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. 46 in B major Hob. I/46 (1772)

Symphony no. 46 has a number of unusual features, starting from the key. For various reasons – which do not amount to a complete explanation – symphonies in B major are rare, though Haydn’s symphonic precursor Georg Matthias Monn wrote one when Haydn was still a child. The mainly cheerful first movement has a number of quixotic passages in two-part counterpoint and two “surprises”. There is a dramatic outburst in F sharp minor just as the music seems to be heading towards a second subject in the traditionally expected key of F sharp major, while the development, after ten bars, restates the opening theme in the tonic. This implies to the unwary listener that there will be no further development and the recapitulation is already under way – which is far from the case. The second movement is a wistful piece in Siciliana style, with the upper strings muted. The minuet is a sturdy affair but the trio, in the minor key, has drawn various speculations from commentators. Ostensibly it is a broad, hymn-like affair, but the absence of a real tune has caused some to suggest that one of the solo instruments (the leading violin, or the harpsichord, which Haydn himself would have played), should improvise freely against this background. Within my cut-off date, nobody has ventured to do this and the best of the performances below leave me convinced that this trio is complete as it stands. The finale contains more quixotic two-part writing for the violins, but the big surprise is the return of the minuet at the end of the development. As a ploy, this obviously pre-echoes Beethoven 5, but the use is very different, since the return of the finale afterwards is very brief. Haydn marks all this to be played twice, but no conductor I have heard has played this second repeat which, if observed, would risk leaving listeners with a feeling of over-exposure to the minuet.

The first recording of this symphony was set down by Walter Goehr in 1952. Originally a limited Concert Hall issue (CH-414), purportedly played by the “Concert Hall Symphony Orchestra”, it was reissued by the Musical Masterpieces Society (MMS-129), with the orchestra described as the “Netherlands Philharmonic Orchestra”. Walter Goehr (1903-1960) was a German Jew, born in Berlin. Forced to leave Germany in the 1930s, he settled in the UK. He made a number of recordings for HMV, including the première recording of Bizet’s Symphony, but Walter Legge tended to relegate him to accompaniments. He became a house conductor for Concert Hall, covering a vast repertoire from Monteverdi to late romantics. I compared Goehr, Märzendorfer and Dorati in this symphony in an earlier article[1], but will enlarge on my comments here. In the first movement, Goehr places much emphasis on contrast, alternating rough-hewn fortes with elegant piano passages. First repeat only. The second movement is taken quite slowly, emphasizing its wistfulness and perhaps hinting at a religioso feeling at times. I did feel the wind were over prominent, dispelling the gentle atmosphere created by the muted strings. The first repeat is played. Goehr’s minuet has a striding, semi-staccato bass line with the pairs of slurred notes in the violins well separated, making for trenchancy rather than elegance. The Presto part of the finale again concentrates on dramatic contrast, with a longish pause before the return of the minuet, which begins hesitantly, emphasizing the surprise.

This was followed, in 1957, by a recording in which Arthur Winograd conducted the Hamburg Chamber Orchestra (MGM E 3426). Born in New York, Winograd (1920-2010) was originally a cellist and, in 1946, a founding member of the Julliard String Quartet. He branched into conducting and was Music Director of the Hartford Symphony Orchestra from 1965 to 1984. During the 1950s he made a number of recordings for smaller European organizations. His Haydn record has been reissued in an excellent transfer by Forgotten Records (FR 2019). He adopts a more elegantly Mozartian approach to Haydn than Goehr. His first movement is quite swift and nicely turned, though with no particularly memorable touches. First repeat only. In the second movement I found the wind once again unduly loud when they entered, but this disturbs less than with Goehr since his muted strings have less magic. His playing of bar 13 and the corresponding bar in the second part as sudden forte interruptions – there is no such indication in the score – suggests he is not seeking to create an evanescent atmosphere in any case. Again, first repeat only. I found the minuet on the fast side though timings show that Hager, for example, is actually faster while sounding slower with his more elegant phrasing. The main body of the finale has a slightly cautious tempo for a movement marked Presto e scherzando. Nor is the opening particularly piano, as requested, so the strong contrasts found by Goehr are lacking. Here too, first repeat only. A decent job but purchasers in 1957 did better to stay with Goehr.

The 1959 recording by Harry Newstone[2] and the Haydn Orchestra (Oiseau-Lyre OLS 6004) came in excellent stereo and this, too has been issued in a well-sounding transfer by Forgotten Records (FR 674). Thanks to the forward wind placement, we hear that in bar seven  Newstone has a bassoon doubling the bass line, though he subsequently uses it as required rather than all the time. He takes a forthright rather than precipitate interpretation of Haydn’s Vivace marking, allowing everything to breathe naturally. The string playing is well articulated though avoiding spiky staccatos. Only the first repeat is given. Newstone adopts an excellent, gently rocking, tempo for the second movement. When I first heard this, I felt that a more backward placing of the wind against the muted strings might have been more effective, but returning to it after Winograd, Newstone sounded fine. He plays neither repeat. Here and elsewhere, Newstone shows a predilection for rather stately minuets, with more legato playing than with Goehr. The playing is sufficiently alert to justify the tempo. He takes a lively tempo in the Presto e scherzando sections of the finale with nimble playing from the orchestra, as well as real pianos where indicated. First repeat only. I am not sure if Goehr’s more robust approach is entirely superseded, but this was certainly excellent and much better recorded.  

Antonio Janigro’s performance with the Zagreb Radio Symphony Orchestra was set down in August 1963 (Vanguard VSD-2146). His opening phrase may be controversial. The staccato notes in the second bar are played diminuendo rather than hammered out forte. No diminuendo is marked but the answering phrase is piano. So do you go for straight dynamic contrast or link the composer’s fairly sparse markings with crescendos and diminuendos? All but extreme purists allow a degree of conductorial “editing” of this kind, but this seems overdone. I try to imagine the opening of Mozart’s “Jupiter” symphony – which likewise begins with a phrase hammered out in octaves and a soft answering phrase by the strings in harmony – treated in this way and it would surely sound weird. However, this is my only reservation over an otherwise marvellous performance. The first movement has much detailed phrasing to match the opening gambit, but this is combined with a fast tempo and a sense of airborne forward movement. The first repeat is given. Like Newstone, Janigro lets us hear clearly in bar seven that he has a bassoon doubling the bass line but, also like Newstone, he appears to reserve it for selected moments. It adds piquant colour to the bass line in bars 42-46 of the second movement, for example. This movement expresses here an intense inner melancholy – the simple rocking movement of the Siciliana countered by long phrases – and the wind do not break the magic. No repeats. The minuet has separated bass notes and well-separated pairs of slurred notes in the violins as in Goehr, but treated more gently. Janigro finds a poignant melancholy in spite of the major key. The Vivace sections of the finale, taken very fast, have scampering vitality and Janigro seems to invest the minuet with even more melancholy on its return. First repeat only. It could be objected that Janigro adopts a style more Mozartian than Haydnesque. Granted his point of view, it could not be better done.

Two other recordings of this symphony appeared in the 1960s. The Princeton Chamber orchestra under Nicholas Haranyi set their version down on 2-3 May 1966 (American Decca MG 10376). A recording by the Masterplayers under Richard Schumacher was issued by Amedeo (AVRS 6174). I have not been able to hear either of these[3].

Märzendorfer clearly has a small orchestra – you can almost hear the single string players. He adopts a fairly steady pace and makes much of details such as the restless quaver motif in the second violins at bar 46, which then plays a major role towards the end of the development. We are made very aware of exchanges between the instrumental parts. The movement becomes a sort of conversation piece, often diffident but with moments of strong passion. First repeat only. Märzendorfer’s second movement is even further from the Siciliana than Janigro’s, with a very slow, but beautifully poised tempo that he justifies with a hushed inner intensity. The wind parts take on the character of a chorale. No repeats. His minuet avoids separating the pairs of slurred notes too strongly and like Janigro, if in a slightly different way, he finds the poetry behind this apparently simple movement. There is infinite sadness to his trio. His finale scampers in a secretive sort of way, the forte outbursts failing to achieve a real conclusion, reaching a hiatus that can only be resolved by going back to the minuet, which is finally brushed aside in the brief final page. First repeat only.

Dorati’s first movement is neatly phrased at an unhurried tempo and nicely observant of the dynamics. While this cheery account has its simple virtues, almost all the conductors previously heard have, in their various ways, found more in the music. First repeat only. His Poco Adagio is more interesting. It is so much faster than the others as to suggest a ghostly, sinuous waltz. On its own terms, it works. It does not move the listener (or not this one) as Märzendorfer’s, in particular, does, but it has a fascination of its own. Played like this, it could easily be a movement from a ballet. The first repeat is played. The trudging third movement sounds like a minuet for footsore pilgrims. Interesting up to a point, but one wearies of it before the end. For the first time, in this anaemic reading, I began to side with those who suggest that the trio is an accompaniment only and needs somebody to improvise a tune on top of it. The Presto e scherzando part of the finale spins along nicely but, as with the first movement, finds less in the music than some other conductors, and there is the return of the trudging minuet to face. First repeat only.

A recording by the Salzburg Mozarteum Orchestra under Leopold Hager is something of a mystery. It is available on YouTube and various places such as Deezer, but the only information they give is that it from an EP. I can find no reference to it in any of the customary places. Extended Play records were widespread in the 1960s but fell from favour as far as classical music is concerned. Hager, though, was the conductor of this orchestra from 1969 to 1981. I once saw him conduct a performance of Beethoven 9 in Milan that suggested he had a liking for fast tempi, but his tempo for the first movement is more Allegro moderato than Vivace. Nicely phrased, the music can take this serious approach, which is just as well since he takes both repeats. The second movement is given adequate expression at a midway tempo. One point of interest is that Hager gives prominence to the violin figuration towards the end which was almost obliterated by the wind in the other performances. First repeat only. The minuet strides along rather formally at a stately tempo (but faster than Dorati’s). The finale is neatly despatched with just the first repeat. The ultimate impression is of a decent rendering under a conductor of limited vision.

In conclusion, Märzendorfer offers the radical solution but does not entirely eclipse Janigro, while there is still plenty to enjoy in Newstone and, if you are not too fussy about the recording quality, Goehr.

© Christopher Howell 2026

[1] https://www.musicweb-international.com/classrev/2018/Sep/Maerzendorfer_Haydn_II.pdf

[2] See my article on Symphony 34 for a brief profile of this conductor.

[3] They are listed in Gray’s Classical Discography, to which I owe most of the information on dates and record numbers in this article.