Music that grapples with demons and is never wholly at ease, even when wings bless the darkness with glimpses of light – when you’re Robert Schumann, angels are terrifying too.

The symphonic models are clear and we hear the ghostly spectre of Haydn, Mozart, Beethoven, Brahms and Mendelssohn – but no one has told Schumann’s material that it needs to conform, and the music cannot help but spill over any frame its composer attempts to place around it. The first movement of his Symphony No 1 reaches an apparent cathartic end-point as a solo flute line marked dolce reconciles grinding harmonic and structural inner tensions. Time to stop this madness. But then brass, percussion and trilling woodwinds unleash a stampeding burlesque march. Baleful chromatic inclines smudge the harmony, like Offenbach or Sousa turned on their dark side, and such instability derives from the restlessness of Schumann’s mind, you think, rather than being an overtly conceptual compositional strategy.

The free jazz of the Second Symphony’s sostenuto assai prologue, C major credentials asserted by having the strings play anything but, as the brass sustain pure C major triads; in the Third Symphony, that extra movement that sneaks in before the finale, a cobwebby and gothic reimagining of the grounding contrapuntal principles of Renaissance music and Bach; and the audacious cyclic structure of the Fourth Symphony, each movement played attacca and dovetailing into the next. This music of demons and angels grapples also with angles – to take on structure, awkward punctuation, Schumann pushing form, his personal mission being to remould the symphony. And when the realisation dawns that Schumann composed the first version of what would become his Fourth Symphony in the same year as his First Symphony, eyes blink in astonishment. The natural order of things would be to presume that Schumann’s streamlined Fourth Symphony is a perfect distillation of the first three symphonies – but the pathway through Schumann’s symphonic journey is filled with unexpected and improbable twists and turns.

Deciding to record a cycle of the Schumann symphonies begs the question: what exactly should be recorded? And complementary but divergent ideas about the Schumann symphonies have been paraded as rarely before, with four major conductors during the past 18 months releasing four major cycles on disc. Sir Simon Rattle, with the Berlin Philharmonic, gives us four symphonies with the early 1841 version of the Fourth, while Yannick Nézet-Séguin (and the Chamber Orchestra of Europe) and Robin Ticciati (with the Scottish Chamber Orchestra) opt for Schumann’s 1851 revised version. But Heinz Holliger and the WDR Symphony Orchestra of Cologne – like Sir John Eliot Gardiner and his Orchestre Révolutionnaire et Romantique, whose trailblazing 1997 Schumann cycle was given the boxed-up DG reissue treatment last year – perceive Schumann’s symphonic evolution in seven stages. When Holliger completes his cycle during the next year and a half, Schumann’s early Symphony in G minor (theZwickau Symphony) will take its place alongside the first three canonic symphonies, both versions of the Fourth, and the often-overlooked mini-me symphony Overture, Scherzo and Finale – Schumann’s compositional twists of fate put into historical context by a composer/conductor/oboist who has been obsessed with the composer’s enigma for more than 40 years.

I made Rattle’s cycle my Critics’ Choice album of the year in the December 2014 issue; I’ve also elevated Nézet-Séguin’s to an equivalent position in the past. Ticciati’s set has given me much pleasure too, and even more to think about. Nézet-Séguin and Ticciati deploy chamber-orchestra string sections, with Ticciati most explicitly evoking period-instrument practice. Rattle’s Berlin set carries its weightier orchestral ballast very elegantly, and his set became my portal back to Schumann after a longer period than I care to admit when my listening had been dominated by Beethoven, Brahms and Bruckner. Rattle’s opulent, rapacious approach – the climax of the First Symphony’s opening movement and the Trio in the Second Symphony’s Scherzo seemingly moving faster than time itself, while the Berlin strings float the slow movement towards heaven – represented the warmest welcome back possible to Schumann’s dream-built chimeric fantasy world, the heartfelt directness of his melodic fancy played out over structural chess moves. Ticciati’s sometimes manically driven, flintier orchestral sound can be unexpectedly austere; Nézet-Séguin’s cycle is the most unashamedly Romantic of the four, fevered-brow gesturing, rubato with attitude.

Despite their differences, though, the sets are unified by one underlying common denominator – none of them could have been recorded 30, or even 20, years ago. Over the phone from his home in Zurich, Heinz Holliger suppresses a laugh when I ask: why now? Why, suddenly, have maestros gone all Schumann crazy? ‘Well, I started conducting him 30 years ago, when too many conductors had problems with Schumann,’ he reflects. ‘He was never a problem for pianists or composers – Debussy and Berg held him in great esteem – but conductors realised that you cannot try to sight-read Schumann; if you do, the music is completely grey.’ And even 50 shades of grey would not be enough to express Schumann’s multiverse of colour? ‘He does not write out everything; he doesn’t tell you which voice is the principal and which accompanies; nor whether one instrument should have a diminuendo while the others crescendo. To make a Schumann symphony sound light and transparent, as he intended, takes a lot of rehearsal. Each player needs to know whether they’re playing part of only the harmony, or whether they are involved in the counterpoint. Schumann was a great writer of words too, and you need to understand how close the phrasing is to speech. But many conductors are not so interested in this background; they just play what they read.’

Holliger reminds me that Schumann never heard more than 12 first violins during his whole life and, in his view, the period-instrument movement has had a very positive effect on how conductors perceive appropriate orchestral weighting and internal balance. And when I talk to Sir Simon Rattle a few weeks earlier, he makes a characteristically smart analogy: ‘We think of Beethoven and Brahms as being the grizzled old lions of Austro-German symphonic tradition,’ he tells me, ‘but Schumann’s symphonies move like a panther. Beethoven plunges his feet forcefully through the ground; but Schumann’s feet sprint and never fully touch the floor.’

Rattle can’t quite explain why Schumann is suddenly so de rigueur, although sometimes, he says, mysterious forces collude to raise the collective consciousness around a particular composer. But the important thing for Rattle is that distinct and informed conductorly perspectives must all be celebrated. Ticciati’s way is not his way, but Rattle admires enormously how he tackles the 1851 revision of the Fourth Symphony: ‘Robin makes a clear case for how the revised version can retain the radical edge of the 1841 version. Still it sounds like a fireball and I take my hat off to him.’

Which Fourth Symphony? That’s the most fundamental decision any wannabe Schumann conductor must make. To programme the 1841 version is to agree with Brahms, who owned the autograph score and wrote: ‘It is a real pleasure to see anything so bright and spontaneous expressed with corresponding ease and grace.’ He found the revised version charmless and stodgy, and Rattle and Holliger concur with Brahms, and each other, that the first version is much preferable – although they choose to do notably different things with that information. ‘Schumann made the revised piece in a depressive state,’ Rattle says, ‘and Brahms was completely right about the relative merits of the two versions.’ Holliger adds that Schumann’s orchestra in Düsseldorf, which premiered the new version, was nowhere near as honed as the standard of playing he had become accustomed to in Leipzig, while Schumann himself ‘was heavier, and moved and spoke more slowly’. But the pertinent point for Holliger is that Schumann retained his high-velocity metronome marks. Rattle chooses to ignore the later rewrite – Holliger gives us both but attempts to play the 1851 version, as he says, ‘retaining the true spirit of the earlier version’.

Holliger reminds me that he met Rattle 40 years ago when the young conductor invited him to perform Richard Strauss’s Oboe Concerto with the Bournemouth Symphony Orchestra. And Rattle clearly remains in awe of Holliger’s status as a Schumann guru – ‘Ask Heinz, when you speak to him, to tell you about the tempo relationships in the symphonies and about his extraordinary discovery in the fourth movement of the Rhenish Symphony.’ And I’m happy to take my cue from the Music Director of the Berlin Philharmonic.

On paper, and in the mind, Schumann’s Second Symphony registers as the most conventionally ‘symphonic’, its four movement groundplan – with a slow introduction breaking into an Allegro trot – putting you in mind of the first two Beethoven symphonies or of Haydn. And as I began to reacquaint myself with Schumann’s symphonic world, I pondered how a composition that felt instinctively unified melodically and motivically could also sound so disparate and varied, like each movement acting as a standalone character piece (not that you would necessarily want that). Holliger provides an answer.

‘The first and second movements,’ he tells me, ‘have the same metronome mark of crotchet=144, and the slow movement is nearly half; then the finale is in a very fast one-beat-per-bar, but still you feel like each bar matches the beat of the slow movement. The whole symphony is in one, like the conception of Mendelssohn’s Scottish Symphony.’ Holliger explains how the music is glued together throughout by a four-note cell, but I ask him to tell me about the music’s disunity. Am I right to hear each movement orbiting independently too, in a way that is uniquely Schumann? Holliger alludes to Bernd Alois Zimmermann, the composer of Die Soldaten, Photoptosis and Requiem für einen jungen Dichter, who died in 1970, and who was famous for pieces that made liberal use of collage and knitted together layers of borrowed material. ‘He was fascinated by the idea of Kugelgestalt – that time is like a ball, and all times of all centuries are focused in one single point. I think Schumann understood this too. You ask about the Second Symphony – well, the beginning could be like 17th-century polyphony and then, suddenly, it looks 120 years or more into the future. You feel this composer knows the whole history of music.’

The Second Symphony’s Scherzo has something of the lightness of Mendelssohn’s A Midsummer Night’s Dream, Holliger explains as he tells me about those angels and demons, ‘but is relentless, a diabolic dance, in the mood of ETA Hoffmann’. And that shockingly abrupt change of mood, the solo flute overtaken by a brutal march as the first movement of the First Symphony reaches its climax, is another characteristic Schumann moment. ‘In the First Symphony the flute symbolises a butterfly which here is overwhelmed by very tragic music. Marches are a frightening thing. Send soldiers to kill, and you’re asking them to stop thinking about what they are doing. Trills in Schumann, like the woodwind trills you mention, often tremor and shiver like music with a high fever – this is not the Baroque idea of a trill as ornamentation.’

Holliger talks about the symbolism of instrumental identity in Schumann’s music. In Overture, Scherzo and Finale a choir of three trombones appear suddenly like a premonition of the role they will take in the fourth movement of the Rhenish. ‘When his brother Eduard was dying, Schumann woke up at three in the morning. He had been dreaming about three trombones, and later he learnt that his brother had died at 3am. Always in Schumann, three trombones is a message about death.’ I mention that Rattle urged me to ask him about this same movement. ‘Well, when I looked at the sketches, I realised that the tempo changes to double the speed two beats later than in the printed score – nobody ever does this, but the difference is essential.’

That Schumann had such specific ideas about orchestral colour and instrumental identity runs triumphantly contrary to that tired cliché about his orchestration being somehow inept and clumsy. In the September 2014 issue of Gramophone, Robin Ticciati revealed that, for him, the attraction of Schumann is precisely because the orchestration is so, as he put it, ‘crazy’. ‘It’s also so controlled, and the palette is extraordinary. And I think when you get to a Schumann score, the first reaction is not to go, “What is all that?” but “What does he want?” and “What’s important here?”’ Ticciati hears clues about how Schumann ought to sound orchestrally in how he ‘orchestrates’ his piano music; and in the booklet-notes accompanying his cycle, Yannick Nézet-Séguin discusses how the defined attack and decay of modern trumpets help balance the orchestration.

And so Schumann wins. The consensus, circa 2014/15, is leave well alone. ‘Schumann learnt lots about orchestration from Mendelssohn, the greatest orchestrator of his time,’ Holliger explains, ‘and he tried to have a very transparent sound in the orchestra. It’s not that very heavy “German potato soup” sound. I never change a single note in any of the symphonies.’ Rattle confirms that Schumann must be ‘light and singing, or the sound can be too brittle – the key word is sostenuto.’ The impulsive and spontaneous side of Schumann is also important to Rattle. ‘The last symphonic music Schumann wrote was the Rhenish,’ he says, ‘and the fourth movement feels like Schumann falling apart, then the finale is an attempt to cradle him in a warm embrace. And for that to work, you can’t micromanage too heavily.’ Music to Schumann’s ears, I suspect – a composer who clearly knew the value of spontaneity: ‘My symphonies would have reached Opus 100 if I had but written them down,’ he said. ‘Sometimes I am so full of music, and so overflowing with melody, that I find it simply impossible to write anything.’

The Recordings

Berlin Philharmonic Orchestra / Sir Simon Rattle

(Berliner Philharmoniker)

'Rattle makes no apology for Schumann’s orchestration, finding the myriad colours with which the composer imbued his music. Neither is there a hint of routine in the playing: the richness of the Berlin strings and beauty of the woodwind go without saying. There’s the expected virtuosity, too, in places such as the Scherzo of the Second...' Read the review

Scottish Chamber Orchestra / Robin Ticciati

(Linn Records)

'Close reading of the score combines with Philip Hobbs’s transparent surround-sound engineering (Perth Concert Hall last November and December) in a recording teeming with revealing detail. Ticciati clearly knows how he wants this music to go and his strong partnership with the Edinburgh players enables him to shape readings notable for their energy and individuality...' Read the review

Chamber Orchestra of Europe / Yannick Nézet-Séguin

(DG)

'Potential pitfalls are seemingly effortlessly skirted around: the obsessive rhythms of the Second, say, or the winding slower sections of the Fourth, which in the wrong hands can sometimes seem aimless...' Read the review

WDR Sinfonieorchester Köln / Heinz Holliger

(Audite)

'Holliger is one of those musicians who hears what he conducts from the inside, a crucial virtue in Schumann and a neat way to disqualify curmudgeonly commentators who wrongly accuse Schumann of ineptitude in orchestration...' Read the review of Vol 1 or Vol 2

Explore Schumann's life and music...

A journey inside Schumann's concertos and trios with Isabelle Faust, Alexander Melnikov and Jean-Guihen Queyras

Faust, Melnikov and Queyras are seeking to change perceptions of Schumann by recording the trios and concertos on period instruments, finds Harriet Smith... Read more

This article originally appeared in the March 2015 issue of Gramophone. To find out more about subscribing to Gramophone, visit gramophone.co.uk/subscribe