Conductors have these pieces, totems you might call them, which guide them at momentous times. For Pablo Heras-Casado, as it was for Dimitri Mitropoulos, that totem is Mendelssohn’s Scottish Symphony. He could hardly have planned to make his debut in January with the Vienna Philharmonic in the piece, in place of the recently retired Nikolaus Harnoncourt. And many of his other important debuts – in London, Salzburg, Munich and Berlin among others – have come with a work that he singles out as the most original and satisfying answer to the problem of the symphony addressed in the difficult generation of the 1830s and ’40s, when so many composers were at once inspired and overawed by Beethoven’s example. ‘This piece has everything,’ he says.

It was a Berlin performance in 2010 that caught the ear of Martin Sauer, Senior Producer for Harmonia Mundi, and led to the conductor’s increasingly fruitful relationship with the Freiburg Baroque Orchestra, which has already yielded discs of Schubert and Schumann. Broadcasts of earlier performances reveal a fairly consistent shape to the piece whose exact dimensions yield to each orchestra’s individual physiognomy in sound: the Berlin Philharmonic’s string sostenuto like rings of teak, or the Bavarian Radio’s smoothly modulated wind band. With the Freiburg band, however, ‘you get many clues,’ finds Heras-Casado, ‘many answers to what Mendelssohn had in mind. In many orchestras the violas tend to lead the opening chorale, they play out, and it never works. The violas need to be a colour of the oboe here.’ This chorale both introduces the first-movement Allegro and also acts as one half of a formally innovative frame around the entire symphony, and it is distinguished by Mendelssohn’s orchestration, ‘fragile, like a leaf in autumn’ and then anchored at both ends by bass and flute in the reprise.

Heras-Casado comes close on the recording to Mendelssohn’s metronome marks – ‘I try to approach and respect them’ – but is not afraid to embrace a flexible pulse that would never work for Beethoven. ‘This comes from the material. You have to read between the lines. In the case of Beethoven, the material is so extreme, so commanding rhythmically, that there are things you cannot do. You feel the pressure of the pulse. We’re talking about another generation. There are descriptions of Mendelssohn as a conductor: he was very modern in his approach, precise, and concerned with togetherness. And he used tempo flexibility as an expressive way of shaping the structure.’

One unifying feature to all four movements is the presence of fanfares and march rhythms, but despite the finale’s public designation of Allegro guerriero, Heras-Casado does not hear a military character to these tropes which, as he notes, were passed down by Haydn and Mozart. ‘They have a heroic element, and his attitude was purely Romantic. He was writing at the same time as the beginning of archaeology, the reconstructions of other civilisations, the discovery of the ancient Egyptians, and ruins, and integrating them into his style – rediscovering Bach and Handel.’

Another striking characteristic of the Scottish is the sustained denial of closure. This is most obviously suggested by the running-together of all four movements without pause, but each of the first three lacks a harmonically identifiable sense of an ending. The high-point of the opening Allegro is placed before the recapitulation – ‘not in the Brahmsian way, where the coda brings the biggest climax. This is spacious and pictorial, moving into something else.’

That something else is Mendelssohn’s old trick of a third theme, here smuggled into the texture on the cellos by preceding it with a recitative based on previously heard material. In the studio, the Freiburg cellos sing the new theme with a prominence not practical in concert conditions. ‘You have the old theme already, you recognise it. Now everything has changed, we need to show it.’ They phrase it accordingly, with the kind of hesitations and breath-marks that are harder to achieve in a full-sized symphony orchestra. ‘The Freiburg Baroque do it naturally,’ observes Heras-Casado, ‘and I think that shows us that Mendelssohn’s orchestra were used to doing it naturally too. At that time they were used to coming from earlier music. They wouldn’t think to play this legato, it was a later development in technique. With the FBO it’s the same situation, they have spent 25 years intensively playing earlier repertoire, and they approach this music as if it is new.’

Mendelssohn again teases expectations with a false reprise of the first theme in the Scherzo, but the Adagio is perhaps his boldest reformation. Tonally ambiguous at the start, the movement’s home key of A major isn’t established until bar 10. The cantabile is a beautiful, expansive melody, but it isn’t allowed to develop before the funeral march intervenes – although that is there from the beginning. It looks like an innocent accompanying rhythm to start with but then it becomes the dominant force. ‘I’m not saying Mendelssohn was a revolutionary but Mendelssohn asked a lot of his musicans,’ says Heras-Casado. ‘The music can be beautiful but sometimes it isn’t sweet – there can be a harshness. I have to stretch the musicians. Mendelssohn was asking his players to go to the edge, struggling with those bows, as any modern composer would. This march doesn’t develop or come to a close. Nothing happens. This augmented chord [at bar 114], it gives you no reference where the music’s going. And then the flow returns.’

Initially conceived in 1829 when Mendelssohn visited Holyrood, it was only in 1841 – having dropped tantalising hints about its progress over the next decade or so – that he made the time to perfect the work. By then, his sense of timing was even more acute. Heras-Casado hears the symphony’s true climax occurring not in the long, hectic fortissimo – teasingly mis-signposted by another false recapitulation, marked by yet more martial upbeats on trumpet and drums – but in its aftermath, winding down but never relaxed ‘like in the overture to A Midsummer Night’s Dream, or the first movement of the Second Symphony, where everything is extinguished. Mendelssohn is a master at creating climaxes of pure suspense.’

Which brings us to the other side of that frame, the expansive, triumphal coda which rolls up all the withheld endings in one, long, neo-Handelian stride of jubilation at journey’s end. It has troubled listeners and performers ever since Mendelssohn wrote it (see some reactions opposite). Heras-Casado gives me a wry smile. ‘This moment is where you have to go to your backpack and look for as much gusto or good taste as possible. And you have to be creative about how you can help this very simple material. The rhythm is quite dance-like, but the texture is dense, so you need to give it some light and spark, elegance and noblesse.’ If it were faultless, how could the Scottish Symphony have everything?

The historical view

Robert Schumann (shortly after the symphony’s 1842 premiere)

The formal basis of Mendelssohn’s symphony is distinguished by the intimate relationship of all four movements…more than any other symphony, it forms a whole that is closely interrelated. 

Otto Klemperer (Otto Klemperer: his life and times. CUP: 1969)

The coda is, indeed, very strange. Has not, perhaps, the clever Gewandhaus Kapellmeister Mendelssohn here overruled the great composer? I believe now that I have the right to make a fundamental change to the coda. 

Sir John Eliot Gardiner (LSO Live booklet-note, 2015)

He had such a refined taste and orchestral palette. With a return of the symphony’s opening theme, now refashioned and ennobled, we see in our mind’s eye the majestic launch of a great warship cruising down the Clyde with cheering crowds.