date is a “tell” that Leeson has simply copied from Caplan.

That Caplan has listed these monograms under Zoffany’s original family name “Zauffely,” with no reference to the name by which the artist is universally known in the specialist literature, suggests a lack of expertise on the artist. Caplan gives no references in the dictionary to the sources for the monograms: the dictionary cites no secondary sources and names no paintings or drawings on which they occur.

In fact, it is one of the peculiarities of such dictionaries of monograms and signatures (there is an entire shelf of them in the Fine Arts Library at the University of Michigan) that they generally fail to cite sources, even though this would seem to be fatal to their utility for any serious investigation of attribution. Paul Pfisterer, Monogrammlexikon 2: Internationales Verzeichnis der Monogramme bildender Künstler des 19. und 20. Jahrhunderts (1995) gives what are apparently the same three monograms as Caplan, without citation and likewise without citing sources (entries ZJ 108, ZJ 109, and ZJ 111, again indexed as Johann Joseph Zauffaly). John Castagno, Artists’ Monograms and Indiscernible Signatures: An International Directory, 1800-1991 (1991), gives a single “JZ” monogram for Zoffany, and does cite sources, one of which is actually useful:  Benezit. Dictionary of Artists. The article on Zoffany in the edition of Benezit that I consulted (2006, vol. 14, pp. 1396–7) includes two monograms which appear to be identical to two of those given in the other dictionaries, but again without direct citations of sources or secondary literature. However, Benezit at least includes an extensive bibliography, which would give a careful researcher a place to start.

It is beyond the scope of the present essay to make an exhaustive search of every detailed published description of Zoffany’s paintings and drawings to track down the sources of these monograms. This is work that Leeson should have done. Suffice it to say that Zoffany did not sign his works often. Webster notes:
Zoffany rarely signed his pictures with his name after he came to England, but sometimes added a signature in the renaissance tradition by including a self-portrait…  (2011, p. 605)
Of the 17 paintings by Zoffany that are described in detail in Oliver Millar, The Later Georgian Pictures in the Collection of Her Majesty the Queen (1969, pp. 148–157), including some of the artist’s most famous works, only three are signed (nos. 1205, 1208, and 1209), none with a monogram. Of the 111 items meticulously described in the catalog of the exhibition Johan Zoffany RA: Society Observed, only 18 are signed or carry any sort of inscription by Zoffany himself.  Seven of these are early works, painted before Zoffany moved to London. After the move, he seems as a rule never to have signed commissioned works; those that are signed are mostly self portraits or miscellaneous works not known to have been done on commission. There is no mention in the catalog of a monogram.

The monogram on the pastel portrait would in any case have been very easy to forge (a possibility that Leeson acknowledges, but fails to pursue). In fact, it is the sort of thing that a forger might have done who had access to dictionaries of monograms, but little knowledge of Zoffany’s actual practice. Leeson provides an image of the monogram on the back of the pastel portrait, but the image is so fuzzy and small (35 x 39 mm, with the monogram occupying a 5 x 5 mm square within that frame) that not much can be said about it—and that is perhaps precisely the point: for providing vague and ambiguous evidence is a classic technique used by those who would rather not have that evidence vetted. Here it is, however, blown up to a size that can be seen (sort of):

Leeson fails to ask one quite obvious question: did Zoffany work in pastels? And the answer to that seems to be: almost never. I have so far found only one reference to a work by Zoffany in this medium, Amphitrite (c. 1760), an elaborate mythological painting in pastel on paper, held by the Museum der Bildenden Künste in Leipzig (Johan Zoffany RA: Society Observed, p. 181, fig. 167), created before the artist moved to London. Stylistically and technically, it bears no resemblance whatsoever to the portrait that Leeson has attributed to Zoffany.

Martin Postle dismisses A Young Girl at Chocolate in a footnote, writing: “[T]he portrait in question is clearly not by Zoffany.” (Johann Zoffany RA: Society Observed, n. 19, p. 48). In the same footnote, Postle mentions yet another Zoffany attribution: “a portrait, said to be Leopold Mozart, was exhibited at Fishel, Adler and Schwarz, New York, 1897, by the art dealer Julius D. Ichenhauser.” An image that may correspond to this painting can be seen online at

Mozart and Musicological Fantasy

To summarize: we have four items (not counting the reputed portrait of Leopold) that have been attributed to Zoffany and been said to depict either Wolfgang or Nannerl Mozart. Three of these attributions are so utterly implausible stylistically and technically as to be essentially impossible. The Cavicchioli portrait seems marginally better in terms of execution, but has no known documentary evidence to support an attribution to Zoffany or Mozart. The arguments for the other three rely on the documented meeting between Zoffany and the Mozarts in 1764 or 1765, and this lonely fact is used to argue circularly for the attribution to both artist and subject. Two of the items (the ink drawing of the orchestra, and the pastel portrait) are executed in media that Zoffany scarcely used. There is essentially no positive case at all to be made for the orchestra drawing, yet it has been the subject of an elaborate musicological fantasy that has propagated itself through the Mozart literature for over 20 years.

How does this happen?  How does a transparently flimsy guess at attribution and subject made by a non-expert 100 years ago become stuck in the scholarly literature like chewing gum in the hair?  The drawing itself is of minor significance and will eventually fade back into the obscurity from whence it came. But how can we account for the failure of a succession of scholars to question the attributions or the extravagant claims that have been made about the drawing?  And what can we learn from that failure?

It is an undeniable but little discussed fact that a very large proportion of the specialist Mozart literature is not subject to anything like independent peer review. The journals, yearbooks, and newsletters devoted to Mozart, the standard reference books and “handbooks,” most of the monographs and collections of papers, and the Neue Mozart-Ausgabe itself have been produced without independent peer review. This is not to say that all of the work in those publications is suspect—indeed, some of it is excellent, and some of the great work on Mozart has appeared in these venues. But they also contain a lot of mediocre and substandard work, and even outright rubbish that might have been improved or forestalled if it had been subject to independent review.

Most specialized research and writing about Mozart takes place in what we might call the “Mozart bubble,” relatively independent of mainstream musicology, and for the most part oblivious to relevant and essential work in related fields—consider, for example, the general failure of anyone writing on the orchestra drawing to open a book on Zoffany. In the world of Mozart scholarship, the composer is the sun around which all else revolves. This obsessive focus can instantly exaggerate the potential importance of even a patently weak attribution like that on the orchestra drawing, bringing vastly more attention to the piece than it would ever have received on its intrinsic merits. Arguments for Mozart attributions almost inevitably become circular, because the focus of the argument is, from the beginning, whether it is Mozart.  A rational approach to a drawing of unknown provenance and subject like this one would be to begin with the acknowledgement that there were hundreds of artists who produced ink drawings in the eighteenth century, and thousands of teen-aged boys who played the harpsichord. Thus we don’t, at the outset, have any idea who created this drawing, or whether it is even sensible to think that the boy in the drawing is intended be an identifiable person. When Mozart’s name becomes attached to something, however tenuously, all hope of rational discourse may be least for a while.

How Rampe’s sloppy and largely fictional account of this drawing could have appeared under the imprint of the publisher of the Neue Mozart-Ausgabe is a question that cannot be answered without knowledge of the inner workings and deliberations of the publisher. But why did subsequent writers simply recapitulate the claims rather than questioning them? One can easily imagine reasons: deadlines, pressure to publish, laziness (it’s time consuming to look all these things up), a reluctance to question what is perceived as authority, a reluctance to criticize at all. And it is virtually certain that none of the writing on the drawing was subjected to independent peer review. The 1991 Salzburg exhibition catalog could perhaps have forestalled some of the subsequent confusion by being a bit more explicit in its doubts, by giving the details on the source of the attribution, and perhaps also by including a question mark or two in the heading. But the organizers of the exhibition are not to blame for the flights of unsupported fancy and the failure to question among later writers, who were perhaps seduced by the promise of enhanced reputation through association with a “new” Mozart attribution.

As for the drawing itself, we are back at square one. We don’t know who drew it, when, or why, or what it is meant to represent.  We cannot even say with confidence that it was drawn by a professional artist. The style is much more plausibly reminiscent of Chodowiecki, as the second expert already suggested in 1914. Daniel Chodowiecki (1726–1801) is known today primarily for his engravings, but many drawings (including drawings in ink) also survive. (Good selections of Chodowiecki’s drawings in a variety of media can be found in Bushart 2002 and Deutsche Zeichnung 1987.) The use of lines and cross-hatching for shading by the artist of the orchestra drawing is reminiscent of the technique used in engraving. The heads in the orchestra drawing—some strongly characterized to the point of caricature, others generic—are reminiscent of Chodowiecki, who similarly had occasional troubles with perspective (particularly with pieces of furniture) and details of anatomy. Chodowiecki occasionally (albeit rarely) included musical instruments in his engravings and other works, and these are mostly not very accurately rendered.

After having examined at a wide range of works by Chodowiecki, I would not want to claim that the orchestra drawing is by him. But a recognition of the resemblance is perhaps a good place to start.


Basso, Alberto. 2006. I Mozart in Italia. Cronistoria dei viaggi, documenti, lettere, dizionario dei luoghi e delle persone. Accademia Nazionale di Santa Cecilia.

Benezit: Dictionary of Artists. 2006. Paris: Gründ.

Boalch, Donald H. 1956. Makers of the Harpsichord and Clavichord, 1440 to 1840.  New York: Macmillan.

[Brunner, Wolfgang.] [1998] 2006. W. A. Mozart. Concerto (“Pasticcio”) in D major for harpsichord and orchestra K40.  Concerto in C major for harpsichord and orchestra K246 (“Lützow Concerto”). Concerto in D major for fortepiano and strings K107/1.  Salzburger Hofmusik.  Wolfgang Brunner.  Profil.  Edition Günter Hänssler. Available online at: Most recently accessed 13 Jan 2021.

Bushart, Bruno. 2003. Die Entdeckung der Wirklichkeit. Deutsche Malerei und Zeichnung, 1765–1815. Published in conjunction with the exhibition of the same name, Museum Georg Schäfer, Schweinfurt, 15 June to 2 November 2003. Exhibition catalog.

Caplan, H. H. 1976. The Classified Directory of Artists Signatures Symbols & Monograms. London: George Prior Publishers.

Castagno, John. 1991. Artists’ Monograms and Indiscernible Signatures: An International Directory, 1800-1991. Metuchen, N. J.

Conway, David. 2012. Jewry in Music: Entry to the Profession from the Enlightenment to Richard Wagner. Cambridge: Cambridge University Press.

Deutsche Zeichnung des 18. Jahrhunderts. Zwischen Tradition und Aufklärung. 1987. Edited by Thomas W. Gaehtgens, Volker Manuth, and Barbara Paul. Published in conjunction with an exhibition from the collections of the Berliner Kupferstichkabinett, 2 July to 11 October 1987. Berlin: Gebr. Mann. Exhibition catalog.

Edge, Dexter. 1996. “Manuscript Parts as Evidence of Orchestral Size in the Eighteenth-Century Viennese Concerto.” In Mozart’s Piano Concertos: Text, Context, Interpretation, ed. Neal Zaslaw. 427–460. Ann Arbor: University of Michigan Press. The drawing is discussed in n. 20, pp. 450–451.

Gétreau, Florence. 2010. “Retour sur les portraits de Mozart au clavier: un état de la question.”  In Cordes et claviers au temps de Mozart.  Actes des Rencontres Internationales harmonique, Lausanne 2006, edited by Thomas Steiner.  Bern: Peter Lang. A color reproduction of the orchestra drawing is included as plate VIII, and the drawing is discussed on pp. 91–93.

Heartz, Daniel. 1995. Haydn, Mozart, and the Viennese School, 1740–1780. New York and London:  W. W. Norton & Company. Heartz gives a black and white reproduction of the orchestra drawing as Figure 8.6 on p. 616, and discusses it on p. 617.

Irving, John. 2003. Mozart’s Piano Concertos. Burlington, VT: Ashgate. Irving discusses the orchestra drawing on pp. 138–40, but gives no reproduction.

Johan Zoffany, 1733–1810. 1976. Edited by Mary Webster. Published in conjunction with the exhibition of the same name, shown at the National Portrait Gallery from 14 January to 27 March 1977. Exhibition catalog.

Johan Zoffany RA: Society Observed. 2011. Edited by Martin Postle. Published in conjunction with the exhibition of the same name, shown at the Yale Center for British Art and the Royal Academy of Arts, London.  Exhibition catalog.

Keefe, Simon P. 2010. Review of  The Mozart Cache: The Discovery and Examination of a Previously Unknown Collection of Mozartiana by Daniel N. Leeson. Music and Letters 91 no. 1: 105–107.

Kunsthistorisches Museum, Wien. Bilddatenbank. [Description and image of a harpsichord by Joannes Daniel Dulcken, 1745]., most recently accessed 20 August 2013.

Leeson, Daniel N. 2008. The Mozart Cache: The Discovery and Examination of a Previously Unknown Collection of Mozartiana. Bloomington, Indiana: AuthorHouse.

Libraire Henri Godts. Catalog of the auction held on Tuesday, 18 June 2013.
catalog pdf:
hammer prices:

The catalog entry for Lot 24 was also published at

All pages most recently accessed 4 Aug 2013.

Lisken-Pruss, Marion. 2013. Gonzales Coques (1614–1684). Der kleine Van Dyck. Pictura Nova XIII. Turnhout, Belgium: Brepols.

Lockwood, Lewis. 1991. Performance and “Authenticity.” Early Music (November 1991): 501–508. The orchestra drawing is reproduced at the head of the article, on p. 501.

McGeary, Thomas. 1981. Harpsichord Mottoes. Journal of the American Musical Instrument Society 7: 5–35, here esp. 20 (motto no. 8). 

Millar, Oliver. The Later Georgian Pictures in the Collection of Her Majesty the Queen. London: Phaidon, 1969.

Mozart-Bilder Bilder Mozart. Ein Porträt zwischen Wunsch und Wirklichkeit. 2013. Edited by Christoph Großpietsch; Internationale Stiftung Mozarteum. Salzburg: Verlag Anton Pustet. Exhibition catalog.

Mozart. Bilder und Klänge.  6. Salzburger Landesausstellung, Schloss Klessheim Salzburg, 23. März bis 3. November 1991. Rudolf Angermüller and Geneviève Geffray; Land Salzburg in Zusammenarbeit mit der Internationalen Stiftung Mozarteum. Exhibition catalog.

Mozart. Briefe und Aufzeichnungen. Gesamtausgabe. 1962. Edited by Wilhelm A. Bauer and Otto Erich Deutsch. 7 vols. Kassel: Bärenreiter.

O’Brien, Grant. 1990. Ruckers: A harpsichord and viriginal building tradition. Cambridge: Cambridge University Press.

Pfisterer, Paul, ed. 1995. Monogrammlexikon 2. Internationales Verzeichnis der Monogramme bildender K[ue]nstler des 19. und 20. Jahrhunderts. Berlin: Walter de Gruyter.

Rampe, Siegbert. 1995. Mozarts Claviermusik. Klangwelt und Aufführungspraxis. Ein Handbuch.  Kassel: Bärenreiter.  Pp. 64–67, with a full-page black and white reproduction of the drawing on p. 66.

Rummenhöller, Peter. 2007. “Zur Orchestrierung von Mozarts Konzerten.” In: Mozarts Orchesterwerke und Konzerte. Das Mozart-Handbuch, vol. 1.  Edited by Joachim Brügge and Claudia Maria Knispel. Laaber:  Laaber, pp. 176 and 230–232.

Schurig, Arthur. 1920. Leopold Mozart Reise-Aufzeichnungen 1763–1771. 27 faksimilierte Blätter. Dresden: Oscar Laube.

Stiftung Mozarteum Salzburg. 2013.  “Latest News,” item for 8 Jul 2013. Most recently accessed on 4 Aug 2013.

The Strad. 2013. Does 18th-century orchestral drawing depict Mozart? 8 July 2013. Most recently accessed on 12 Aug 2013.

Treadwell, Penelope. 2009. Johan Zoffany, Artist and Adventurer. London: Paul Holberton.

Walther, Hans. 1963. Proverbia Sententiaeque Latinitatis Medii Aevi. Lateinische Sprichwörter und Sentenzen des Mittelalters in alphabetischer Anordnung. Carmina Medii Aevi Posterioris Latina, II/1.  Göttingen: Vandenhoeck & Ruprecht. 

Webster, Mary. 2011. Johan Zoffany, 1733–1810. New Haven and London: Yale University Press.  Published for the Paul Mellon Centre for Studies in British Art.

My thanks to Michael Lorenz for offering me the opportunity for this warm-up exercise after an unwanted hiatus from writing on Mozart. I am grateful also to Ethan Allred, David Black, Laura Blanken, David Buch, Dick Mackey, and David and Enid Sutherland for their comments, suggestions, and insights.

© Dr. Dexter Edge 2013. All rights reserved.