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. 43 in E flat major Hob. I/43 “Mercury” (c.1771)

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

No explanation has ever been found for the nickname “Mercury” attached to Haydn’s 43^rd symphony, almost certainly long after his own day. I am not sure exactly what a “Mercury Symphony” ought to sound like – you would be hard put to find in this work a portrait of the “winged messenger”. It is, on the whole, a serene and contented work, with an intimate tone that suggests it might have grown out of a string quartet. Indeed, if the four principal string players were to play their parts as a quartet, no harmonic or melodic interest would be lost[1]. Such a performance would, however, provide a fine lesson in the strokes of colour to be obtained from judicious use of two horns and two oboes. The opening theme continues for some time in a conversational manner, in spite of the three forte chords that begin and mark out its progress, until more brilliant writing breaks in. The development has one of Haydn’s false recapitulations, gulling the listener to suppose he is back on the home stretch already, when in fact Haydn has a good deal more to say. The strings are muted for the Adagio second movement, a songful piece in the homely (but unique in Haydn’s symphonies) key of A flat major. Some hesitant chromatic touches add poignancy, though each part ends in a more flowing section, with added sonority from the oboes and horns – the former are otherwise silent throughout. The minuet is poised between elegance and forthrightness. An unusual feature of the trio is that the second part begins in exactly the same way as the first – you would normally expect it to begin in the key reached at the end of the first part, and then find its way back. The finale begins hesitantly, before breaking into the most “mercurial” music of the symphony. Hesitancy returns continually, though, not least at the end. Most unusually for a finale in binary form, after the two parts have been repeated – if the conductor sees fit to do so – there is a coda in which the music seems almost to be petering out in isolated phrases for the strings, until a brief tutti flurry brings it to an end.

The first recording was issued in 1951 by the Haydn Society (HS 9071, issued in the UK by Nixa HLP 1041). Its Haydnesque credentials were declared with a sleeve note by Robbins Landon, a further note by the editor of the score, Jens peter Larsen, listing the differences from the old Breitkopf edition, and by the employment of Mogens Wöldike and the Danish Radio Chamber Orchestra. Wöldike had already recorded rare Haydn symphonies before the war and later entered the Haydn Pantheon with his fiery readings of the last six symphonies. He well understands that this is not a fiery symphony, capturing its intimate tone to perfection. His tempi for the first two movements may seem a little slow at first, but as the music reaches the forte passages and there are more notes to accommodate, we realize that the tempi were exactly right. His minuet and finale are beautifully poised. I shall be interested to see what emerges later in this survey, since I cannot imagine a better performance, except for the matter of repeats. Wöldike plays the first in the first movement and finale, neither in the Adagio. The interesting thing is that the Robbins Landon edition gives a timing of c.23 minutes for the symphony. Wöldike takes 22:56 – but with all repeats this would have stretched to something like 35, too long for an LP side in those days. So did Robbins Landon actually believe the other repeats should not be taken?

It was fortunate the first recording was so good, since nothing else came along for fifteen years – though I doubt if the Wöldike remained in the catalogue all that time. Two new recordings appeared in 1966.

Leslie Jones and the Little Orchestra of London (Pye GSGC 14046) seem determined that the outer movements shall sound mercurial at all costs. The opening movement goes at a lively one-in-the bar. This does not preclude sensitive shading of the initial theme and the forte passages are certainly refreshingly brilliant. The trouble is, Haydn gets reduced to froth in the process. The Adagio goes at a flowing two-in-a-bar tempo and it is marked 2/4 not 4/8 after all. Here, too, this does not preclude sensitive shading of the chromatic passages. In the flowing passages that end each section, though, Wöldike’s experienced wisdom in taking his tempo, not from the beginning and then sticking to it regardless, but from a passage further in which has a lot more notes, is all too evident. Jones’s oboe soars above the texture attractively, but the tune is in the strings and it cannot be heard. This is partly a matter of balance, but I feel the real trouble is that Jones sees the violin writing here as unimportant decoration rather than the real melodic interest. One might ask, at this point, whether Adagio is strictly an indication of tempo, or whether it indicates a feeling or a mood. Judged by the metronome, this can doubtless be called Adagio, but the mood created is that of an agreeable serenade, Andante or even Allegretto. The minuet, however, is a delight, with a lovely lilt. In the finale, the same problem arises in the first movement. The tempo at the beginning may not seem all that fast, and I cannot say that Jones barges through Haydn’s stoppings and startings heedlessly, but when the running quavers start the music assumes a meaningless brilliance. Again, there is the problem of a tempo that sounds Allegro when you start but sounds like Prestissimo further in. As with Wöldike, first repeats in the outer movements, none in the second.  

The recording by the Hungarian Chamber Orchestra led – not conducted – by their founder Vilmos Tátrai  also appeared in 1966 (Qualiton LPX 1087)[2]. My reactions to this performance add another chapter the reflections I have made from time to time in this series on the difference between real tempi and perceived tempi. My impression as I listened to the first movement was that this was more spacious than Jones, closer to Wöldike. The feeling is definitely three-in-a-bar and the players have all the time in the world to express the cadences of the opening paragraph. When the more brilliant music arrives, there is clarity and a steady swing. It is a little punchier than Wöldike, but in similar mould. Haydn is not reduced to frothiness. And yet it lasts only four seconds longer than Jones! I think the difference comes because Tátrai, as the founder and leader of one of the great European string quartets, is leading the performance at a tempo at which he, as a virtuoso string player, can properly express the music. No doubt the leader of the Little Orchestra of London was an excellent musician, but his task was to fit his playing into the conductor’s beat, with the result that the music charged along breathlessly. This is all the more striking in the finale, where the Hungarians are actually faster than Jones, let alone Wöldike, but the feeling is still that of a brilliant Allegro, not a free-for-all Prestissimo. The Hungarian performance also handles the coda very imaginatively. Tátrai also offers a forthright, rustic minuet, enabling them to make a complete contrast with the trio, which is taken very gently without any change of tempo. But what really sets the seal on the Tátrai performance is the Adagio – a real adagio in spirit as well as actual pace. At the beginning I wondered if the music would bear such a grave approach, but it emerges with greater stature. This symphony is actually a contemporary of the Sturm und Drang symphonies, many of which begin with profound and extensive slow movements. No. 43 is generally described as in light-hearted contrast to these, but in the hands of Tátrai and his players, this movement reveals a kinship with these more “troubled” works. By placing such a slow movement at the heart of their performance, the Hungarians give extra context and point to the “mercurial” outer movements. I began by saying I could not imagine a better performance than Wöldike’s, but I seem to have found one – and better recorded, too.

So that is how things stood when the two complete cycles hit the ground. Märzendorfer takes a fairly broad view of the first movement, closer to Wöldike, though in the forte passages his energized quavers in the bass and rasping oboes provide plenty of forward motion. More than any of the others, Märzendorfer emphasizes the contrasts in this movement. His Adagio is slower even than Tátrai’s. He holds the attention with his obviously deep feeling, but perhaps this is a movement that responds especially well to the intimacy of a conductorless performance. His minuet is trenchant rather than buoyant and, like Tátrai, he maximizes the contrast with the trio, taken very gently. His finale is similar to his first movement. This is definitely an Allegro not a Presto and he brings extra weight to the movement by playing the second repeat – otherwise he, like the others mentioned, observes just the first repeat in the first movement and none in the second. He shapes the coda very well. This fine performance takes an honourable place in the Märzendorfer cycle but Tátrai and the Hungarian Chamber Orchestra – who have not always satisfied me greatly in these articles – seem to have something special here.    

Dorati’s performance, recorded in November 1970, is masterly, though not always in ways that I can fully enjoy. His first movement is the fastest of all. The repeated notes that begin the main theme are rapped out staccato, yet he demonstrates that, with strict conductorial control, fine shading can be attained even at a high speed. When the semiquavers burst in, this is nothing if not mercurial. The question is whether, considering that Haydn did not call it thus, it ought to be. Dorati’s control enables him to be less frothy than Jones, but I could not avoid feeling that I was getting a whistle-stop tour through Haydn’s landscape. Like everyone else, he plays only the first repeat. Maybe the movement would have regained some of its lost gravity by playing the second too. Dorati’s Adagio feels like an Andante to me, yet there is no feeling of haste as with Jones. The playing is full of delicate expression, almost balletic in feel. I could imagine shadowy dancers coming in and out of view in a nocturnal woodland scene. Tátrai and Märzendorfer found more depth here, but this is a fascinating alternative. The first repeat is taken. Dorati’s minuet has an archaic grace. Well as he shades it, I found it difficult to remain interested as it dragged on. Better for formal period-style dancing than for listening, maybe. The finale, with the first repeat only, is splendid. Dorati takes a proper Allegro without any attempt to make it mercurial. He treats Haydn’s little hesitations with some imagination. A mixed bag then. Some may appreciate his mercurial first movement more than I do, but the minuet is surely a trudge. I shall come back to this for the second movement.

Not many live performances have come to light – just two, in fact, and both conducted by Harry Blech. The Cameo Classics/Richard Itter box[3] contains a broadcast studio performance from 27 November 1956 with the London Mozart Players[4]. The opening paragraph of the first movement is phrased with that touch of graciousness that seems to characterize Blech’s work. He shades its phases carefully. When the forte passages enter, though his tempo is not as mercurial as some that have been tried, there is abundant vitality. As with conductors of an older generation, he brings added dynamics to Haydn’s sparse indications, grading the full orchestra passages with crescendos and diminuendos. Purists may object. Others may feel, as presumably Blech did, that any musical player of any generation would do this naturally. The overall impression is of warmth and generosity, except in the matter of repeats – it was naughty of him not to give us even the first. Warmth and generosity likewise characterize the second movement. In the full orchestra passages that conclude each part he obtains a gloriously rich sound from Haydn’s apparently simple orchestration. There is much detailed phrasing in the more intimate sections. I thought of Schumann at times, even momentarily of Elgar just after the double bar as the violas soar into the texture with their harmonic suspensions. Anachronistic, some will say, but I loved every minute – just as well since this is the most expansive account of all. No repeats. The minuet strides along purposefully, contrasted with a trio that mingles Schumannesque wistfulness with Brahmsian warmth. The finale, with the first repeat, opens quixotically while the forte passages have a twinkle in the eye that seems to have escaped other interpreters.

The other Blech performance comes from what, so far as I can discover, was the first of only two concerts he conducted with the Orchestra Alessandro Scarlatti di Napoli della RAI, on 24 February 1959. This orchestra was originally an independent body, formed in 1949 with aims similar to that of the London Mozart Players – the performance of 18^th century repertoire with a scaled down band though without too much deference to musicology. At the end of 1956 it was adopted by RAI, which seemed a good answer to their financial worries at the time, less so when, at the end of 1992, RAI decided to disband it. Given the similarity of aims, it is perhaps surprising that Blech was not invited more often, but he seems to have limited his appearances with orchestras other than his own. One might wonder how Blech’s sometimes empirical conducting technique worked with an orchestra not used to it, but there seems to be no problem here[5]. The remarkable thing, really, is how interchangeable the two performances are, the only differences being the observance of the first repeat in the first movement and a finale that has a little more dash and a little less twinkle in the eye. As things stand, the Naples performance, which I recorded on cassette from RAI’s Filodiffusione channel many years ago, is less well recorded. If the original tapes still exist in RAI and sound well, I think I would marginally prefer the Naples version. 

It may be evident by now that the performance I like most is Blech’s – either of them – but elsewhere in these articles I have described Blech’s readings as a guilty pleasure – maybe to younger ears than mine they are all too warm and generous. Tátrai’s performance is exceptional, with Märzendorfer not far behind, while Wöldike’s serener conception has its attractions.

Christopher Howell © 2026

[1] Though the putative string quartet would need to lower the cellos an octave here and there, where Haydn has them playing above the violas and depends on the double bass to complete the harmony.

[2] Gray’s Classical Discography https://classical-discography.org/ gives a date of 18 December 1969, but the EMG Monthly Letter reviewed the recording in March 1966. However, they list only a mono version, while Gray gives a stereo number, SLPX 11458, so perhaps Tátrai recorded it again in stereo. If this is so, I have probably heard the 1969 recording. 

[3] Details here https://www.musicweb-international.com/classrev/2020/Dec/Haydn-sys-CC9119.htm, though Stephen Greenbank’s review does not specifically discuss this symphony.

[4] I gave a brief account of Harry Blech and the London Mozart Players in my article on Symphony no. 49.

[5] Nor in his performances with them of Symphonies 6-8, which I have discussed in separate articles.