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. 15 in D major Hob. I/15 (c.1760)

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

Symphony no. 15 has a number of unusual features. It begins with an Adagio in which the first violins spin an arioso line over a delicate pizzicato accompaniment from the other strings, who play with their bows only for a few bars at the end, and with interjections from the horns. Is it a slow introduction or is this a symphony that begins with a slow movement? A bright Presto follows, adding oboes to the orchestra, in something like sonata form, though returning to the tonic for the beginning of the so-called development, and without any marked repeats. But the beginning was not a slow introduction after all, since the Adagio returns, varied at the end to finish in the tonic. An initial slow movement with a fast middle section, then. The minuet and trio are placed second, one of the few occasions when Haydn did so. There is a baroque feel to the perky dotted rhythms of the minuet. The trio is led by a duet between solo viola and solo cello, with commentary by the full complement of violins – not that the full complement was very large at that stage in the Esterhazy court. Now comes a delicate Andante for strings only, mostly pianissimo though with a few stronger passages. This is in conventional binary form with both parts marked to be repeated. The Presto finale is in the form of a further dance – though far faster than any minuet – with a trio in the minor key. This latter is particularly attractive and has the first violins sketching out an often syncopated melody over running semiquavers from the second violins. As with the first movement, no repeats are marked. This symphony betrays its baroque roots, but is nevertheless full of initiative as well as being highly attractive.

The first recording was part of Max Goberman’s aborted attempt to record a complete cycle with the Vienna State Opera Orchestra (HS 12M/12S, though I have it on a CBS LP, 32 16 0166). Goberman has the oboes and horns well forward, the former with a slightly rasping sound, and has a modestly audible harpsichord continuo. If the idea was that some of the textures are harmonically thin without this latter (I am not entirely convinced of this), perhaps the engineers should have given it greater presence. Goberman takes a spacious but not dragging tempo for the Adagio, invests the Andante with gentle charm, is upfront but not overdriven in the faster movements and gives a bouncy account of the minuet. In short, this symphony entered the catalogue with an exemplary first recording. We are told that Goberman recorded the symphonies with all repeats, but that many were omitted from the LP issues. My LP has only the first repeat of the Andante.

This was followed not long after by Wilfried Boettcher’s recording, issued in 1966, with the Kammerorchester der Wiener Festspiele (Turnabout TV 4092/34092S). Boettcher’s record of Symphonies 6-8 was popular in its day. This one seems less remembered – it is not listed in Gray’s Classical Discography but the 1968 EMG Art of Record Buying included it with one star (out of a possible two). I daresay this lesser diffusion is because the symphonies on this record had no names, rather than a reflection of its quality. I am grateful to a fellow member of a private group who has provided me with a nice transfer.

Like Goberman’s, Boettcher’s record contained Symphonies 15-17 and, also like Goberman, his record gives only the first repeat of the Andante. Unlike Goberman, he has no audible harpsichord – I cannot say I missed it – and, also unlike Goberman, he has a more string-based sound with sweet-sounding oboes. This, if you like, is a more traditional sound, and to my ears, it seems more like an early classical sound whereas Goberman seems more baroque.

Boettcher is considerably faster in the Adagio parts of the first movement, and I think he is right not to try to extract too much gravity from this very charming serenade-like piece. He is marginally slower in the Presto, giving us a little more time to appreciate the view. I found myself noting the various harmonic excursions better – they were rather brushed away by Goberman’s single-minded drive. I also prefer Boettcher in the minuet. The tempo is about the same, but Boettcher is less insistent with the dotted rhythms. Again, at least to my ears, more early classical and less baroque. He also finds an intimate quality in the exchanges between solo viola and cello and the violins in the trio. In the Andante, though, I prefer Goberman, who finds a gentle songfulness where Boettcher, with a faster tempo and a more deliberate manner, sounds heavy-handed. Tempi are about the same in the finale, but I prefer Boettcher’s lighter touch, though I realise that for some, Goberman’s more abrasive sound will be closer to 21^st century ideas about Haydn interpretation.  

There were no further versions of this symphony (or of 16 and 17) until the two complete cycles arrived.

Märzendorfer, in the Adagio, goes even further than Boettcher in treating it as a light serenade. He has the horn interjections played staccato – Haydn did not indicate how he wished to have them played. This is utterly delightful, though I wonder if this is not more Andante, even Allegretto, than Adagio. His Presto is very swift, but effervescent rather than driven in the Goberman manner. For some reason, the first violin quavers in bars 15 and 119 are played pizzicato, which is not marked in the Robbins Landon score (nor in the old Breitkopf edition). As we have seen elsewhere, Märzendorfer uses independently researched Urtexts, so I take it there is some justification for interpreting the source in this way[1]. It does not seem very effective. Märzendorfer sides with Goberman in emphasizing the baroque roots of the minuet, but he is jaunty rather than ceremonious, as though Haydn is cocking a snook at the old style. He takes the trio a tad slower, and here he is closer to Boettcher’s more intimate interpretation. Märzendorfer takes the Andante gently and gracefully and is perhaps more attractive still than Goberman – but he does not play either repeat. His finale goes at about the same tempo as the other two, but the outer sections do not seem very dashing. However, he characterizes the middle section best of all by dint of skilful placing of the syncopated notes. There is no definite winner out of these three performances – and no definite loser either.

So what about Dorati? Dorati uses a string-oriented sonority and any harpsichord used is kept well out of earshot. His first movement is extremely interesting, in that he applies in full measure the 20^th century concept of conductorial control. His Adagio is slow – though not so slow as Goberman’s – and each phrase is exquisitely shaped, with internal crescendos and diminuendos. The horns play with a warm legato and they, too, point their phrases with hairpin dynamics. At the end, when the lower strings take up their bows, Dorati swells the dynamics – the first time but not in the repetition after the Presto – to create a rather ominous effect. The Presto is swift and again makes much of dynamic manipulation. This is almost the sort of performance that Stokowski might have given and, in its old-fashioned way, it is certainly imaginative. But then the minuet! In a slow tempo, every phrase is a simper and a curtsey. Possibly valid if – maybe in a ballet – you can actually see the bewigged dancers going through their stately motions. Agreeable even to dance to, if that sort of deliberately artificial dancing is to your taste. But irritating to listen to. The trio continues in this vein. And so does the Andante. At first, it seems charming, but the curtsey on every first beat is self-defeating. First repeat only, and quite enough. The finale goes nicely enough, but the imaginative approach with which Dorati began the symphony dissipated over the following movements and does not recover here – even the middle section is bland.

So the situation is as before. Any of the former three, despite small reservations, reveals a delightful symphony. Dorati, after the first movement, does not.

Christopher Howell © 2026

[1] The only source is a copyists MS in the Fürnberg-Morzin Collection, see Robbins Landon: Haydn: Chronicle and Works Vol. 1, Thames & Hudson 1980 p. 281.