Pioneering Haydn On Record
by Christopher Howell
An occasional series dedicated to commercial and live recordings of Haydn symphonies up to and including those of the first two complete cycles by Ernst Märzendorfer and Antal Dorati
Symphony no. 44 in E minor “Trauersymphonie” Hob. I/44 (1772)
A pdf version of this article can be downloaded for offline reading and printing here.
Symphony no. 44 has been among the more frequently visited of Haydn’s works from the Sturm und Drang period, perhaps because it satisfies our expectations as to how a Sturm und Drang symphony should be. The outer movements are both uncompromisingly tough, each beginning with a forte phrase in bare octaves and remaining in the minor key through to the end – Haydn’s later minor-key symphonies generally work their way to the major before the first movement is out, as in no. 83. It is true that the “second subject” – inverted commas because it is really a continuation of the first, not an independent theme – goes into the relative major, but at the end of the exposition Haydn ducks back to a B major chord, enabling him to begin the development in the dominant minor. This procedure is mirrored in the finale. The minuet is a severe canon in the minor key, so major-key relief is left for the trio and the Adagio. These are both in the tonic E major, rather than in the relative G major, as might have been expected of the slow movement at least, and offer a gentle balm all the more welcome for the context. Moreover, by placing the minuet second, instead of after the Adagio, Haydn caused the trio to act as a premonition of the longer Elysian calm to come. Alas for his efforts, in the old Breitkopf edition this “anomaly” was corrected and the minuet put into its “proper place” as the third movement. It is alleged that Haydn wished this tenderly serene slow movement, for muted strings with a few contributions from the wind, to be played at his funeral, giving rise to the symphony’s nickname. Another give-away, if you want to know if an Urtext edition is being used, comes at bars 60-61 of this movement, where the Breitkopf had D naturals instead of sharps in the violin’s melody line.
The first recording was an early effort from the Haydn Society, but they had evidently not yet got into the Urtext business, so we have the minuet placed third and the D naturals in the Adagio. This was conducted by Jonathan Sternberg with the Vienna SO (June 1949, HSLP 1003). I have often found Sternberg’s Haydn heavy-handed but here his deliberate, rough-and-ready approach gives a fair idea of the music and he handles the Adagio with some sensitivity. The long gap between the two parts of this movement may be an editing error. I do think he might have driven the finale a bit faster. A few string portamenti will sound out of place to modern ears. C. G. Burke, in his 1952 survey of Haydn recordings, found less to enjoy than I did: “The Symphony is serious and important, the performance limping and haggard, the sound veiled and scattering”[1].
The 1950s were quite productive for this symphony, beginning with Hermann Scherchen and the Vienna State Opera Orchestra, who set down their account in June-July 1952 (Westminster WL5206, WN 18613). The Breitkopf edition, with reversed order for the middle movements and D naturals in the Adagio, is used. It will come as no surprise that the conducting as such is on an altogether higher level – witness the way Scherchen contrasts the different thematic ideas in the opening paragraph without losing continuity. I had the impression that his tempo was slower than Sternberg’s, since he finds so much more space to express the changes of mood. Interestingly, the stopwatch tells me he is marginally faster.
However, conducting genius sometimes comes at a price. One of Scherchen’s specialities, in music of all periods but especially in baroque and classical, was to choose a tempo that is theoretically ridiculously fast or ridiculously slow, then do it with such mastery as to have you wondering, for the duration of the performance, whether it is everybody else who is wrong. But even he sometimes went too far, and he does so in the Adagio of this symphony. It is in 2/4, after all, so two beats in a bar not four. Phrase it as he will, there is simply not enough happening to hold the attention at this slow-motion crawl. The dotted rhythms in bars 9-15, in particular, just sit side by side and seem to go on forever. This, if you like, is “limping and haggard”. The minuet goes rather better, but even here Scherchen’s insistence on a strong accent on every first beat becomes tiresome. The finale returns to the standard of the first movement – rather slow but strongly argued. As a whole, though, the performance has to be dismissed as a curiosity.
Another major conductor entered the fray the following year – Ferenc Fricsay, whose recording with the Berlin Radio SO was set down on 20 June 1953 (DG LPM 18180). This recording corrects some aspects of the previous ones but adds a few oddities of its own. It would be interesting to know what edition was used. The middle movements are in the correct order, but the Adagio still has the D naturals and there is a rewritten violin line in the first movement (at bar 130). Most puzzling of all, in the Adagio a flute, or maybe two, makes an unscripted appearance – there is no flute in this symphony in either the old Breitkopf edition or the Robbins Landon Urtext, which in any case did not appear until 1966. It doubles the violin line at several points, sometimes at pitch, more often an octave above. It was by no means unheard of, in Haydn’s Esterhazy Court, for the oboist to turn flautist in a slow movement, so I would not rule out the possibility that a contemporary copy had this added part, but I find no mention of it anywhere.
That said, the performance has a lot in its favour. The first movement adopts an elegiac tone, rather more warmly romantic than either of its predecessors. The initial attack, rounded in the Furtwängler manner rather than fierce and gutsy, sets the tone and note how Fricsay eases into the melody from bar 4 with almost Brahmsian warmth. The horns, too, are carefully balanced to suggest an air of gentle foreboding. Maybe the time has come to mention that Haydn marked the movement Allegro con brio, implying a bright and breezy tempo. So far, every performance I have heard has got further away from this marking than the last, but I find it difficult to imagine that this movement ever could sound really bright and jolly. In the major-key passages, Fricsay slightly increases the tempo, finding a touch of elation the eluded the others. The minuet is quite swift, with a tenderly expressed trio. I supposed that the penalty to pay for playing the repeats in the outer movements was the forced removal of the second repeat in the minuet, first time round – 1953 technology being unable to accommodate any more music on a single side. However, this repeat is also missing in the live performance discussed below. The Adagio flows serenely and naturally. Unfortunately, the flute is not always perfectly in tune, or completely synchronized with the violins, making it harder to accept this freebooting intruder on its own terms. The finale is a proper Presto, with splendid drive. Despite reservations, there is a lot to enjoy here.
Audite have issued a further Fricsay performance from the same year, this time with the Cologne Radio SO (Audite 95584). The edition is the same, with added flute. Interestingly, as already remarked, the minuet still loses its second repeat first time round, so this was an interpretative decision not a question of restricted side-length. Differences are marginal but go in favour of the Berlin performance where, in the Adagio especially, Fricsay’s own orchestra makes a more beautiful sound. The somewhat raw oboe timbre and less effulgent string tone of the Cologne orchestra might have benefited a certain type of Haydn interpretation, but not this one.
Less than a month after Fricsay’s Berlin recording, Mogens Wöldike set this symphony down with the Danish State Radio Symphony Orchestra (6 July 1953, Decca LXT 2832 in UK, London LL 844 in USA). This seems to be using the old Breitkopf edition, with D naturals in bars 61-62 of the Adagio (and no added flute), but with the inner movements in the right order. Wöldike tended to be a connoisseurs’ conductor rather than one known to the many, but to Haydn enthusiasts he was a hero, above all for his set of the last six symphonies. I have to say this performance does not attain the same level. The opening lacks bite, though Wöldike finds some of his customary energy later on. The minuet is definitely on the slow side, enough to make me wish he had followed Fricsay’s book and dropped a repeat. The Adagio is nicely played at a flowing tempo, but fails to engage the ear as Fricsay’s did. The finale is on the steady side for a Presto though, as in the first movement, Wöldike fires things up later on. On the whole, I say regretfully as an admirer of the conductor, this is improved Sternberg rather than anything very memorable[2]. The Record Guide’s description of it as “carefully, if rather stiffly, played”[3] seems to stand the test of time.
The following year, Paul Sacher recorded the Trauersymphonie with the Vienna SO (6-8 January 1954, Philips A 00212L). I have discussed several Sacher performances in this series of articles and have usually found him competent rather than inspired. He must have had a special feeling for this work. He starts on the right foot, with the middle movements in the right order and D sharps, at last, in bars 61-62 of the Adagio. He is urgent in the opening paragraph of the first movement, adopting a tempo that allows the major-key portions to stride ahead with real brio. The strings play warmly and cleanly in a highly reverberant acoustic. The oboes add a rustic rasp and the horn pops through the texture attractively, though here the one neo is that the player is sometimes very fractionally behind the beat. Still, the drive and vitality of the reading carry the day. The minuet is also urgent, with an optimistic trio. The Adagio flows beautifully and warmly at a real two-in-the bar, which is what Haydn wrote. This is practically double Scherchen’s speed, but there is not a trace of haste and it sounds absolutely right. The finale is a tad slower than Fricsay’s – his pace would maybe not have worked in this acoustic. It nevertheless has splendid drive, concluding in fine style the first fully satisfactory recording of this symphony.
The 1950s ended with a recording by the Netherlands Chamber Orchestra conducted – or more probably led – by Szymon Goldberg (13-15 October 1958, Philips A 00212L). I was well impressed by this combination’s version of Symphony 83 and there is a lot to be said for their 44. The edition seems to be the same as with Wöldike – basically Breitkopf with the middle movements in the proper order and with D naturals in bars 61-62 of the Adagio – but with more bite than Wöldike found on this occasion. The opening statements are more dramatically separated than with Sacher, who prefers continuity. The upbeat to bar 5 is separated from what follows, so Goldberg avoids sinking into the theme in a Brahmsian manner and from then on he proceeds in a taut 18th century style – the major-key passages have real brio. The minuet is slower than Sacher’s, but there have been slower ones still and the doleful, limping tone is attractive. Goldberg emphasizes the dynamic contrasts in the trio, making it sound more restless than usual. The Adagio has a fairly flowing tempo, but here I feel that Sacher’s real two-in-a-bar hits the nail on the head, removing the last residue of stickiness. No reservations about the finale, which is fiercely driven with both repeats. I have to say that, marginally, I still find Sacher the best so far.
The Trauersymphonie continued to attract record companies’ attention during the 1960s. The Hungarian Chamber Orchestra, led – not conducted – by Vilmos Tátrai, set down their version on 11-12 July 1961 (Qualiton LPX 1059). This seems to be another case of Breitkopf – hence D naturals in bars 61-62 of the Adagio – but with the middle movements in the right order. I have commented before on the way the performances by this group often suggest an enlarged string quartet rather than a scaled-down symphony orchestra – to a greater extent than with Goldberg and the Netherlands Chamber Orchestra even though the scale and methods were theoretically the same. After the stern opening there is certainly the impression of an intimate exchange between the players and the piano phrases that follow. When the music heads into the major key, though, the playing becomes finely assertive, with an almost euphoric brio. The Hungarians are especially effective in driving optimistically right up to the moment of recapitulation, when the music crashes back to the minor key, followed moments of uncertainty, until the brioso music returns to conclude the movement with a demonic air, being now in the minor key. The minuet is just sufficiently faster than Goldberg’s to remove any suggestion of a trudge, and the trio is very gently and beautifully handled. The Adagio struck me as finely poised between Sacher’s swifter tempo and Goldberg’s slower one. Another example of the difference between real and perceived tempi, since the stopwatch tells me it is barely any faster than Goldberg’s. However, the sense of intimate communication between the players avoids any sense of heaviness. I have praised Sacher for his flowing tempo, but Tátrai shows that a slower speed, combined with a feeling of inner serenity, can be equally moving. The finale is full of fiery vigour. This seems to me a very worthwhile alternative to the Sacher recording.
A version was recorded on 3-12 July 1962 by something called the Orchestra San Pietro conducted by Renato Ruotolo (US Decca DL 710069). Very probably, this was the hardworking Orchestra Alessandro Scarlatti di Napoli della RAI, which Ruotolo conducted regularly, under one of its several extra-contractual pseudonyms, but I have been unable to track down this disc.
August 1963 saw a recording by Antonio Janigro and the Zagreb Radio Symphony Orchestra (Vanguard VSD-2145), part of their series of Sturm und Drang symphonies. Janigro joins Sacher in using a score that has the D sharps in bars 61-62 of the Adagio as well as the middle movements in the correct order. Either he or the editor made some changes to the phrasing here and there.
I have sometimes found Janigro’s Haydn over-phrased, and the opening statement of the first movement is a case in point. The staccato crotchets at the beginning of bar 2, instead of being punched out forte, are given with the first forte and the second pianissimo. This seems to me to blunt the effect as well as drawing attention to itself unnecessarily, especially since Janigro applies this every time the phrase occurs – and it does so many times. In the major-key sections, his insertion of crescendo-diminuendos in all the semi-quaver phrases seems to me to detract from the rough vigour most other conductors have found here. I also feel that the omission of even the first repeat is rather amiss. There are also some good things here – the oboes and horns are well forward and contribute a rustic colour not always heard elsewhere. I also had the fleeting impression, here and there, that a bassoon was doubling the lower string lines. There should be one in all these early symphonies, but not many conductors trouble to bring one in. If there really is one here, though, why not make it more evident? While on the subject, there should probably be a harpsichord in these early symphonies too, but if any of the recordings I have heard so far has used one, the engineers have kept it well under wraps.
Conversely to his treatment of the first movement, I found Janigro’s minuet rather too smooth and gracious, cushioning the ear from the poignancy of the music. His Adagio is slow by any standards except Scherchen’s. With a fine string player on the rostrum he manages to hold the attention. Unlike any other conductor so far, he separates the quavers in the bass line at the beginning, and also those of the melodic line in bar 20. Playing the opening bars like this gives the impression that this will be a movement underpinned by a steady baroque tread. This perhaps falsifies the character of this beginning, since the music does not continue in this way. The finale is brisk and lively but rather oddly generates an air of good-humoured bonhomie rather than fierce drive, in spite of the minor key. Coming from a musician I very much admire in other repertoire, I found this performance puzzling. Not for the first time, as items from Janigro’s Sturm und Drang series emerge in these articles, I am left wondering whether the decision to concentrate on Sturm and Drang symphonies was his own or that of the recording company. I cannot help feeling that he might have been better suited to some of the more brilliant major-key symphonies.
Leslie Jones and the Little Orchestra of London recorded their version in 1964 (Pye GCG 4006 / GSGC-14006 in UK, Nonesuch H-71032 in USA). Though the Robbins Landon Urtext was published only in 1966, perhaps Jones had access to Robbins Landon’s research since he seems to be using a similar text. Like Janigro, he has a bassoon, maybe two, and it is rather more audible. Indeed, the wind cut through the texture in a way not heard so far in this symphony, even to the extent of giving prominence to lines that are purely harmonic rather than melodic. His first movement is brilliant and energetic. There seems to be a general tendency, as we move from the 50s to the 60s, to take Haydn’s Allegro con brio at face value, and the semiquaver-dominated forte passages are extremely exciting. If this looks ahead to the sort of Haydn younger listeners in the 21st century expect to hear, they may not be so happy with the minuet, which is rather slow. It is nonetheless alert, and Jones sees to it that we really hear the canon. Earlier conductors, or their engineers, often had the lower strings so much lighter than the upper ones that a listener not primed to hear a canon might not even have noticed there was one. Here, Jones makes imaginative use of the bassoon timbre to make the imitation properly evident. The trio, though, falls rather flat at this tempo. Jones is also slower than anyone except Scherchen in the Adagio. Whereas Janigro, almost as slow, got away with it thanks to refined string phrasing, Jones stagnates in places, and the dotted rhythms in bars 9-15 sit inertly side by side as in no version except Scherchen’s. The finale is another matter, full of rude vigour and with both repeats. A mixed success, then.
According to Gray’s Classical Discography[4], the French conductor Henri-Claude Fantapié recorded this symphony on 22-23 March 1966 in the Salle Adyar, Paris (SFP 30002, Baroque 2867). The orchestra was named as the Orchestre de Chambre Henri-Claude Fantapié. A CD transfer of what I presume to be the same performance (Divine Art DDA 21212) gave the same venue but dated the recording 1964 and described the orchestra as Les Solistes de Paris. According to the French Wikipedia, Fantapié (born in 1938), after studying conducting with Bigot and Markevitch and composition with Dutilleux, formed Les Solistes de Paris in 1964 and remained its conductor until 1982. If the date of 1964 is correct, perhaps this was a pick-up band forming the nucleus of the chamber orchestra formed later in the year. We are also told that Fantapié had a close association with Robbins Landon, leading to the rediscovery of forgotten works by Haydn and others. This association no doubt explains why the performance corresponds to Robbins Landon’s 1966 Urtext edition.
Fantapié is swift and energetic in the first movement. The orchestral sound is lithe – the strings do not seem very numerous – and the wind are well forward. These latter, the horns in particular, retain the characteristic French vibrato that we relish in old recordings of Debussy and Ravel. Whether it is so welcome here is another matter. In one or two places, moreover, the players do not pass cleanly from one note to another. The omission of the first repeat is a pity. The minuet has a sad lilt, with the canon clearly audible. The trio is not so characterful. The Adagio is rather slow and about halfway into the first part has settled into a marginally slower tempo still. I found this rather too reverential. The finale is spiritedly done.
I see that my MWI colleague Robert Hugill[5] felt that “in many places the string playing is just too untidy … and in the Adagio problems of tuning render this account completely out of court”. I find this puzzling, since string articulation seems to me as precise as in any other version heard so far. There is a mistake from one of the oboes at bar 38 of the Adagio which should have been remade and, to be carpingly critical, one or two sour spots of sour intonation from the oboes in this movement, but otherwise I had no complaints. I do, however, continue to prefer Sacher and Tátrai.
The Trauersymphonie was recorded by Yehudi Menuhin and the Bath Festival Orchestra in 1967 (ALP/ASD 2297), an LP that enjoyed a maximum two-star rating from the EMG Art of Record Buying but has never been transferred to CD and which I found impossible to trace. Perhaps more understandable is the disappearance of the version by Alexander Brott and the McGill Chamber Orchestra (8 March 1969, Decca SDD2160), a conductor and orchestra well regarded in their native Canada but little known elsewhere.
Carlo Zecchi set down his version with the Slovak Philharmonic Orchestra on 22-23 November 1969 (Supraphon S.110 0963). Zecchi (1903-1984) entered the world of conducting almost by stealth. He trained as a pianist with, among others, Busoni and Schnabel. A handful of 78 sides testify to an artist of exceptional tonal beauty and range, coupled with a dazzling technique[6]. An accident in 1938 curtailed his career but won him handsome compensation from his insurance company. A resumption of his solo career would have meant losing this latter source of income so he performed thereafter only as accompanist, particularly with the cellist Enrico Mainardi. He evidently reckoned, and correctly, that the insurance people had insufficient musical knowledge to realize that the piano parts of the Brahms cello sonatas are just as taxing as his solo music, so the accident need not have stopped his solo career. At the same time, he studied conducting with Hans Münch and Antonio Guarnieri and embarked on a considerable career as a conductor. His records are few and tend to remain out of the catalogue, but are all the more sought after and revered by the cognoscenti for that. They could be supplemented by a trawl through Italian and other radio archives. He was also a revered teacher.
In later years, at least, Zecchi’s performances were characterized by slow tempi, and that is certainly the case here. His outer movements are big and bold, but he has a Klemperer-like ability to create an ongoing, powerful rhythmic impulse so that, after a while, you do not notice the slowness. The repeated notes in the lower strings from bar 5 of the first movement, for example, have an urgent, quasi-Beethovenian pulsation. His minuet does not seem all that slow because he gives a feeling of three-in-bar, so there is a sense of an underlying, repressed energy. He has a bassoon or two on the bass line, balanced well forward and adding droll colour to the canon. His Adagio is slower than anyone except Scherchen, but with refined phrasing to carry the ear forward, I found this sublime rather than stagnant. In the dotted-rhythm passage, for example, the pairs of notes sing as a longer phrase, rather than sitting separately alongside one anotheras they do with Jones and Fantapié, even though Zecchi is slower than these. He is apparently using the Breitkopf score, with D naturals in bars 61-62 of the Adagio, but with the middle movements in the right order.
Such a performance obviously divided opinion. The anonymous critic in the EMG Monthly Letter, while unhappy with the sound quality and noting that the orchestra was not in the top rank, felt that “No one who truly loves Haydn should be without this disc … the slow movements are all done with such extraordinary depth of feeling and purity of style, with such courageously slow tempi, beautifully sustained, as to render this recording unique in our experience of Haydn playing”[7]. Edward Greenfield, in Gramophone[8], was less enthusiastic, noting “little attempt to relate this middle-period Haydn to the eighteenth-century background. But for performances which use a fairly large and certainly opulent-sounding body of strings … these are polished and agreeable performances”. He felt that the Adagio was too slow – “the chains of notes in dotted rhythm almost lose their continuity”. Oddly enough, he went on “unhesitatingly” to prefer Jones, whose performance to my ears lacks continuity at that point even if it is not quite so slow. Jones, for Greenfield, was “not always quite so polished in the playing but much more in style with more imaginative phrasing”. My own impression is that this is big, bold Haydn well worth having as an alternative to more chamber-style performances, without necessarily displacing them[9].
John Pritchard set this symphony down on 10-11 February 1973 (CFP 40021). This would put it just beyond my cut-off date, but I shall stretch a point since I very much enjoyed Pritchard’s earlier recording of no. 80. This, like the Menuhin, has fallen into a black hole, having never made it to CD, but a kind correspondent has sent me an excellent transfer. Pritchard appears to be using the Robbins Landon Urtext and gives both repeats in the outer movements, but not those in the Adagio. He adds some dynamics of his own, particularly in the finale.
The first movement is swift, with clear articulation of the semiquavers and excellent overall drive. About the minuet, I am not so sure. Readers will have got the idea that I prefer minuets not to go too slowly, but even the fastest of those heard so far – Sacher’s – nevertheless sounds like a swift minuet. Pritchard’s sounds like a steady-paced jig, well sprung if you like it that way. His Adagio is also among the faster ones, though less so than Sacher’s, and I found this flowing performance very beautiful. The finale sprints along nicely, with the insertion of several unwritten diminuendos and sudden pianos to make it refined rather than galumphing. If you want a big-band performance without Zecchi’s slow tempi, this may be for you, though I would in that case stay with Sacher. Admirable, even excellent, as Pritchard is, I was left with a sense of incomplete satisfaction, and was bound to wonder why. Perhaps I found the answer when I reflected that, at much the same time as this recording, and with the same orchestra, Boult set down the Brandenburgs and a pair of Mozart symphonies. Here, too, we have an unashamedly full symphony orchestra, yet I do not get the impression that this is blunting my enjoyment. The point is that Boult, even in his last years when his fires were burning lower, was ready to challenge the orchestra, to fire it up. Pritchard, in this case, has the orchestra play, very nicely, within its comfort zone. I say “in this case” because I have not investigated Pritchard’s recorded legacy in any depth, and I must say his Rawsthorne disc for Lyrita was pretty vital.
Noting the repeats in the outer movements, Edward Greenfield, in a generally laudatory review[10], raised an interesting question. “There is also the question of a proper balance of length between movements. After the double length of the first movement, Pritchard opts for a very fast tempo in the canon minuet, fast and light and pointed … He also opts for a fastish tempo in the Adagio, cool and detached but beautifully phrased, and observes the repeat of neither half. With both halves repeated in the finale, it means that two rather long movements frame two rather short ones.” Listeners of Greenfield’s generation were used to hearing classical symphonies with all repeats in the minuet and trio[11], none in slow movements and, where both halves are marked to be repeated in the outer movements, the first only, if that. Add to this the tendency of earlier conductors to take minuets slowly, this meant that the minuet and trio was as long as the other movements, or occasionally longer. If all repeats are observed, and the minuet not taken too pompously, the minuet and trio become a relatively brief interlude. So the answer to one part of Greenfield’s observation is that Haydn, assuming he played all his repeats himself, must have known that the minuets and trio were shorter movements and have intended it that way. The answer to the other part of the remark might be countered with another question: why did Pritchard not play the repeats in the Adagio? In earlier times we could blame omitted repeats on 78 and LP side-lengths, but by my calculations the entire symphony with all repeats, at Pritchard’s tempi, would have lasted a little over 28 minutes, which was perfectly possible in 1973. Several other conductors in this survey would have needed to wait for CD technology if they wished to give all repeats. In any case, if Haydn took it into the reckoning that the minuet and trio was a brief interlude, he would also have expected the slow movements to come out as the longest ones. His apparently arbitrary habit of indicating, or not indicating, repeats in his later works might be worth examining in the light of the relative lengths between movements that result. This is a study more easily undertaken using relatively recent recordings, many of which play all repeats as a matter of course.
Märzendorfer uses the Urtext edition and adds very little to the dynamics Haydn himself wrote, showing that they are completely sufficient. His first movement has a tempo that allows him to swing naturally from tragic minor to bright major, leaving a sense that Haydn is poised between the two. His minuet has a steady, ominous tread, with a trio that does not entirely dispel the unease. In the Adagio, which he takes fairly slowly, the quavers in the lower strings are separated, but the effect is not of a baroque bass line but of a steady, patient pulsation that looks ahead to Beethoven or Schubert. The finale is not taken at breakneck speed but, with both repeats, it unfolds with a rugged vigour that allows certain imitative details between the strings to register. If I have written relatively little about this performance, it is because I find it difficult to fault, unless you want all repeats, which you will not find until later in recorded history. And yes, it does give the idea, missing in Pritchard, that the conductor is pushing his little band beyond its comfort zone.
Dorati’s first movement is remarkably similar to Märzendorfer’s – he, too, finds a good mean course that accommodates both the tragic and the brighter elements. If this is Dorati’s Haydn at its best, in the next two movements he is at his incomparable worst. The minuet is not only slow, it is so smooth as to suggest a pastorale for a particularly gloomy Christmas. The trio is unable to lighten the mood. The Adagio has its first repeat, but the price to pay is that Dorati seems to want to demonstrate that, if you play it insanely fast, it will still be shorter than most performances that have no repeat at all. To be fair, it is done with a tripping delicacy and the passage with the dotted rhythms certainly does not fall apart at this tempo – it sounds like an elegant butterfly waltz. The passage in semiquaver triplets from bar 27 sounds busy as in no other performance. It is extraordinary that Greenfield, in the review quoted above, should have described Pritchard in this movement as “delightfully detached, classically cool [thus far the comment is plausible] against the comparatively expressive styles of Dorati and Jones”. The Jones, considerably slower, is a reasonable comparison but as for Dorati’s dismal travesty, I can only suppose that Greenfield, writing against a deadline, saw the timing and supposed that, because it was longer, it was slower and “comparatively expressive”. Dorati’s finale starts slowly and deliberately. Then, after a patch of rocky ensemble, suggesting that the orchestra were rebelling against the tempo, it settles into a fairly vital account, though without Märzendorfer’s steady growth, and without the second repeat.
Märzendorfer’s recording, then, seems to me comfortably the best up to this time, though I will still go back to Sacher and Tatrai, while keeping Zecchi for especially serious moods.
I have just one live recording to discuss, by Vittorio Gui, from a concertwith the Orchestra Alessandro Scarlatti di Napoli della RAI on 26 January 1968. Interestingly, this took place just three weeks after Barbirolli conducted them in Symphony no. 83, and I have to say that the players are a little less attentive towards the elderly Italian maestro, then in his 83rd year. There are some minor ensemble slips here and there, though there is life in the playing, especially in the last movement. Gui seems to be using the same edition as Fricsay, or something similar, since there is a flute added to the Adagio as well as the D naturals in that movement customary in older editions.
Gui conducts the opening Allegro con brio with rugged vigour. Like Märzendorfer, he finds a route through this movement that gives due weight to its dramatic elements while allowing proper brio in the major-key sections. His minuet is rather slow, but I find it one of the most convincing of the slower traversals, since Gui gives it a robust lilt that prevents it from sounding merely doleful. His trio is nicely shaped and he conducts the Adagio with a very natural flow. His finale is strong and urgent, even if not especially fast, and it is here that Gui most successfully sheds his years to rekindle the sort of fire he regularly brought to his work a decade earlier. I am glad to have heard this and it suggests that a Gui recording in the 1950s would have competed worthily with others of the day.
This symphony appeared only occasionally in RAI programmes – I note it was done by Erminia Romano in Turin in 1963 and by Rafael Frühbeck de Burgos in Milan in 1967. Not that it fared much better with the BBC within my cut-off date – it became quite popular from the later 1970s. It was performed by Rae Jenkins and the BBC Welsh Orchestra in 1965 as part of the BBC cycle of Haydn symphonies, divided between their regional orchestras with a range of conductors. It was given twice by Gary Bertini and the BBC Northern SO (1969 and 1971), by Emanuel Hurwitz, leading a section of the BBC Scottish SO (1970) and by the Swiss conductor Erich Schmid with the BBC SO in 1972.
| I | II | III | IV | |
| Sternberg | 05:19 neither repeat | 05:18 (placed after Adagio) | 06:09 neither repeat | 02:56 neither repeat |
| Scherchen | 05:00 neither repeat | 05:37 (placed after Adagio) | 09:04 neither repeat | 04:20 first repeat only |
| Fricsay Berlin | 07:26 first repeat only | 04:01 second repeat of minuet omitted first time | 05:36 neither repeat | 03:36 first repeat only |
| Fricsay Cologne | 07:16 first repeat only | 04:04 second repeat of minuet omitted first time | 05:10 neither repeat | 03:25 first repeat only |
| Wöldike | 05:12 neither repeat | 06:03 | 05:43 neither repeat | 02:52 neither repeat |
| Sacher | 06:49 first repeat only | 05:01 | 04:38 neither repeat | 03:41 first repeat only |
| Goldberg | 06:40 first repeat only | 05:28 | 05:31 neither repeat | 04:45 both repeats played |
| Tátrai | 06:40 first repeat only | 05:08 | 05:27 neither repeat | 03:38 first repeat only |
| Janigro | 04:49 neither repeat | 05:12 | 05:58 neither repeat | 03:41 first repeat only |
| Jones L | 06:11 first repeat only | 05:47 | 06:28 neither repeat | 05:05 both repeats played |
| Fantapié | 04:42 neither repeat | 05:42 | 06:06 neither repeat | 03:42 first repeat only |
| Zecchi | 05:17 neither repeat | 05:46 | 07:41 neither repeat | 03:59 first repeat only |
| Pritchard | 08:40 both repeats played | 04:06 | 05:12 neither repeat | 04:59 both repeats played |
| Märzendorfer | 06:42 first repeat only | 05:43 | 05:29 neither repeat | 05:15 both repeats played |
| Dorati | 06:47 first repeat only | 06:18 | 06:15 first repeat only | 04:00 first repeat only |
| Gui | 07:02 first repeat only | 06:04 | 04:50 neither repeat | 03:43 first repeat only |
Christopher Howell © 2026
[1] High Fidelity, November-December 1952, p. 72.
[2] My MWI colleague reacted more positively: http://www.musicweb-international.com/classrev/2012/Jan12/haydn_woldike_DACOCD703.htm
[3] Edward Sackville-West, Desmond Shawe-Taylor, with Andrew Porter and William Mann: The Record Guide, enlarged and completely revised, Collins 1955, p. 351.
[4] https://classical-discography.org/
[5] http://www.musicweb-international.com/classrev/2009/apr09/Haydn_Stabat_dda21212.htm
[6] See my review of a Cetra compilation of his piano recordings: http://www.musicweb-international.com/classrev/2004/July04/zecchi.htm
[7] July 1972, p.2
[8] May 1972, p.1888
[9] This recording was coupled with Symphony 49. When I wrote my article on that symphony, a reissue date misled me to believing it fell outside the period covered. At some stage I will revise these articles since new material and information continually comes to light.
[10] Gramophone, February 1974, p.1558.
[11] Though not in the Da capo minuet, which some of today’s Historically Informed performers think necessary.














