The Mahler Symphonies

A synoptic survey by Tony Duggan

Symphony No.3

 

The Third Symphony is Mahler’s hymn to the natural world and his longest work. It was largely composed in the summer of 1895 after an exhausting and troubling period that pitched him into feverish creative activity. Bruno Walter visited him at that time and as Mahler met him off the ferry Walter looked up at the spectacular alpine vistas around him only to be told: "No use looking up there, that’s all been composed by me." Mahler was inspired by the grandeur around him at the very deepest level of feeling and also by visions of Pan and Dionysus. In fact by a sense of every natural creative force in the universe infusing him into "one great hymn to the glory of every aspect of creation", or, as Deryck Cooke put it: "a concept of existence in its totality."

To deliver a convincing performance of the Third I believe the conductor must do two things before anything else. Firstly, in spite of the fact that the work falls into Mahler's "anthologising" strand, along with Das Klagende Lied, the Second and Eighth Symphonies, the overriding structural imperative linking the six movements must be a pattern of ascending steps based loosely on the evolutionary ladder within broadly-based Pantheistic cosmology. In these terms the six movements are:

1] Inorganic nature summoned into life by Pan, characterised as summer after winter
2] Plant and vegetable life
3] Animal life
4] Human life represented as spiritual darkness
5] Heavenly life represented as childish innocence which, when combined with 5, brings
6] God expressed as, and through, Love.

Mahler’s original titles for these movements were:
1] "Summer Marches in"
2] "What the Meadow Flowers tell me
3] "What the Creatures of the Forest Tell Me"
4] "What Night Tells Me"
5] "What the Morning Bells Tell Me"
6] "What God Tells Me"

The conductor who fails to see this "ladder of ascent" and make it manifest is one who makes the mistake of concentrating too hard on getting the first and last movements right and neglects the movements in between, treating them as interludes rather than steps on the journey to perfection fashioned out of the world around and beyond. The first movement must also retain a degree of independence since Mahler designates it Part I with the remaining movements Part II. This leads to the second thing I believe the conductor must do and that is render the seemingly disparate elements of the first movement into a rigorously-wrought whole when the nature of its thirty-five minutes sets it on course for structural failure. There must be no doubt on the part of the conductor as to the movement's greatness and this includes an awareness of, and an ability to bring out, the rougher edges woven into it. Any attempt to "prettify" or "smooth out" the first movement leads ultimately to a blunting of its special power and so to failure. It’s a hard thing to quantify but it’s something you know is there at a deep level at certain "way points" and in the way you can give in to its atmosphere, hallucinatory qualities and lack of doubt in itself. I think it’s also true that a conductor's confidence in the rightness of Mahler's vision in the first movement stands him in good stead for the rest. Those conductors who get the first aspect right tend to get the second right, and are therefore, for me, the greatest interpreters of this symphony.

It is very hard many decades after a first performance to try to gauge the effect a piece of music first had on its early audiences. When something has become so familiar, loved, venerated even, to try to imagine "the shock of the new" that must have seized people at the time is a tall order. But it is an idea we should try to bear in mind if we can and so should the performer. When Mahler wrote his Third Symphony he was a young man wanting to make a very big noise in the world, to try to shake people out of complacency. In the first movement it has always seemed to me that Mahler was saying to his audience, to use modern slang, "Eat my score!" and any performance of the piece that falls short of giving an impression of that attitude is just not trying hard enough. Or at least is trying too hard to be accepted in now more polite circles. So I think it takes a particular kind of conductor to turn in a great Mahler Third. No place for the tentative or the sophisticated, particularly in the first movement which will dominate how the rest of the symphony comes to sound no matter how good the rest is. No place for apologies in that first movement especially. The lighter and lyrical passages will largely take care of themselves. It’s the "dirty end" of the music - low brass and percussion, shrieking woodwinds, growling basses, flatulent trombone solos - that the conductor must really immerse himself in. A regrettable trait of musical "political correctness" seems to have crept into more recent performances and recordings and that is to be deplored. The edges need to be sharp, the drama challenging, Mahler’s gestalt shrieking, marching, surging, seething and, at key moments, hitting the proverbial fan.

Sir John Barbirolli passes this test impressively. In March 1969 he recorded the work under studio conditions for the BBC and this is now available on BBC Legends (BBCL 4004-7). No matter what observations one might care to make about his treatment of individual sections, matters of phrasing, dynamics and expression, his vision of this work was emphatically of this journey upwards in carefully graded steps. He also grasped completely the first movement's totality with no doubt as to its validity and he wasn’t ashamed of it or its rough edges and elemental texture. The opening on eight horns is vigorous, rude and raucous. The recording then allows us to hear grumbles and groans on percussion as primeval nature bestirs, even though the crucial uprushes on lower strings are a little disappointing when compared with some where they are made to really "kick". The section that introduces Pan himself contains a ripe delivery of the trombone solo and when other members of the section join in, forward and close-miked, the effect of their lament comes over black as doom. The role of what passes as Exposition is the delivery of the brassy "in your face" march meant to signify summer's arrival. Though with this being Mahler he insists on hurling the workaday world into the maelstrom. Mahler loved his marches as much as Elgar did and this one is his most joyous and so it comes over under Barbirolli. The moment of its arrival in this recording has a particular quality which I can't imagine any other orchestra bringing. If workers in Vienna inspired Mahler, Barbirolli seems to have had in mind the holiday resorts in the north of England at the height of summer some time in the past, the forties or fifties, perhaps. There's a hint of the Promenade at Blackpool: the whiff of fish and chips, the sun catching the silver paper on the "Kiss Me Quick" hats, the tang of petrol from charabancs depositing mill girls from the looms of Manchester on Bank Holiday Monday. Then at 347 we are dragged back to the natural world with all its splendour as the horns roar out the theme from the start and the Development is underway. I like the way Barbirolli balances his brass sections here. It shows the value of the orchestra having played in "live" performances before. The important passage at 530-642 is where Mahler develops on the idea of marching and he marks each section differently, something a conductor must take note of. "The Rabble", "The Battle Begins", "The South Storm" are all acknowledged by Barbirolli and this has the effect of making the music seem to comment on itself. I was also put in mind of some of the wilder sections of Ives in the way the marches, broken down into constituent moods, seem to criss-cross each other in mesmerising half-nightmare. There is some lovely playing from the cellos prior to the return of the march proper. The portamenti the players indulge are quintessential Barbirolli. But this is swept away because the march has one more appearance to make. This time I was more aware of the long crescendo that will bring about a conclusion to the movement. The frenzy of the coda, starting at Figure 74, where the orchestra explodes into wild and crazy vistas, is well brought off. Though not even Barbirolli can match Horenstein here whose LSO brass are absolutely shattering.

There is enough of a sense of contrast between the first and second movements to mark the change from Part I to Part II but not too much to deny this is the next "step" in our ascent. There's certainly no question of treating the movement as a lightweight interlude and the second movement is a lot subtler than is sometimes realised, so the conductor must lavish the same care on it he would everything else. Barbirolli’s walk through the flowers in the meadows doesn't take the pretty route. There are stinging nettles beyond the blooms and we stumble into them in the way the woodwind allows spiky sounds to come through. The rhythm is also nicely pointed when the tempo picks up, which means when it relaxes into lyricism the effect is that much more nostalgic. Barbirolli next adopts a slightly slower tempo in the third movement but this allows a little more room to make rhythmic points and bring out character. I don't think I've heard the rollicking brass descents two bars before 9 and likewise before 23 delivered quite so loudly and with such precision at such volume. Barbirolli must have drilled his players meticulously. The crucial posthorn episode, our first glimpse of humanity, is beautifully prepared but the first posthorn is closer than we are used to. However, the section between the two appearances of the posthorn makes up for any misgivings by being gloriously raucous. If the posthorn represents the first appearance of humanity then nature has the final word with the unforgettable passage at bars 529-556: a crescendo from ppp to fff followed by a diminuendo back down to pppp replete with harp glissandi. This passage has at its centre, a development of one of the bird call motifs to become "The heavy shadow of lifeless nature", rearing up on horns and trombones. It links back to the first movement and forward to the end and is a key moment of crisis that should be marked with special emphasis so we feel threatened. Barbirolli prided himself on being able to recognise highlights and climaxes in each Mahler symphony and there's no doubt he gives this passage everything it can stand. I would have liked a little more Stygian gloom for the fourth movement which is a setting of Nietzsche's "Oh Mensch" and the first appearance of the voice. Kerstin Meyer is a fine singer but you can hear too much of her for her contribution to be as mysterious as it ought to be. I did like the way Barbirolli appears to want us to make the connection between her accompaniment and the start of the first movement, though. A nice contrast arrives with the boys and women in the fifth movement and a return to the Wunderhorn world heralding dawn with bells tolling. The boys of Manchester Grammar School are nowhere near the angelic voices we are used to. These are urchins from the mean streets of Manchester and give an earthier quality to match the purer sounds of the women and the darker, warmer tone of Meyer. Compared with some, Barbirolli is more expressive and "heart-on-sleeve" in the last movement and the big-heartedness of it all is overwhelming. This is a true journey's end that couldn't have been won by this conductor in any other way. Notice Sir John’s expressive rubato and the singing line of cello portamenti. His inability to resist speeding up at moments of release later on spoils this movement's serenity just a little, though. But take that away and it would not have been a Barbirolli performance at all. The end is built to masterly fashion within Barbirolli's warm-hearted view. He presses forward in the closing pages and can't resist almost a luftpause before the last chord of all. But he keeps his timpani under control, just as he should, and justifies his view of the end as a safe harbour nobly won.

A couple of months after Sir John’s death the Mahler expert Deryck Cooke declared this "one of the finest Mahler performances I have ever heard" and I certainly concur with that. A sentiment confirmed by an international jury of critics at the Mahlerwoche in Toblach in 2000 when they gave the recording the award for best stereo Mahler recording of 1999. It's quite a close-in sound especially made for broadcast, almost a conductor's balance with every detail clear. Some may find the reproduction of the brass troublesome but with good remastering it comes over bold, brassy and exuberant like the symphony itself and Sir John's interpretation which more than makes up for any shortcomings in the Hallé’s playing. They are some way from the finest but you would have to have a heart of stone and a pair of ears to match to let occasional lapses in ensemble and fluffed notes bother you very much. There is poetry here, there is drama, and there is a performance that reflects a world of feeling now gone.

Testament have given an official release to a "live" Berlin Philharmonic recording of the Third conducted by Barbirolli from 1966. Even though this is the Berlin Philharmonic the standard of the playing falls below what you would expect from that orchestra and, as with their Mahler Second with Barbirolli, there is just not enough familiarity with the music for this to challenge the Hallé version on BBC Legends.

Another of the work’s greatest interpreters was Jascha Horenstein whose Unicorn recording of 1970 is, for the moment, still available (UKCD2006/7 and also in a boxed set of symphonies by various conductors on Brilliant 99549). The playing of the London Symphony Orchestra is remarkable for character, unfailing alertness and ability to reflect every aspect of Horenstein’s view of the work. The result of a number of "live" performances. The introductory section of the first movement is gutsy and elemental, not at all a comfortable start. Just the kind of impression Mahler must have had in mind when he pointed Walter’s attention to the mountainous landscapes. Notice how the first trombone solo, heavy with funeral dread, conveys a sense of expectancy. Notice too how Horenstein can vary his approach straight after to take in delicacy. It’s Horenstein’s total grasp of every aspect of the first movement and his matchless sense of structure that welds the movement into an expressive whole and rivets the attention throughout. It also allows him to mark a real spiritual aspect in the episode of the march in the way it approaches from a distance before bursting on us and coming to a climax that is, like the opening, raw and rugged. I’ve always believed Horenstein was aware there is a lot more than mere programme music here. Notice how order and chaos seem genuinely pitched against each other in the central section where the marches meet. In this we can witness an aspect Arnold Schoenberg drew attention to. That this movement (and the symphony as a whole) is a struggle between good and evil. Horenstein certainly conveys struggle here to a greater extent than many conductors do. The close of the movement sees the performance emerge on the side of the angels but not before Horenstein delivers the most breathtaking account of the closing pages themselves. At Fig. 74, where harp glissandi introduce an explosion of brass, Horenstein grades the brass dynamics from fortissimo, through piano and then up to triple forte, with the latter absolutely shattering. No other conductor on record quite matches this moment. The crashing and pounding percussion that follows are really abandoned also. Magnificent.

The second movement is, as with Barbirolli and as we will find with Leonard Bernstein, the perfect Prelude to Part II and distinguished again by the playing of the LSO’s woodwinds. Horenstein also notes the darker sides of the movement, realising these are not just pretty blooms in the meadow being depicted, but weeds too. In the third movement there’s a hazy, nostalgic feel in evidence, but when muscularity is called for, as with the first movement, Horenstein is not found wanting. The posthorn solo is played on a flügelhorn making this one of the most distinctive accounts before us. Notice also how Horenstein pays attention to the phrasing of the woodwind around the solo. The great "way point" of this movement, the rearing up of raw nature prior to the gallop for home, finds Horenstein and his players really on their toes. The "Oh Mensch" fourth movement is dark and atmospheric but detailed also. This is a perfect tempo for this movement and so Norma Proctor is given all the space she needs to make every word clear. Clarity is also the keynote in the fifth movement where the boys are a joy – sharp and cheeky in the way they burst in on the silence. Though intensely beautiful in parts, Horenstein doesn’t neglect the drama and tension implicit in the sixth movement and doesn’t stand in the way of the great beauty and sense of contemplation. This great Brucknerian also brings out the qualities the movement seems to inherit from that composer in the music’s sense of slumberous growth. The end emerges naturally with the final timpani notes very prominent, a feature of this recording, which leads me to say the sound balance is not ideal. It favours the winds with the lower strings especially further back in the picture than they should be. But this is the only cloud on the horizon of this classic recording. In lesser hands this symphony can sag in parts. Never once under Horenstein is there any sense of that. His concentration is stunning and every bar seems to have something to say. This remains one of the greatest recordings of any Mahler symphony ever set down and I think it always will.

Over the years my high regard in this survey for these two recordings by Barbirolli and Horenstein have generated more critical comment than any of my choices across the whole synoptic survey both in private e-mails and in public internet forums. True, there are more who will go along with my estimation of the Horenstein recording, but even I have to admit I plough quite a lonely furrow where the Barbirolli recording is concerned. So it goes. I will carry on singing the praises of both these recordings in the general profile. I can do no other but write what I feel and hope those interested will listen with open ears. As I say in my Preface, this survey is a personal selection.

Less disagreement greets my high regard for Leonard Bernstein in this work, of course. Of his two studio recordings with the New York Philharmonic I prefer his first one on Sony (SM2K 47 576). It’s much the same interpretation as on the later DG release but the playing of the NYPO in 1961 has more sense of discovery. I also think the earlier recording, though showing its age, is still a better sound picture overall. Bernstein is alive to every nuance of the score but, as in his recording of the Seventh from the same period, he lets the music speak for itself right the way through. That isn’t to say his reading doesn’t have distinctive qualities, not least in the first movement. At the start there’s a definite feeling of a journey beginning as the horns roar and there’s also a sense of latent energy. This is a feeling that will persist and is what infuses the great uprushes from the lower strings in the opening pages which are projected with superb attack from the New York players. As too are the woodwind choir’s squawks, like birds on a wire startled into life by some noise at dusk. Following the great trombone solo Bernstein segues seamlessly into the main exposition material where the march of summer finds him in characteristically exuberant mood. If Barbirolli’s march was Blackpool Promenade on August Bank Holiday, Bernstein’s is New York’s Fifth Avenue on St. Patrick’s Day, and none the worse for that. His explicit sense of the march material means his treatment of the crucial central episode (530-640) comes off splendidly with the tensions of "live" performance and more than a nod towards Charles Ives, a composer who must have been in this of all conductor’s mind. There is tremendous frenzy whipped up here with every subtle change of tempi taken care of and a definite sense of danger that seems most appropriate. The conclusion crowns the movement with power, grandeur and excitement combined. I could imagine some finding Bernstein’s exuberance in this movement just over the top, but this music is "over the top" to start with so Bernstein and Mahler just about balance this time.

There is lovely attention to detail in the second movement and a real sense of flight in the quicker passages. Most important of all Bernstein realises this is a prelude to what follows and there is no sense of relaxation, even though the felicities of the score make their nostalgic effect. The latter also applies to the third movement that finds a slightly more relaxed tempo than some recordings. This allows the woodwind especially to convey the charm of the music by articulating every note and for everyone else to get across a real swing to the animated passages. The rollicking brass should bring a smile to your face if you have ever heard more sober views. The posthorn solo is sweet and mellow proving, as elsewhere, that Bernstein can relax when he needs to. Around the second appearance of the posthorn he also gets his strings to throw a shimmering haze around the player which is magical. Then when raw nature rears up at 529 the effect is even more big-boned, sexy and dramatic than it might have been. The fourth movement’s "Oh Mensch" brings some rapt playing and Martha Lipton is a veiled witness. The fifth movement with the boys and women comes over remarkably restrained for Bernstein. A bigger choir might have helped and maybe this is the only movement where I feel any great sense of disappointment. Bernstein takes the last movement slowly and with great dedication. However, unlike some, he brings that kind of tempi off because he never overloads it with too much emotional weight. He seems to have realised the music has plenty of its own already in it. All flows from within, just as it should, and the attention is held from first bar to last and an ultimate triumph that is natural and solid.

Rafael Kubelik’s excellent DG studio version is currently available only as part of his complete cycle (463 738-2) but has always been for me on a par with Barbirolli and Horenstein. It has one main drawback in that the recorded balance is, like the rest of his Munich studio cycle, close-miked and somewhat lacking in atmosphere. It never bothered me that much but just occasionally I felt the need for a little more space. As luck would have it, since the first version of this survey appeared, an Audite release (23.403) in their series of "live" Mahler performances from Kubelik’s Munich years in the archives of Bavarian Radio has now appeared. It even comes from the same week as the DG studio version and must have been the concert performance mounted to give the players the chance to perform the work prior to recording in the empty hall. It goes some way to addressing the problem of recorded balance in that there is a degree more space and atmosphere and also more separation across the stereo arc. It thus offers an even more satisfying experience whilst still delivering Kubelik’s gripping and involving interpretation with the added tensions of "live" performance. There is a little background tape hiss but nothing that the true music lover need fear. So, like with the Barbirolli, (and the Scherchen and Martinon recordings) dealt with below), here is yet another "not originally made for release" broadcast recording of Mahler’s Third for the list of top recommendations.

Like all great Mahler Thirds it has a fierce unity and a striking sense of purpose across the whole six movements, lifting it above so many versions that miss this crucial aspect. Tempi are faster than you may be used to, let me stress. It also pays as much attention to the inner movements as it does the outer with playing of poetry, charm and that hard-to-pin-down aspect, wonderment. In the first movement Kubelik echoes Schoenberg’s belief that this is a struggle between good and evil, generating the real tension needed to mark this. Listen to the gathering together of all the threads for the central storms section, for an example. Kubelik also comes close to Barbirolli’s raucous, unforgettable "grand day out up North" march spectacle and shares his British colleague’s and Leonard Bernstein’s sense of the sheer wackiness of it all. (Why are modern day conductors so afraid to see this aspect?) Listen to the wonderful Bavarian basses and cellos rocking the world with their uprushing basses and those raw, rude trombone solos as black as an undertaker’s hat and about as delicate as a Bronx cheer or an East End Raspberry. Kubelik also manages to give the impression of the movement as a living organism, growling and purring in passages of repose particularly, fur bristling like a cat in a thunderstorm. Too often you have the feeling in this movement that conductors cannot get over how long it is and so they want to make it sound big by making it last for ever. In fact it is a superbly organised piece that benefits from the firm hand of a conductor prepared to "put a bit of stick about" like Kubelik does.

In the second movement there is a superb mixture of nostalgia and repose with the spiky, tart aspects of nature juxtaposing the scents and the pastels. Only Horenstein surpasses in the rhythmic pointing of the following Scherzo but Kubelik comes close as his sense of purpose seems to extend the chain of events that was begun at the very start, still pulling us on in one great procession. The pressing tempi help in this but above all there is the innate feel for the whole picture that only a master Mahlerian can pull off and frequently only in "live" performance. Marjorie Thomas is an excellent soloist and the two choirs are everything you would wish for. Though Barbirolli’s Manchester boys are just wonderful. Like Barbirolli, though warm of heart, Kubelik refuses to indulge the music of the last movement and wins out as the crowning climax is as satisfying as could be wished. This is a firm recommendation for Mahler’s Third and another gem in Audite’s Kubelik releases.

Whilst dealing with earlier interpreters on record, the name Charles Adler might be unfamiliar to many people today but he was a Mahler pioneer too who made the first recordings of the Third and Sixth Symphonies, as well as the Adagio of the Tenth, for a label he financed from his own resources. He also might have known Gustav Mahler as he’s thought to have been one of the assistants who helped train the choruses for the first performance of the Eighth in Munich in 1910. Adler’s recording of the Third was made in Vienna in 1951 with the Vienna Symphony Orchestra and, on first release, boasted sleeve notes by Alma Mahler herself. Remember that until 1960 this was the only recording of the Third available so it helped form the impression of this work for a generation of Mahlerites giving it a firm place in the history of Mahler recordings. It’s now available on French Harmonia Mundi (HMA 190501.02) or Tahra (Tahra 340). It is claimed that the occasional deviations from the score an experienced listener will notice came from Mahler himself. If true it adds interest to the recording over and above the considerable virtues to be found in it. There is spaciousness and weight to the first movement which, when allied to the distinctive Viennese playing style and sound still preserved in 1951, takes us back to another world. This can be heard especially well in the sound of the horns and in the aching lyricism of the contrasting sections in the introduction. The summer march then builds from very gentle beginnings to emerge in grandeur. All through Adler justifies his weightier, muscular approach by a miracle of concentration and by the response of his players who, whilst never the last word in security, have this music in their bone marrow. There is, I believe, a hint of what this work might have sounded like under Mahler himself especially in the mellow horns and in a hundred different ways in which the strings turn a phrase. Maybe the rougher mono recording helps but the contrast of toughness and lyricism is most engaging too. The close of the movement is built up to over a huge span and rises to a massive climax to seal a deeply impressive account.

The second movement stresses lyricism again with some perky woodwinds. Again the way the strings phrase their contributions is the kind of playing you really don’t hear today and might sound quite unfamiliar to younger listeners. But I believe it tells us a lot about this work we might otherwise miss. The third movement is rather held back in tempo but, as with Barbirolli and Bernstein, benefits from this in having time to allow the myriad details to make their effect, woodwinds especially. There is real atmosphere conveyed, not least in the posthorn played by its Viennese soloist "to the manner born". There is one bad edit after the second posthorn where two sessions seem to have been spliced together with two different tempi to match, but try not to let that bother you. Hilde Rössl-Majdan gives a surprisingly passionate performance of "Oh Mensch" and, rather like Bernstein’s recording, this leads to a much gentler account of the fifth movement. The last movement under Adler is then very pure and ethereal in parts. The body of strings is not as large as it might be and it’s in the last movement this shows most. However, I still want you to be aware of this recording for all that it can tell us about Mahler performing practice. The mono sound shows its age a little, but a few minutes getting used to it is all that’s needed to adjust and enjoy a fine performance with many virtues.

Whilst on the subject of Mahler pioneers you should be aware of a "live" recording by Hermann Scherchen and Leipzig radio forces from 1960. This is now available in a Tahra release (TAH 497-498 coupled with the Tenth Adagio) giving you a "live" performance by that most individual of Mahler conductors that should provide you with a fascinating alternative view well off the mainstream. Let me also at this point mention in passing another superb radio broadcast recording of this symphony that I think demands general release. It’s by Jean Martinon and the Chicago Symphony Orchestra and has been commercially available as part of a large and expensive commemorative box. For that reason I will not deal with it in any detail. Suffice it to say that I consider it the equal of the great classic recordings for all the reasons I have tried to set out. Surely a label could be found who would release it singly.

So far I have dealt only with recordings from before 1970. So I think I am justified in calling them recordings from a previous era of conductor and sound. They are certainly analogue, all of their conductors are now dead, but they still impress, still seem in touch with a view of this work that seems to have gone. When writing the first version of this survey I was hard pressed to then find recordings from the more recent past and in digital sound that I thought came anywhere near the achievements of the versions dealt with above. But I did have a go and I see no reason to strike any of those that I included out as they are fine achievements and still worth consideration.

Giuseppe Sinopoli with the Philharmonia Orchestra on DG (447 051) is as good a place as any to start. The sound recording is superb - bold and rich, with lots of unobtrusive atmosphere. There is also splendid playing from the start with clear and "up-front" lower string uprushes that have an extra element of impetuosity about them. The trombone solos are splendidly ripe, and note also the cracks from the bass drum here and right through which are wonderfully caught. Sinopoli is aware of every colour in Mahler’s special sound palette as he is also of the rhythms in the march whose tension he builds inexorably. When the summer arrives, Sinopoli delivers exuberance but just stops short of Bernstein in this. Note the superb trumpet playing prior to the start of the development and also in the central crisis where the marches join battle. Here Sinopoli’s structural grasp is as sure as Horenstein’s, aiding his ability to convey the struggle for good over evil that Schoenberg noticed. The closing pages are a culmination not just of the re-start of the march but of the whole movement.

Can Sinopoli maintain such a promising start? Indeed he can. Delicacy is the watchword in the second movement, especially the care shown to the inner string parts and the way the music is moulded, but not excessively so. Sinopoli can frequently be heavy-handed in Mahler but here his touch is a light one. The same applies to the third movement but this doesn’t prevent Sinopoli from bringing great swing to the heavier scored parts that emerge with life-enhancing drama. When the posthorn solo arrives, the delicacy already noticed carries us into a dream landscape, enhanced by one of the best accounts of the solo on record (John Wallace?) leading to a genuinely awesome delivery of the Nature arrival ushering in the fourth movement. Under Sinopoli and sung by Hanna Schwarz, this is suitably crepuscular which makes the bright and breezy fifth movement a real wakening to the day. I feared the last movement would be where Sinopoli’s judgement would desert him and he would spoil everything by pulling the music around too much. Not so. What he gives is a noble and warm account with climaxes that don’t overwhelm, rather that seem perfectly natural parts of the whole, as do the final pages’ timpani contributions which are never allowed to swamp the texture. Sinopoli’s Mahler Third is one that should be a leading contender for the library. Superb sound, playing and interpretation.

Jesus Lopez-Cobos and the Cincinnati Symphony Orchestra on Telarc (80481) give a lively opening that brings distinctively grainy trombone sounds like nothing else on any other version. There are also some great "kicks" from the lower strings as they burst from the depths. This is not a pretty opening, in fact it’s quite ugly. It’s one of those in the tradition that sees this music cut from the landscape with the bluntest of instruments. Lopez-Cobos then has another surprise or two in store with a dance-like quality to the lyrical passages that accompany the opening and then a somewhat agitated account of the first trombone solo with violence lurking in the background. Summer itself has an airy, open-air quality and some energy to it that’s refreshing and the very immediate recording balance helps him in what appears to be a much more radical, Wunderhorn view of this work. No attempt to smooth out the shifting moods and sounds. The March sections are superbly done, prepared for with some tension and delivered with vigour and the close has all the architectural security of Horenstein and the colour and blaze of Barbirolli. Notice too how Lopez-Cobos and his engineers make you hear all the woodwind contributions.

The second movement is fleet-footed and very precise. A refreshing account indeed which puts Lopez-Cobos in with those who lavish care and attention on this short movement. I especially like the character of the woodwind and the transparent textures, which are carried over to the third movement. Here there is a lovely rhythmic snap in the more animated passages and a post-horn solo dreamy and distant. In all, an account of this movement that covers all aspects. I also felt Lopez-Cobos had in the back of his mind the sound of the Fourth symphony in these two movements, reinforcing in my mind the impression that he doesn’t lose sight of the fact that this is a Wunderhorn period piece. Michelle de Young is a rapt and sonorous soloist in the fourth movement with Lopez-Cobos in excellent support. In the fifth movement, he shows again his ability to illuminate elements others miss. Like the string accompaniments, which seem to receive special attention and some fine playing from the Cincinnati orchestra. This adds to a fine sense of flow that carries over into the last movement making the kind of culmination that it ought to be. Lopez-Cobos is a touch more detached in his textures than others are here but not so much that it detracts from his flow. It’s a fine alternative to the more "heart-on-sleeve" conductors since Lopez-Cobos has a lighter touch that pays dividends in that the optimistic side to Mahler wins out in the end. I’ve always believed this to be Mahler’s least troubled work and it’s good to hear Lopez-Cobos appears to have reached that conclusion too. The sound recording is less sumptuous than, for example, Sinopoli. But I enjoyed its detail and musical sense and hi-fi fans should note this is encoded in Telarc’s "Surround Sound".

Simon Rattle's recording on EMI (56657) with the City of Birmingham Symphony Orchestra followed live performances but I wish they had issued one of those. Having heard a broadcast of one I felt the presence of the audience gave the players a greater sense of unfolding drama. The sound here is rich, deep and well upholstered. Very much a concert hall balance with a wide spread left and right and good front-to-back perspective. The CBSO horns open the work with a sense of space, both physical and musical, with each note spaced out more deliberately and the horns themselves sounding less penetrating than I think they should. Overall the brass of the CBSO are more cultured and cushioned than the Hallé for Barbirolli or the LSO for Horenstein, so they offer a better blend. But there is some loss of character. The trombone solos are little too well mannered too, I think. The strings are well balanced and there appear to be enough of them for the uprushes from lower strings to really shudder from the depths. Rattle's main march is well done in terms of tempo and weight and is also very grand. But I think it misses the greater swagger of Bernstein and Barbirolli and the sense of the approach from far distance. In the passage at 530-642 where Mahler develops on the marches Rattle could have learned a lot from the example of his older colleagues. Horenstein and Bernstein never lose track of the plot where Rattle seems to have done at the start. He redeems himself in the "Ivesian" frenzy but then lets the music sag again in the long, dreamy section before the march resumes for the Recapitulation. With Rattle my attention wondered whereas with Barbirolli, Horenstein and Kubelik I remained riveted. Rattle also seems to cushion the climaxes at the end. The impression is that he might want to save something in his arsenal for later.

The second movement gets a lovely performance. Then in the third the opening woodwinds of the CBSO show great character and a more cultured and refined delivery. It's really a question of taste as to whether you prefer a more homespun sound like that for Barbirolli and Horenstein. Rattle seems anxious to luxuriate in the details of this movement where others prefer to be more extrovert. This does lead to an unforgettable delivery of the posthorn solos in the Rattle recording. The lead up is given a deliberate slowing down which in "live" performance was a piece of concert hall theatre worthy of Furtwangler and then in the recording the player impinges into our aural imagination from a huge distance. In the interlude between the two passages Rattle then coaxes his muted brass players to cluck like an expectant hen house - "What the animals tell me", indeed. Rattle is more reined-back in the passage at the end of the movement where Nature rears up and, again in comparison with others, disappoints. He seemed more concerned with the beauty of sound that can be drawn from this moment rather than its earthy, elemental ugliness. The backward depth in the sound stage means that the fourth movement starts with a considerable advantage as Birgit Remmert emerges from way back, singing with greater insight into the words and character of her part than most counterparts. The most noticeable difference with Rattle in this movement concerns his renowned zeal for bringing out every detail of the score because this leads to a controversial decision. There is an important solo for principal oboe and cor anglais and Mahler's instruction to the player is "hinaufziehen". A friend who was at one of the live performances described the sound produced by CBSO principal oboe Jonathan Kelly as "an extraordinary upward glissando". Rattle may have interpreted exactly what Mahler asks for but hearing something I'm so familiar with played in a way I'm so unfamiliar with makes me wonder if this is a detail to far. Rattle learned the effect from an off-air recording by Berthold Goldschmidt and the Philharmonia Orchestra from 1960. Rattle also adds girls to his boys choir and so there is a difference between his fifth movement and many others. There is more warmth but I feel less contrast. I prefer Barbirolli's unvarnished honesty though Rattle's orchestral accompaniment is very telling. Rattle is restrained in the last movement. He finds a degree of expression a few notches beneath Barbirolli’s and Bernstein’s and supplies more of the inner spirituality of Horenstein which the movement benefits from. He doesn't slow up too much, though. He agrees the movement should have ebb and flow, but his ebb and flow is within narrower limits than Barbirolli’s or Bernstein’s. The string players in Birmingham have more weight of tone and seem better able to deliver a true pianissimo and more levels of dynamic than their Manchester counterparts who were, perhaps, given a separate agenda. At the close Rattle is very satisfying. If you are looking for a modern version of Mahler's Third, superbly recorded and played, with a care for detail that takes you deep into the complexities of this remarkable work, Rattle is a contender though not a leading one.

Klaus Tennstedt’s version was mentioned in passing in the first version of this survey and I realise now this was a mistake and one I am pleased to correct. He always seemed to approach Mahler’s music from its past rather than its future. Under him the symphonies emerge as works by a composer standing at the culmination of a 19th century tradition of romantic symphonies rather than at the start of its disintegration in the 20th. Sonorities are often richly and grandly presented, romantic and expressive opportunities are likely to be grasped with alacrity, astringency and harshness tends to be underplayed and tempi are frequently, though not always, expansively presented. Tennstedt still has a legion of admirers for whom he can seemingly do no wrong and though I’ve never counted myself among them I have always admired his Third in spite of reservations relating to those characteristics I have outlined. (EMI 5 74296 2 coupled with the Fourth Symphony.) One of the aspects of Tennstedt’s Mahler conducting that always concerns me most is that he seemed surer of himself when the music was dark and tragic. That he often appeared unable, or unwilling, to deliver as convincingly as others did those passages when Mahler lightens his mood and tone. It seemed that Tennstedt was "marking time" in those passages until the next chunk of tragedy or drama came along. Perhaps we could say that when Tennstedt turned to Mahler there was "something of the night" about him. But if a performance of Mahler’s music is going to do it justice it must bring out every aspect in equal measure. Only then are the full implications of Mahler’s unique qualities, his world-embracing visions, likely to emerge best and most especially in movements where he changes frequently from one extreme to the other. I never really felt Tennstedt really did that. The first movement of the Third is one such movement, perhaps a paradigm, and is therefore a "graveyard" for conductors who cannot bring this aspect off. The opening paragraphs see him as a Sisyphus pushing his rock with the accent on weight and drag. Few versions are as doom-laden as this and it is certainly a memorable account of this part of the score. The problem is that when the lighter music arrives, with woodwinds chirruping and squawking in the dovecotes and strings lifting the music aloft as if those birds have flown, the mood seems to remain dark whereas it should change profoundly to signal the pattern for the rest of the movement. The great trombone solo is also surprisingly tame where it really ought to be rude and raucous. It’s as though Tennstedt wants to keep this as a creature of the dark also. Likewise in the build up to the march crisis in the development there is the sense of Tennstedt waiting for the moment when he can unleash his forces in mass attack which he does do with great effect. So I think he again misses the musical equivalent of montage film editing that gives equal attention to every passage rather than some. There are impressive things in this movement, though. The sound of the LPO horns roaring at the climax of the exposition’s march of summer, for example, and the close of the whole movement with brass and percussion sweeping all before them. But Horenstein, Barbirolli, Kubelik and Bernstein all have a better grasp of every feature of this movement.

The rest of the symphony under Tennstedt works much better, though it cannot be said too often that an account of the Third where the first movement doesn’t convince is a Third with one hand tied behind it’s back. It might well be because each of the following five movements essentially has just one mood which Tennstedt can therefore stick to. Just to prove he is capable of the light touch the second movement is warm and beautifully pointed with a carefree air. The playing and recorded balance is alive to every colour and this carries over to the third movement where a nice feeling of urgency also gets injected into the system. The two posthorn solos are superbly atmospheric and notice the violins in the passage between them and the splendid woodwind squeaks just prior to the second entry. It should go without saying Tennstedt manages the great rearing up of nature’s power at the close of the movement with awesome effect. The fourth movement finds Ortrun Wenkel a more open and expressive soloist than we are used to though I would have liked, once again, more contrast for the entry of the boys in the fifth movement. However, the real surprise and pleasure comes in the last movement where Tennstedt confounds expectations to deliver one of the best accounts I have heard. Too many conductors take the arrival of this movement as the signal to slow down, even seeming to try to outdo each other as to how slow they can take this music, some stringing it out into glacial progress. But this music is an anthem not a wake and Tennstedt keeps things moving forward so that the underlying tension is never allowed to flag and neither is the attention of the listener. I’ve heard accounts of this music where I have frankly become bored by it. By keeping his eye firmly on the closing pages and when these arrive delivering them without overheating the emotion, Tennstedt brings the work home on a really triumphant note. Not at all like Klaus Tennstedt, in fact.

These were the most outstanding of the modern versions that I included in the first version of this survey. The intervening years since have been very good to this symphony and whilst I still think Sinopoli and Lopez-Cobos in particular still deserve their leading places there are now other newer recordings that are as fine and, in three cases, even finer. The question at the top of my mind this time around in this survey is whether any of these could now challenge the old guard, give me the Holy Grail of a modern digital recording by a conductor of our own time worthy to be listed in the same breath, something I failed to find last time. I will tell you now that whilst some new recordings come very close there is one, at last, that I think does meet that formidable criteria, but more of that one later. Let me first deal with two new recordings that I think don’t quite make the grade but are included here because they are borderline and they illustrate better the virtues of the ones that do.

Above I wrote that I think it takes a particular breed of conductor to turn in a great Mahler Third. No place for the tentative and no place for the sophisticated. The greatest interpreters have all knocked about the world and been knocked about by it. Andrew Litton with the Dallas Symphony (Delos DE3248) gives every impression of not falling into this category as what he gives us is an all too sophisticated, contrived and ultimately complacent reading that makes me wonder if he really believes in Mahler’s vision or whether he isn’t, in effect, rather embarrassed by it all. Attention never flags in the immense first movement but neither is there what you could call an attitude. Which means the performance is not marked out for distinction from those who have gone before. Rather that Litton appears daunted by the forces Mahler’s imagination unleashes and he has decided the best thing to do is get out unscathed, which he does and with much aplomb. But is "aplomb" appropriate in this movement? A crucial passage is between bars 530 and 642 where the March that dominates the animated sections does battle with the primeval forces to see who is dominant. It should be the scene of abandon, danger and struggle. Under Litton it’s just an example of fine orchestral playing and sound recording where the level of attack seems blunted. So often in other passages there is the feeling Litton cannot bear to let things get too much out of control. The usually awesome climax at 367-368, where the enhanced horn section is left bellowing at the universe, Litton again hangs fire. The second movement has elegance and charm and seems to suit Litton’s style more. Mahler wrote about the third movement: "This piece really sounds as if all nature were making faces and sticking out its tongue. But there is such horrible, panic-like humour in it that one is overcome with horror rather than with laughter." A tall order for the conductor which only the best come close to matching. Litton’s animals all sound too Beatrix Potter to me. The lovely Posthorn solo is well brought off, but even here I thought Litton and his player go for the saccharine bringing us music more suited to a candy advert. Then in the passages between the two solos we are faced again with that problem we faced in the first movement. There is no sense of the dangerous abandon for it to bring us close to Mahler’s "horrible, panic-like humour." There is also that crippling habit Litton has of holding back when he should let rip which shows itself especially in the amazing passage from 529-556, a crescendo from ppp to fff followed by a diminuendo down to pppp that Mahler describes as "the heavy shadow of lifeless nature". Nathalie Stutzman has a full and verdant tone and fine sense of words in the fourth movement but the problem is Litton pushes her and the orchestra along too much. The reading of the great last movement that Litton then gives is sweet and intense to start with. It is possible for the attention to be allowed to wonder unless the conductor has a clear idea of where is has come from, where he is now, and where he’s going. The only aspect I’m aware of with Litton is a desire to beguile the ear. The orchestra plays well but doesn’t, as yet, have the ability to convey the idea they are reaching back into a real tradition of playing this music.

After listening a number of times to the version by Benjamin Zander (Telarc 3SACD-60599) I made the "mistake" of listening again to the recording by Hermann Scherchen and also the unreleased one by Berthold Goldschmidt already mentioned, both from 1960. Straightway I knew what I had been missing. These two great Mahler interpreters of the past may not be blessed with the kind of rich and detailed digital sound that Zander is given but such is their uncanny and innate understanding of the deep structures of this work that matters of sonics cease to matter. In Scherchen’s case he is even labouring under the disadvantage of conducting an orchestra that would struggle to be called second rate. No matter. Such is the players’ grasp of what Scherchen is doing that even their technical shortcomings cease to matter all that much. In the case of Goldschmidt he had before him what was then one of the world’s best orchestras - in fact the same one as Zander, albeit of forty-three years ago. I must say that on this evidence the Philharmonia of 1960 knew their Mahler more intimately than their counterparts of 2003. Surprising because in 1960 they had never played the work before yet still had it within them to bend their collective spirit in a manner of playing, a tone of musical voice, that now seems lost. Both the older conductors project a symphony full of ambiguity, cocky self-confidence, naïve poetry, warmth of heart, wonderment and an emotional richness that comes not from an outside-in imposition but percolates out from the core, all in overarching, urgent, forward-moving structures that have you on the edge of your seat from first note to last. Just like my preferred stereo versions from pre-1970, in fact. It is a Mahlerian truth that a performance of the Third that fails to bring off the first movement successfully and idiomatically is fatally wounded. That is the case with the Zander recording. The horn-led opening under him is powerful, leonine and vividly projected, but not nearly elementally seismic enough. The high woodwind trills which become scattered right through the movement seem far too regimented and cleanly delivered to approach the demented squawks Mahler surely intended. The trombone solos are well played but, as with the woodwind trills, are still too contained, not rude enough. All of this is symptomatic for me of Zander not really "getting" this symphony. Under Zander there seems in the whole, long introductory passage of the first movement too literal a presentation of the material, a feeling the desire is to present the notes rather than what lies beneath them. The great march of Summer sees the bands beautifully turned out and well-drilled, though there is in the recorded sound an edge to the brass when playing full out that is tiring on the ear. Following the horn section’s crowning of the climax at the mid point of the movement the lead-back to the return of the march and the stormy variation of it leaves me with the impression that Zander didn’t really know what to do with this transitional passage. That he’s just longing for that storm to come up. The battle of the storms is not as bone-shaking as it could be. It’s a stiff breeze rather than a hurricane. The coda, capable of being the most exciting music that Mahler ever wrote, is ruined. Zander presses so hard down on the accelerator I was put in mind of the way Furtwängler used to conduct the coda to the last movement of Beethoven’s Ninth. The orchestra just about hangs on, but all nature-storming grandeur is knocked down in the rush.

The second movement does contain some nice touches, in the string playing especially, but the slides are strictly controlled, the phrasing calculated. The third movement fares better with more of what has been missing in warmth and involvement, though there is still an impression of the metrical. Every rhythmic jump and jerk superbly prepared and executed. But are the animals in the forest really like that? Then there is the post-horn solo. This wonderful effect in the third movement is one of Mahler’s greatest master-strokes. In this recording Zander calls for his soloist to use a genuine post-horn and the instrument is even described for us in the notes. The problem is that the player is set so far in the distance that you can barely hear what he is playing. You can, of course, turn up the volume control but you would then have to turn it down again quickly when the whole orchestra joins in. In the two choral movements it was a pleasure to hear the warm tones of Lilli Paasikivi and the vitality of the Tiffin boys who all lead into a consoling and grand final movement where, at last, there is a glimpse of what a great Mahler Third really can be. The Philharmonia Orchestra plays well throughout and the recorded sound is rich, though the dynamic range is huge as I indicated when discussing the post-horn. Fix a volume setting to contain the all-out passages with comfort and you lose detail in the quiet passages.

The newer recordings that now follow do, I believe, challenge the older ones more closely than the above two. Michael Tilson Thomas’s first version with the LSO on Sony boasted the best contralto of all in Janet Baker and a wonderful coupling of Baker singing the Kindertotenlieder. The Third recording had virtues but without quite convincing me it deserved promotion to front rank. However, Tilson Thomas has now re-recorded the Third as part of his ongoing San Francisco cycle (SFSO/Avie 821936-0003-2) and this one is a definite improvement. Make no mistake, this is a well-played, well-recorded, enjoyable and involving performance. It is only when it is compared with the older recordings that you start to hear what is missing. If you must have a supplement in the very latest recorded sound then you might consider Michael Tilson Thomas’s new version, but read on. There is in the first movement still not quite enough of the rough-edged, rude banality I’m sure Mahler meant us to hear which must have so shocked his first audience. This shortcoming is all the more sharply felt when contrasted with the nature painting Mahler provides to go with it. Cases in point are the great trombone solos. In the older recordings mentioned above these come over almost as a force of nature stressing bloated fecundity. Tilson Thomas’s soloist is a fine musician but his relatively backward placing in the sound picture and his straight-faced delivery of this rude, cheeky music is not powerful or coarse enough. Under Kubelik an unforgettable raw assault bears down on you like the earth being ripped apart. Horenstein and Barbirolli also pull this effect off. This is a small aspect, you may say. However I think it indicative of the overall tone of the first movement under Tilson Thomas which, by a crucial gnat’s whisker, fails to convey the "life or death" struggle Schoenberg noticed. Maybe it’s the space Tilson Thomas gives the music in the first movement that makes it fall short on the urgency aspect. Just over thirty-six minutes is long even for this movement. I can admire the grandeur, though. Taken with his care for the lyric aspects it certainly engages right the way through. There are some carefully prepared string tremolandi in the introduction, and the woodwinds squawk tunefully on cue every time their dovecotes are disturbed. I always think Mahler’s birds should be more Alfred Hitchcock than Percy Edwards. This is certainly the case from Kubelik, Barbirolli and Horenstein and all the better to round out the picture. The great march of Summer which crosses and re-crosses the movement is done with gusto and panache, as you would expect from this conductor, though I found his tendency to over-control detracted from the "in your faceness" Mahler surely wanted. This march should just let rip and be its rude self no matter how coarse it might get. All of this remains the impression to the end of the movement: grandeur contrasted with lyricism, urgency and edge are downplayed by too much control. From Kubelik there’s terrific forward momentum, even in the repose passages, and no lack of the uglier, coarser aspects of nature to go with the lyric ones. From Tilson Thomas there are a few of the colours missing, the primary ones, and not enough sense of danger.

Tilson Thomas’s control of the second movement is strong too, which gives it an admirably taut quality but then detracts from the sense of intermezzo that perhaps it should have. There are some impressive things from the orchestra here, though. The third movement emerges naturally from the second and is most enjoyable. The post-horn solo is a little lacking in character, both in sound and delivery. Beautifully played but no real attempt to "sound-paint" a mood. The great coda to the movement, where nature rears up to bite our heads off, is delivered splendidly with tremendous portent and fear. Full marks to the horn section for the lungpower. Michelle de Young sings the fourth movement with a matronly operatic vibrato I didn’t take to at all. Something more disembodied is called for here. Whilst the boys in the fifth movement are pure and bell-like to suit the words but I miss the Manchester lads from Barbirolli or the Wandsworth boys from Horenstein for their sheer cheeky edges. One of the many appeals of Mahler’s music is how close it takes itself to edges without quite falling over them. This puts conductors on their honour to save Mahler from himself when they can. The Andante to the Sixth always seems to me a step short of kitsch. Likewise the last movement of the Third seems to me a step short of mawkish if not handled correctly. Like the slow movement from Bruckner’s Eighth this is, for most of the time, a meditation not a confession. I think Tilson Thomas’s "heart on sleeve" is too close to his cuff so the music palls rather. I’m well aware that many of you will love it and will swoon at this kind of treatment. I wish you well with it. For me something a little more detached goes a longer way, saves Mahler from himself, prevents his music being turned into our own personal psychiatrist’s couch. At the start of the music part of me thought I was listening to the opening of Barber’s Adagio and that can’t be right at all. Go back to Kubelik for the right balance of "heart on sleeve" and cerebral repose and you will see what I mean. But that’s not the whole story of this movement, of course. The end should be triumphant and under Tilson Thomas it really is just that. The heart is warmed by the journey’s end and this goes some way to making up for any reservations I may have over the rest. I’m happy to stress pros rather than its cons here. The San Francisco Orchestra is on fine form and they are recorded with depth and spread in a realistic sound picture that packs a punch when needed but can pare down to intimacy too. It must be said that they don’t have the last few ounces of tone colour variation that mark out the greatest Mahler orchestras from the others, woodwind especially. Their brass section too is rather soulless, especially when playing all out.

Michael Gielen has been recording Mahler Symphonies for a number of years with his Baden-Baden orchestra and the results have been rightly admired. He approaches Mahler essentially from a 20^th century viewpoint, seeing him as a composer looking forward rather than backward. In that aspect we hear Mahler in the clear light of day, instrumental lines clear, the sharp edges in his sound palette thrown into relief, the romantic and emotional effects not so much played down as left to their own devices. Some may find this last characteristic disappointing: a barrier between them and music that they think should move and thrill them more. But when so many conductors seem happy to connive with those who wish to use Mahler as their own personal consulting room I believe Gielen, like his predecessor at Baden-Baden Hans Rosbaud, presents an important and refreshing point of view and would urge you to try it (Hänssler Classics CD 93.017). One of the most notable aspects of the long first movement of the Third under Gielen is his deliberate tempo for the march that dominates it, crossing and re-crossing like armies over a familiar battlefield. There can be few recordings where this is given with such swagger and emphasis as here and I liked it very much. I also liked the fact that Gielen encourages his trombones to really observe the written glissandi at the start that others seem to almost wilfully ignore. These are the kind of touches you would expect from Gielen: examples of his gimlet eye for radical detail that also means he is never dull, always with something to say. The recording balance helps too and I was especially impressed by how much you can hear of Mahler’s dense textures. The many string tremolos shimmering and the woodwinds squealing and squawking above the heaviest scored of passages are examples of this. The attention to the kaleidoscopic textures is shown at its best in the section of the development where Mahler pitches the March material into furious battle. Gielen keeps track of every line of the score for us. Not for him any attempts to smooth out the music into more palatable form.

The second movement shows a nice contrast between the pastoral minuet material and the more energetic trios. Again notice the snaps from the strings and the squeaks from the woodwinds. The third movement then seems to grow directly out of the second with some perky, cheeky woodwinds at the start but a very pure and ethereal trumpet solo in the remarkable central sections. Not for Gielen a flügelhorn here, as in Horenstein’s recording, for example. Perhaps that would present a little too much charged nostalgia. However Gielen manages plenty of power in the extraordinary passage at the end of the movement where Mahler depicts nature rearing up like a great prehistoric monster. In the fourth movement the contralto Corneila Kallisch is placed forward and sings well but the most notable sound you will take away from this movement is that of the oboe. Gielen instructs his soloist to observe Mahler’s marking "hinaufziehen" and perform upward glissandi, as with Goldschmidt and Rattle, though in Gielen’s recording this effect is a little less obtrusive and I could be persuaded to accept it as played here. The mood of the fourth movement should then be broken sharply by the entry of the boys intoning the "Bimm-Bamms" of the bells in the fifth movement but in this recording they really don’t do that, appearing to be set too far back to make much impact. I also think the boys sing too politely and sweetly even in a recording where we are kept at greater distance. I longed for Horenstein’s urchins at this point in an object lesson in how this movement should sound. After this the last movement is played with great restraint. A restraint many will find runs dangerously close to a detachment for music that has so much heartfelt emotion at its core and can stand some coaxing even in an interpretation like this. Certainly Gielen misses the inner spirituality others bring. But that is not the effect Gielen is aiming for overall and we must accept that or ignore his recording completely. The playing of the orchestra remains true and committed to the end rounding off what is still a fine and interesting performance. This is a worthwhile and challenging recording of Mahler’s longest work, fresh and clear. Not a first choice, but certainly one to compliment versions offering more personal involvement by the conductor and I believe it to be worth your consideration.

Claudio Abbado has now recorded the Third Symphony for DG for a second time (471 502-2). His first version was studio made in Vienna and notable for its grasp of detail, even though I always felt there was something crucially missing in the direct communication department. Something that a "live" performance has every chance of redressing alongside offering a more mature interpretation. This new performance was given by Abbado and the Berlin Philharmonic in London in October 1999. It was first broadcast by BBC Radio 3 and DG acknowledges the BBC in the liner credits so this could be said to be a harking back to the type of performance not meant for release that I praised earlier. The audience is impeccably behaved and the orchestra on top form. Perhaps they tire a little towards the end of the long evening, but that is what happens in concerts and only adds to the sense of occasion and really should worry only those who always demand the often clinical perfection of the studio. The extraordinary introduction section to the first movement is outstanding here for the acutely perceptive balancing of parts and sections and for the sense of a slow, inexorable forward momentum projected beneath the considerable degree of portent that Abbado brings. The lower string uprushes could kick a bit more but this might well be more to do with the recorded balance. Then notice the way the tone of the music lightens in the pastoral interlude at bars 57-131. The BPO delivers this material with a bright, golden tone so that when the terrific snarls arrive from the bass drum as the opening material reasserts, it is that much more vivid when seen in such contrast. The fact that Abbado is so convincing in these two most important faces of this movement bodes well. The performances of the first movement that come off best are those that don’t shy away from the kaleidoscopic nature of a piece brimming with youthful exuberance and, most especially, sheer nerve. There had never been a symphonic movement like this before, after all, and you know Mahler knew it. Another equally important face to the movement is the great march of summer that comes so much to dominate everything that it should, in the very best performances, give the impression of even threatening to take it over. Under Abbado it seems to begin from far away and then advance towards us before bursting out in its summer glory. However what I don’t hear, certainly not to the same degree, is the sheer bumptious effrontery of it all that I do get with Barbirolli, Horenstein, Bernstein and Kubelik. They also deliver better the primeval elements of the movement, all that dirty bass-end grumbling and shuddering, that must also have come as such a shock to the first audiences. The climax to the march’s first procession (347-368) where the massed horns roar to the skies comes off very well under Abbado because here is a horn section that can be both powerful and retain great beauty of tone. But again the previous versions I mentioned manage it better because they seem not to care how bold or crude they sound here. The development then begins with that lyrical, golden music Abbado gives with even more warmth than before, allowing him to then segue effortlessly into the return of the march where the battle between good and evil that Schoenberg so perceptively noted can really be enacted. Notice here the Berlin double basses’ precision and the woodwinds’ shrieks. Nowhere does Abbado’s modernist soul allow him to smooth out or prettify Mahler, let me assure you. The "battle of the marches" (530-642) is suitably exciting with the impression of forces champing at the bit to be released and I think the fact that this is a "live" performance helps here. From recapitulation to coda we are taken in one grand arch but there is a real lean towards the grandeur of the music under Abbado - a "grandstand" end to the extraordinary musical events we have just heard which begins even in the solo trombone. Then when the coda swells to its massive climax, broad and with plenty of space, Abbado’s expansive approach is capped and justified. In all the first movement certainly holds the attention across its immense span because Abbado has the belief written through his interpretation that you cannot and should not try to contain this music. He is also blessed with an orchestra that is on top of the movement’s demands even under these concert hall conditions and seems to respond to that challenge.

Abbado appreciates the importance of the second movement and so makes it memorable by paying it the same attention to detail he has to the first. He sets out the five-part structure very particularly. He also achieves by his colouring of the winds the important fact that whilst the flowers that are being portrayed in this movement can smell nice they can also sting. The playing of the Berliners is again beyond praise in giving pin-sharp ensemble and great beauty of tone, shifting and darting between the various episodes, responding to Abbado’s little dabs of colour and to his minute, but so telling, changes of tempo. All of which are carried over to the third movement which Abbado, quite rightly, sees as the next step up the level of ascent he has now set himself upon and seems to grow naturally out of what has just gone. Under him this movement manages to be both energetic and lyrical at turns and pretty well covers all bases, though Barbirolli, Horenstein and Kubelik yet again take the more raucous passages even further than Abbado who holds them in by comparison. I also feel the crucial posthorn sections, that most evocative sound in all Mahler, whilst admirably played and positioned in the sound picture are a little stiff. In the fourth movement Anna Larsson is superb in her delivery of Mahler’s night song to Nietzche’s "O Mensch!". Since his first recording Abbado has also has come over to the school of thought that believes the oboe soloist (and later cor anglais) should interpret Mahler’s hinaufziehen marking in the solos as an upward glissando. The two local choirs sing well in the fifth movement but there is some attack missing from the children who are not helped by their backward balancing. That said, Abbado does catch the feeling of a fresh day awakening the symphony demands here and provides a fine prelude to the delivery of the last movement. Just when you thought this performance couldn’t get any better, it does. The last movement has all the concentration of chamber music playing in a noble and spiritual reading that grows in emotion and warmth and it progresses. Notice especially how in the later pages Abbado manages to correctly recall moods from the first movement, binding the vast structure together prior to an ending that is uplifting and focussed - pulling on the heartstrings but never in danger of snapping them. The enthusiastic applause from the full house at the end is given an extra track on the disc so you can programme them out if you want. The sound is spatially very wide with impressive left/right and front/back spread and, crucially, much more air around the sound than we are used to in this hall. Instrumental detail is still very clear but I do wonder whether some of the impact of certain passages may have been better left as this hall usually delivers them to microphones. Dynamic range is wide but comfortable and believable. The effect is like sitting in a seat quite far back in the hall and contributes to the concert hall atmosphere.

There are, of course, many other versions available in the catalogue but those I have dealt with in detail are for me representative of the best in this work. As so often we are comparing the excellent with the outstanding and the outstanding with the immortal. Nevertheless, let me round up a few more that I would not include at all in case anyone wonders where they are. James Levine on BMG is good but I always feel there’s too much gloss and polish, especially in passages that demand a more "homespun" approach. He also drags badly in the last movement. The last movement is also the main problem in Lorin Maazel‘s version with the Vienna Philharmonic on Sony, though not the only one. This is all a pity because the Vienna Philharmonic is magnificent and the same applies to the sound recording. Bernard Haitink is slow in the last movement on his second recording with the Berlin Philharmonic on Philips, though not as slow as Abbado was in his old VPO version. The main problem with Haitink is his first movement where he’s just too "foursquare", phlegmatic and sane for music that should retain at least a touch of madness. I’ve heard him better "live" in this work, as he is on a performance preserved in the multi-disc Phillips collection of his TV Christmas Mahler concerts where the Concertgebouw Orchestra plays to the Mahler manner borne. Sir Georg Solti on Decca whips up too much excitement in parts of the last movement and, as always under him, I find the brilliant Chicago Symphony brass section inappropriately brilliant all too often. Esa-Pekka Salonen’s Los Angeles version on Sony has been widely praised. For me it’s let down by too smooth an approach in the first movement especially, and this is aided and abetted by a bass-rich recording more recessed than most with the result that too many details get "melded" into the whole and pass us by. Riccardo Chailly’s Decca version is another from their Amsterdam production line - impeccably played, too backwardly recorded, all the appropriate boxes ticked, but ultimately uninspired and badly lacking in the effrontery department. Much the same applies to Kent Nagano on Teldec. In the case of Pierre Boulez on DG (4742982 - a hybrid CD/SACD) the presence of the Vienna Philharmonic and a DG recording team on top form makes this one of the best sounding Mahler Thirds now before us. However, I really cannot this time empathise with Boulez’s seeming unwillingness to engage with the very elements of the work which I find so crucial. A creative detachment, so admirable in other symphonies, just seems, to me, misapplied here. Interestingly, his performance of the last movement is transcendently moving but his portrayal of the first seems wounded by his inability to bend with its many contours and byways and some of the tempi are on the slow side. The second movement too doesn’t seem to possess enough warmth of heart though the third is more appealing. So, with regret, and with praise for the sound and the playing, I shall pass over this version even though I am certain Boulez has delivered the very performance that he meant us to have.

For those on a limited budget let me draw attention to the version by Zubin Mehta and the Los Angeles Philharmonic on Decca (443 030-2). This is a "Double Decca" set and couples a reasonable account of the First Symphony with Mehta conducting the Israel Philharmonic so clinching its bargain status. Mehta’s Third is a ripe and vivid account, well played and brightly recorded, though not in the front rank. Also for bargain hunters there is Erich Leinsdorf Boston Symphony account on BMG (63469) who also couples a good First Symphony so again is worth the price asked.

As I indicated above, at the time of writing the first version of this survey I was of the belief that I would probably wait in vain for a modern digital recording by a living conductor to challenge the great versions of the past - Barbirolli, Horenstein, Bernstein, Kubelik, Scherchen etc. One was certainly needed because this symphony really benefits from the latest sound as the digital versions proved in spades. Gielen and Abbado come the closest of all of those so far dealt with, closer than the digital versions dealt with last time, and yet there is still one recent digital recording that I do think is worthy in performance terms to be spoken of in the same breath as the immortals and I turn to it now. If what I have to say about it seems briefer than some of the reviews above it is because I have so few reservations about it. Quite simply it is so good that it is just a sheer pleasure to listen to from start to finish. The first thing to be said about the Semyon Bychkov’s version on Avie (AV 0019) is the detail of the sound recording that lets you hear every aspect of instrumentation of this great score in excellent proportion and balance to an extent that is still surprisingly rare even in the digital era. Not just a question of the fact that it is digital but also because the balance engineers have done their jobs properly. I suppose some might call it a "close-in" balance. For me the description "no frills" springs better to mind. It is as if you have a seat in the hall near the front of the platform. There are, after all, recordings of this work where a too reverberant balance robs us of hearing just what a revolutionary canvas Mahler presents us with. Highest to lowest frequencies are accommodated with thrilling definition here and the highs and lows in this symphony are very high and very low indeed. Next is the excellence of the principal players of the WDR Sinfonie-Orchester Köln whose contribution is heard to thrilling effect by the sound balance. This is the orchestra that recorded the superb Shostakovich symphony cycle under Barshai and the Mahler cycle under Bertini.

The opening massed horn call is a call to attention, almost like a fanfare here, that sets out the stall from the start with clear intent. "You will listen to us," Bychkov appears to want to say before a refreshingly sharp delivery of the opening slow march follows with rhythms very sharply pointed and the up-rushes of the basses articulated with razor-like precision. Spring may be trapped by Winter but this is a Winter with real bite. I like the lyrical contrast of the second theme that comes next but I like even more the way its fundamental precision seems to be an appropriate counterbalance to the opening. As, for example, in the bracing way in which the important trombone soloist has been instructed to play with a rude health that is so often missing in the more polite interpretations. This spills over into the march which, with the closer recording, the excellence of the players and Bychkov’s sense of the unadorned, puts me in mind of Bernstein and Barbirolli in its proletarian kick and real sense of lift and determination. Notice especially the middle of movement climax, just prior to where the horns come back and blast to the four corners, how the woodwind choir lets out a great sustained high shriek that ushers in the horns with a climax to take your head off. Now there is the shock of the new. Or should that be the "shock and awe" of the new? In the return to the opening material again the deep frequencies are superbly rendered with bass shudders to shiver anyone’s timbers before the march comes back with renewed swaggering, ballsy confidence. The end when it comes is carefully prepared, never rushed and again every detail is heard. With a first movement like this we have in front of us a Mahler Third of rare quality. From the start through to the end you become aware of a conductor who has thought through this movement anew and has the sense that he is telling a story. This is a "live" performance before an audience too, though you would hardly know it.

The instrumental contributions to the second movement maintain the quality of the first. It is a fine contrast to what has gone, just as it should be, and also pays as much attention to the sharp, tart elements as to the warmth. The third movement serves the early Mahler song on which it is based very well seeming to twist it slightly almost as if it is being "sent up" which is quite probably what Mahler intended when he wrote: "This piece really sounds as if all nature were making faces and sticking out its tongue. But there is such horrible, panic-like humour in it that one is overcome with horror rather than with laughter." The posthorn is closer than it often is but this seems well in keeping with the general approach and there is no diminution in nostalgia. The rearing up of fecund nature at the close is as dark and lowering as you could wish for, frightening in its immediacy here. I also like Marjana Lipovšek in the fourth movement’s "Oh Mensch" movement very much. Especially her dark, portentous tone and care for the words. The oboe and cor anglais plays the sliding glissandos now the fashion but the effect does not seem to jar as much as it did. Maybe I’m getting more used to it. Fine and lusty boys usher in a finale that is judged to near perfection with radiant. glowing strings, light and dark perspectives, a beautifully sustained line that is crowned by a truly liberating and life-enhancing coda. You will have gathered by now that I rate this Bychkov recording very highly indeed and I write not with a new review disc that has just arrived on my desk and I have heard for the first time only in the previous few days. I have lived with this for many months and many playings and look forward to doing so for years to come. Here at last is a Mahler Third from the modern era worthy to go with the greats of the past and this is now my top recommendation for performance and sound. Often the most difficult reviews to write are the ones where the performance just seems to work and be right. You just put down your notes and listen as if anew to a work you thought you knew so well. That is the case with Bychkov in Mahler’s Third and I recommend it to you enthusiastically.

In the end, in terms of performance and interpretation alone, Horenstein and Barbirolli still remain supreme in this work for me with Bernstein and Kubelik (now available singly and in good open sound on Audite) very close behind. Hermann Scherchen is hors concours occupying the same kind of place that Stokowski‘s Second does in my survey of that work. For outstanding performance and interpretation that is in modern digital sound Semyon Bychkov is triumphantly top of the pile and he is followed by Abbado and Gielen in their differing ways and these are certainly honourable mentions as still are Sinopoli and Lopez-Cobos and Tilson Thomas too. Of all the recordings I have heard since writing my first version of this survey, though, only Semyon Bychkov’s is the equal to the previous generation versions and only time will tell if he surpasses any of them. For a great and enduring Mahler Third, for now, the old guard do still have it. But all of these recordings meet the criteria I outlined at the start whilst still reflecting different aspects but never once losing track of this huge work and its special qualities. They bear out what the young Arnold Schoenberg wrote to Mahler about this symphony:

"I saw your very soul naked, stark naked…. I felt your symphony. I shared in the battling for illusion; I suffered the pangs of disillusionment; I saw the forces of good and evil wrestling with each other; I saw a man in torment struggling towards inward harmony…. Forgive me, I cannot feel by halves."

And, where Mahler’s music is concerned, neither should we.

Tony Duggan

Barbirolli Hallé  BBC Legends BBCL 4004-7   Amazon UK   Amazon US
Horenstein LSO UKCD2006/7   Amazon UK   Amazon US
Bernstein NYPO Sony SMK47590       Amazon UK  Amazon US
Kubelik Bavarian Radio Audite 23.403 Amazon UK  
Adler VSO  TAH340341 Crotchet
Sinopoli Philharmonia DG 4470512 Amazon UK  Amazon US
Jesus Lopez-Cobos Cincinnati SO Telarc 80481 Amazon UK  Amazon US
Rattle CBSO  CDS5566572    Amazon UK  Amazon US
Tennstedt LPO EMI 5 74296 2 Amazon UK  Amazon US
Zander LPO Telarc 3SACD-60599 Amazon UK  Amazon US
Tilson-Thomas SFSO Avie 821936-0003-2 Amazon UK  Amazon US
Gielen Baden-Baden Hänssler Classics CD 93.017 Amazon UK  Amazon US
Abbado BPO DG 471 502-2 Amazon UK  Amazon US
Mehta LAPO 443 030-2   Amazon UK   Amazon US
Bychkov WDRSO Avie AV0019 Avie