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. 26 in D minor – “Lamentatione”,  Hob. I/26 (c.1768/69)

A pdf version of this article can be downloaded for offline reading and printing here.

H. C. Robbins Landon has shown, with musical examples, how this symphony incorporates, in its first two movements, the simple chant-like melodies of an old Passion play well known to Haydn’s contemporaries[1]. The earliest source bears the title “Passio et Lamentatio”. This is not Haydn’s own MS, which is lost, but it dates from only a few years later than the composition of the work. The chorale quotations are labelled in the Urtext score and are given by the second violins doubled by the oboes. Since the first violins play other music, often elaborate, the question rises as to whether the chorale melodies should cut through the texture or just “be there”. We will see shortly how different conductors deal with this.

The first movement opens with a powerful syncopated passage for the violins above a treading bass line. After a few soft chordal bars, the syncopated music returns piano. There is no transition, as would be normal in a sonata form movement – the second subject, based on the chorale, simply explodes in the relative major. When the recapitulation comes, the second subject is in the tonic major. This became a common practice for Haydn in his later minor-key symphonies, such as nos. 83, where relatively little of the music is actually in the minor key at all, but was unusual at the time. This must also be one of the first symphonic first movements in which the second half is not marked to be repeated. In the second movement, the chorale melody is combined at first with a flowing countermelody on the first violins, which later branch into elaborate figuration in triplet semiquavers. These in their turn are combined with the chorale melody. Towards the end, the second violins also play triplet semiquavers while the horns join the oboes with the chorale theme. The last movement is a minuet notable for curious pauses and assertive passages in bare octaves. The second part contains a passage in canon, starting with the lower strings. The trio is in the tonic major, its apparent calm punctuated by sharp forte chords. This enigmatic ending led some earlier commentators to hypothesize that a final movement has been lost, but Robbins Landon shows there is no evidence for this. This troubled symphony is counted among the earliest from Haydn’s Sturm und Drang period.

The first recording was a Haydn Society LP with Anton Heiller conducting the Vienna Chamber Orchestra. It was set down in May 1950 (HSLP 1019). Heiller gives a fiercely passionate account of the first movement. I only wondered if the oboes might have been given more prominence – the chorale tune sounds to be played by the second violins only. I definitely would have liked greater prominence to the oboes in the second movement – when we do hear them they sound gravely attractive. But never mind, Heiller finds an almost Bachian spiritual depth here. No repeats. With both repeats, the movement would have lasted close on thirteen minutes, but I could have taken it. Heiller also captures the almost unbearable poignancy of the minuet, with even the major-key trio smiling through tears. Those who got to know the symphony through this recording were left in no doubt that it was a visionary masterpiece.

According to Mike Gray’s Classical Discography, a version was set down by the Saarländisches Kammerorchester for Les Discophiles Français on 19 March 1954[2]. No conductor is named and no disc number is given. It would be strange for this orchestra to record without their founder-conductor Karl Ristenpart, though I would not doubt their capacity to play an early Haydn symphony without a conductor. A member of a discussion group has kindly consulted a biography of Ristenpart[3] and tells me that Symphonies 90 and 91 were recorded on that date, but no recording of no. 26, on this or any other date, is listed. Nor do I find any reference to a recording of Symphony 26 by this orchestra, with or without conductor, in any other source. Probably Gray is mistaken, though there remains the possibility that the symphony was set down at the same sessions as for 90 and 91 but not issued.

This brings us to Max Goberman and the Vienna State Opera Orchestra, recorded some time between 1960 and his untimely death on the last day of 1962. He takes a much broader, forcefully austere,  view of the first movement. This is impressive, though I wonder if Haydn’s unusually detailed indication Allegro assai con spirito does not mean something closer to Heiller’s hell-for-leather treatment. With Goberman, as with Heiller, I would have preferred the oboes to cut through a little more in the chorale theme, though the second violins certainly make it come across. My other reservation is that the syncopated notes in the upper strings are often played fractionally late, with the result that the movement sometimes seems to be in 12/8 rather than 4/4. A pernickety quibble, some may say, but I think that this does blunt the effect to a certain extent, even for the non-score-reading listener. Goberman’s Adagio is slightly faster than Heiller’s, but actually seems slower. This is because Goberman’s well-separated, unphrased bass line increasingly weighs upon the proceedings. In addition to this, the triplet semiquavers are also separated and absolutely even, with the result that they tick over rather mechanically. Lastly, since readers will have got the idea that I did not enjoy this greatly, the chorale theme is given such prominence by the second violins that the counter-melody in the first violins is almost inaudible. All these three features mean that the sublimity revealed by Heiller is kept well at bay and I was thankful that the engineers did not give us the repeats, which Goberman almost certainly played. More successful is Goberman’s minuet, though even here, comparison with Heiller shows a rigidity that makes the trio sound excessively formal.

The Milanese cellist and conductor Antonio Janigro made something of a speciality of Haydn’s Sturm and Drang symphonies. His version of Symphony 26 was set down on 22-24 June 1964 (Vanguard VSD-71126) with the Vienna Symphony Orchestra or rather, from the sound of it, a small section of that band. He takes the first movement fairly swiftly with clear, light textures and the oboe comes through clearly in the chorale passages. This is a strictly classical version, a Haydn who would not have shocked the Esterhazy Court and some may feel this is as it should be. In their different ways, Heiller and Goberman are far more involving. In the second movement, Janigro does not have the oboe cut through – though the horns state the theme proudly at the end of the movement. He nevertheless has a good balance between the chorale theme and the first violins’ counter-melody. His bass line is not gruffly staccato, as is Goberman’s, but his more flowing tempo does not approach Heiller’s sublimity. It is interesting that, while he takes the first repeat, his performance is barely longer than Heiller’s. When the semiquaver triplets arrive, he has the bass notes separated more and the triplets themselves are kept detached. However, he has the players modulate their tone as they do not with Goberman – this is perhaps a case where the presence of a fine string player on the rostrum is a definite plus – and avoids the mechanical effect of this latter. Janigro’s minuet is very nice indeed – his delicate touch allows the poignancy to register, though he avoids being too assertive in the forte passages. The trio is slightly slower and has the sort of lilt that would be difficult to obtain outside Vienna. There is the usual problem after a slower trio – the minuet seemed at first to be hurrying. In the last resort, this is a nice performance of a symphony that requires more than just niceness.

1964 also saw the issue of Leslie Jones’s recording with the Little Orchestra of London (Pye GGC 4014/SGGC 14014). Like Janigro, he keeps the first movement within strictly 18^th century bounds. The opening sounds more bluff than passionate, becoming jaunty when the chorale theme enters. Excellently done if you like it that way. His second movement, taken quite fast, has a gentle, easeful movement. Rather interestingly, when the triplet semiquavers arrive, he has the lower strings slur some of their notes, evidently believing that the principal melodic interest is here rather than in the triplets. There is a pleasingly proto-Schubertian flow, taking us farther than ever from Heiller’s sublimity. The minuet is more perky than poignant, though Jones makes more of the dynamic contrasts than Janigro. Another performance that is rather too nice for the job in hand.

A recording of Symphony 26 by the Bath Festival Orchestra conducted by Yehudi Menuhin was issued in 1967 (ALP/ASD 2297). It was awarded the maximum two stars by EMG so someone, at least, thought it was good. Under the circumstances, and given the eminence of the musician in charge, it seems strange that this LP, also containing Symphonies 44 and 48, has disappeared into a black hole. It never migrated to a cheaper label in the latter days of vinyl, has not been issued on CD and I could not find it on any of the various blogs and YouTube channels that are often helpful with this sort of material.

The version by the Hungarian Chamber Orchestra led – not conducted – by Vilmos Tátrai, was recorded during sessions on 12-13 and 16-20 September 1969. As I have remarked à propos other Haydn recordings by this ensemble, the impression is of an expanded string quartet rather than a scaled-down symphony orchestra. The oboes and horns, in fact, are kept in the background, though the chorale melodies come through clearly as a result of careful balancing between first and second violins. In a way, this is strictly classical Haydn no less than those of Janigro or Jones but, perhaps because the players are on their own, listening and responding to each other like a string quartet, there is nonetheless a sense of passionate engagement. I did not feel I was sold short. Tátrai’s group take the Adagio faster even than Jones but there is no impression of too easy a flow. The bass line is quasi legato. Interestingly, when the semiquaver triplets arrive, Tátrai does not add slurs to the bass line, but adds some to the triplets themselves, which emerge as more gently melodic than usual. If I missed Heiller’s sublimity at first, I found here instead a miraculous serenity which was equally moving. The minuet sets the seal on a wonderful, even great, performance. The slowish tempo is fully alive to the poignancy of the music and the trio has an affecting wistfulness. A dance, maybe, but a last dance like that of Schubert’s Leiermann.

Raymond Leppard set down his version with the English Chamber Orchestra on 31 May 1970 (Philips 6500 084). He can probably  claim credit for the first recording to be issued with all repeats[4]. In his intentions, his first two movements are remarkably similar to Tátrai. However, he joins Goberman in allowing the violins to play their syncopations in the first movement fractionally late, creating the impression that it is in 12/8 not 4/4. It can easily happen, but the reason for holding rehearsals with an orchestra of such expertise that it could sight read any Haydn symphony is to sort things like this out. In the end, this is another likeable rather than involving reading. In the Adagio, Leppard has the lower strings play their quavers detached and very evenly. When the semiquaver triplets arrive, he has some of them slurred in a similar manner to Tátrai, but the overall effect is of placidity rather than Tátrai’s serenity and, with the unvaried bass line, the inclusion of both repeats proved less than welcome after all. In the rest between the two parts of this movement a hitherto inaudible harpsichord is heard, several rooms away by the sound of it, playing a scale. It makes no further contribution to the performance that I could hear and, so faint was it that I played the passage again to be sure it was really something on the record and not some kid practising scales on the other side of the street. Not for the first time I am led to protest, if you think a harpsichord should be there, then let us hear it properly, all through. Leppard’s minuet is very nicely managed, at a tempo that allows its poignancy to come across, though the trio is bland. Perhaps I would have appreciated this performance more had I not just heard Tátrai’s, but even so, I do not think I would prefer it to Janigro or Jones if a strictly classical reading is required.

The stopwatch tells us that Märzendorfer was the first to match Heiller’s rapid tempo for the first movement, but he does so to very different effect. The oboes are well and truly forward here, dominating the texture not only in the chorale quotations but in the opening syncopated passage too. The horns are also well forward, with the result that, in the soft string chords following the first statement of the syncopated theme, the strings sound a mere accompaniment. This must be a miscalculation, since the horns are only doubling the inner harmonies. For the rest, this is fascinating and revelatory. This symphony used to have an alternative name “Christmas”, leading commentators to suppose that a further movement, maybe a Pastorale, has been lost. As stated at the beginning, Robbins Landon has shown this to be mistaken, but was he in time for Märzendorfer to know this? With the chorale sounding on the oboes like pastoral shepherd’s pipes and the violin quavers sounding delicately, we seem to see a Nativity scene through the falling snow. Whatever the rightness of this, it is wonderful. In the Adagio, too, Märzendorfer has the oboes projecting the chorale melody clearly. His bass line is detached, but he has them play so gently as to avoid any sense of heaviness. When the semiquaver triplets arrive, he does not add unwritten slurs, but has them play very tenderly. Here, too, we seem to see falling snowflakes. In this movement, we have had sublimity from Heiller and serenity from Tátrai. From Märzendorfer we have an atmosphere of spellbound contemplation. His minuet sounds very forlorn, as if depicting someone excluded from the scene, allowing just a glimpse of warmth in the trio. I think, on balance, I prefer this even to Tátrai and Heiller – the prominently balanced oboes decide it for me.   

Dorati’s oboes do not dominate as with Märzendorfer, but they are audible. He takes the first movement slower than any and there is an air of world-weariness in the opening bars from which the music never really recovers, despite some fidgeting with the dynamics in the development section. In the second movement, his oboes again do not cut through, but their colour is skilfully mixed with that of the violins. In Dorati’s hands, this movement has the air of a pilgrim’s march, but with a light, proto-Mendelssohnian bass line. When the triplet semiquavers arrive, he adds slurs to some of them. I can only wonder, not for the first time, what was the point of using an Urtext edition and calling in its editor as advisor to the series, if you are going to do your own thing with it. The impression is that Dorati is aiming here at agreeableness rather than any deeply spiritual experience and, if that is enough for you, I suppose you will find him successful, with warm-sounding horns towards the end. His minuet has a gentle elegance that allows its poignancy to come across. In the trio, he makes more than most conductors of the sudden offbeat forte chords, but it seems a matter of obeying orders rather than conveying an expressive reason why they should be there. Not that other conductors have really succeeded in this, preferring to keep them a gentle mezzo forte.

So Dorati leaves the situation unchanged. Märzendorfer is the one for me, but Heiller and Tátrai will also have you marvelling at the sheer range of a composer who could produce a masterpiece so unlike the later works for which he was, for many years, almost exclusively known.

I have no live/broadcast versions to discuss, but the BBC Programme Index, as always, shows some interesting performances if they exist anywhere. Pride of place would go to George Enescu conducting the Boyd Neel Orchestra in 1952, a concert that also included the Missa in Honorem Beatissimae Virginis Mariae, with Jennifer Vyvyan leading the solo quartet – a glittering prize if anyone could find it. Also of interest would be performances by Rae Jenkins (BBC Welsh O, 1964), Graham Treacher (BBC Scottish O, 1965) and Kenneth Montgomery (BBC Scottish SO, 1973). Over in Italy, the RAI archives show that Ettore Gracis conducted a performance with their Rome SO in 1958.

Christopher Howell © 2026

[1] H. C. Robbins Landon, Haydn. Chronicle and Works Vol. II, Thames and Hudson 1978,  p. 291

[2]  https://classical-discography.org/,

[3] Charles W. Scheel/Damien Erhardt: Karl Ristenpart und das Saarländische Kammerorchester (1953-1967). Karl Ristenpart et l’Orchestre de Chambre de la Sarre (1953-1967), SDV 1999.

[4] Goberman almost certainly played them, but his engineers habituallly excised many of his repeats to save space. Moreover, we do not know the exact date of the Märzendorfer recording, which has both repeats.