mahler symphony bis vanska

Gustav Mahler (1860-1911)
Symphony No. 9 in D Major (1909)
Minnesota Orchestra/Osmo Vänskä
rec. 2022, Orchestra Hall, Minneapolis, USA
Reviewed as a download
BIS BIS-2476 SACD [82]

In 2016, BBC Music Magazine polled 151 conductors around the world and asked them to rank, in each conductor’s opinion, their top twenty symphonies from the orchestral repertoire and Mahler’s Ninth was fourth.  This represents a huge turn-around from 50 years earlier, when Mahler was just barely starting to become core repertoire, having previously been spared from obscurity only by a few conductors such as Bruno Walter and Otto Klemperer, who did what they could to keep Mahler in front of the public.

It used to be said that an orchestra had truly made its mark on the recording industry when it had a set of the complete nine symphonies of Beethoven in the catalogue, but that has changed today, as everybody wants a complete set of symphonies of Mahler in the catalogue.  That gives Mahler fans a great richness of choice but it also increases the odds of missing out on the best the market has to offer.

One such set of Mahler symphonies currently in progress is from BIS Records with Osmo Vänskä and the Minnesota Orchestra.  As of now, they have released everything on the list, including Deryck Cooke’s “performing version” of the unfinished Tenth (review and review) , except for Mahler’s monumental Eighth Symphony, which was recorded in July 2022, and the Third, which was recorded in November 2022.  They will hopefully both be released later this year.  I am not aware of any plans for them to record Das Lied von der Erde, although there is an earlier 1994 recording of Vänskä conducting Schoenberg’s chamber arrangement with the Sinfonia Lahti Chamber Ensemble (BIS-CD-681, still available).

Regarding the Ninth before us in this review, there is a wealth of very stiff competition from many of the world’s greatest orchestras.  Mahler’s Ninth seems to bring out the best, both from conductors as well as orchestras, as there are very few truly “bad” recordings of it.  The 1982 Karajan-Berlin live concert recording (Deutsche Grammophon 4390242, still available) is probably the best-known ever since Richard Osborne in Gramophone declared it to be “one of the seven wonders of the modern musical world”.

There is certainly a lot of good to be said for that recording, but for the purpose of this review, as a comparison reference I am using Riccardo Chailly’s 2004 recording with Amsterdam’s Royal Concertgebouw Orchestra on Decca (review) I chose this because, firstly, there is no orchestra in the world with better Mahler credentials than the Concertgebouw and also because I could compare the SACD’s high-resolution multi-channel surround sound with the equivalent from BIS in the Vänskä-Minnesota recording – although sadly, the SACD version of the Chailly is no longer readily available.

I downloaded the Vänskä recording from BIS’ retail website,  It is also available for download in high-resolution stereo and standard CD audio, and there is also a physical SACD disk available, whose multiple layers also contain all three incarnations of the audio presentation.  The booklet notes, by Jeremy Barham, give a thorough historical background for the Ninth and a well-considered discussion of its musical attributes and formal structure.

Looking back over MusicWeb International reviews of previous Vänskä-Minnesota Mahler recordings, I find that a fairly consistent pattern emerges:  BIS Records’ quality of sound engineering is described as reliably excellent.  The Minnesota Orchestra acquit themselves brilliantly, proving to be an excellent Mahler orchestra and that the pattern holds true with this latest recording; their performance is virtually flawless. BIS’ sound has a completely natural-sounding front-to-back transparency where everything is audible, from the softest pianissimo to the loudest fortissimo, all in complete, unstrained clarity.  The perspective seems to be from the middle of the front seating section, with an ideal spread from side-to-side and from front-to-back.  We are close enough to hear an enormous amount of detail, but also distanced just enough to get sufficient amounts of the hall’s ambience and bloom for the louder passages to have the most effective impact.  In short, we have the best seat in the house.

I should mention that I also listened to some sections of the high-resolution stereo and CD audio versions to compare with the multi-channel surround version.  As to be expected when removing the surround channels, there is not quite as much background ambience from the hall’s acoustics, but otherwise, the stereo layers both maintain similarly excellent sound reproduction, so whatever your listening setup is, you need have no concerns about the quality of audio on this recording.

Vänskä’s overall timing is 82 minutes, which is fairly conventional and puts him in the middle of the road, alongside Karajan and Haitink, although just a tad brisker than Klemperer.  Timings can be deceptive, however, as they reveal virtually nothing about the underlying interpretation.

The beginning of the symphony is well-judged, the tempo quite suitable for Andante comodo.  As with previous Vänskä-Mahler recordings, the violins are divided left and right, allowing us to better experience the dialogue between the 1sts and 2nds.  As we progress through the exposition’s revealing of the two conflicting theme groups (one in D Major, the other in d minor), the clarity of BIS’ recording becomes apparent; when the texture of Mahler’s orchestration begins growing thicker and more complex, we can still hear it.

A problem also begins appearing at this point, however, one that will not abate for the course of the whole recording.  It’s all too poised, too controlled, too restrained-sounding.  It’s all very proper and clean, with balances never being broken between sections of the orchestra, or anybody allowed to do anything ugly with their instrument.  When the score calls for the trumpets to rise above the orchestra and express the first cry of desperation in Mahler’s quest, Vänskä keeps them firmly inside the overall texture, never allowing the trumpets to pierce the shell of general balance.

Chailly-Amsterdam, however, seem to throw off all restraints right from the outset, clearly taking a damn-the-torpedoes approach.  The sound is somewhat more recessed than with Minnesota, but Decca still provides plenty of clarity, while the world-famous acoustics of the Concertgebouw’s Grote Zaal give the sound plenty of bloom for a massive impact in the louder passages.  When the score calls for the trumpets to rise above the orchestra, Amsterdam’s trumpets are fully up to the task, punching through with all the desperation Mahler is asking for here.

As we begin the development section, Vänskä gives us a nice sense of mystery and brooding darkness, but compared with Chailly, it becomes apparent that Minnesota is still holding back.  Muted brass interjections are not as nasty-sounding as they could be and sudden dynamic swellings indicated in the score are produced only just audibly.  In Amsterdam these are all rightly exaggerated completely beyond the bounds of “good taste”.

As the conflict between the exposition’s theme groups begins in earnest, Vänskä makes the mistake of going too fast.  It’s exciting, true, in a way, but it also causes a problem; going this fast means that the textures are now more crowded because the notes are zooming by faster in the audio “picture”.  This creates an issue with Vänskä’s desire to always maintain clarity, so, in order to keep the brass from swamping the rest of the orchestra, he has to back them off, which causes the impact of their parts to be seriously watered down.  This gives the development’s war-between-theme-groups a sense of merely watching a documentary instead of living the real thing. With Chailly-Amsterdam, a better-judged tempo gives them plenty of flexibility to assert the brass parts with all the power the Concertgebouw can muster while still keeping the other parts appropriately audible.  In addition, the Concertgebouw shows no regard for considerations of maintaining control or individuals not daring to stick out from the overall texture. 

As we approach the second half of the development section battle, with the final, frenzied ascent and crash into the recapitulation, Vänskä again takes a tempo that is too fast for his own good, and yet again, backs off the brass so as to not overpower the orchestra. Also, while I’m not entirely certain about this, it sounds as if the Minnesota orchestra might possibly be at the limit of its strength here.  I can hear something of a strained quality, as if they’re about to run out of steam – although there is no doubt in my mind that Vänskä is holding things down. It’s all very clean, and you can hear virtually everything on the page of the score, but we’re not looking for an academic score-study session here.

Amsterdam being Amsterdam, however, it’s no surprise that they “fire at will”, unleashing everything in their arsenal as the world crashes and burns around them, and then a Mahlerian Armageddon explodes out of our speakers with the low brass.  It sounds like one of the ugliest, most unnerving passages in the whole orchestral repertoire – exactly what it should be. In the aftermath, as rampant chromaticism and numerous parallel tritones inform us that the enemy has won this time, Amsterdam’s muted horns and dissonant woodwinds are nasty in our face, with all the distaste and ugliness of the moment, while Minnesota continues keeping it all under control and polite.

As we reach the recap proper, it gets better for Minnesota since things are less strained in the score.  They don’t have the measured, heavy-sounding weariness of Amsterdam, however, who sound tired of the world here.  The coda of Vänskä’s first movement is poised and beautifully rendered by the Minnesota Orchestra, with a wonderfully mournful flute soloist who is not afraid to linger and savor the moment.

As the second movement gets underway, Vänskä’s apparent predilection for control and balance becomes an advantage in this instance – that is, until Mahler starts putting things in his score like horn trills, over-exaggerated accents, or one part of the orchestra playing pointed, staccato figures while another is simultaneously playing legato passages, which clearly have no business being together on the same page of any well-written music.

Minnesota renders all of this perfectly and dutifully, but Amsterdam revels in it, deliberately putting all the “bad” writing front and center.  I honestly have to wonder if Vänskä believes Mahler was really serious about all the extreme markings and raucous drunkenness of this movement.  He absolutely will not let things get outside the bounds of his so-carefully delineated balance and sound texture.  Everything on the page is there, clearly audible, and the orchestra’s performance is as clean as the bleach load – but that’s not enough.

When the horns are marked sforzando, Minnesota’s horns hit them enough to push right up to the edge of the overall balance, but Amsterdam’s horns “tastelessly” blat them out to the point of buzzing their bells, bullying their way right to the front of the sound picture.

At the end of the second dance section, just before the ländler’s return, Mahler marks the low brass with ff and sf markings.  Minnesota plays them loud enough to be clearly heard amid everything else going on without ever getting uncouth about it, but Amsterdam’s trombones and tuba are positively (and deliciously) flatulent here.

In the third movement Rondo-Burleske, Minnesota’s rendering is, not surprisingly, very clean, very precise, well-controlled, and never ever improper, whereas Amsterdam’s Rondo-Burleske belongs squarely in the red-light district on New Year’s Eve – and in Amsterdam, that’s really saying something.  Everybody jumps in to do everything they can to “spoil” the proceedings, whether it be flutes with fortissimo flutter tongues, brass players occasionally and deliberately over-blasting all the strings, or clarinets using the most unsavory barnyard tone quality

In the much quieter center section, beginning with the trumpet giving us a Wagner Turn-infused melody, Minnesota is more up to the task since this part is not so irreverent.  The thing that hurts them here, however, about their overall approach is making the melody “sing”.  Most people do not understand how to make a melodic line “sing”; there has to be a constant ebb and flow of the melody’s volume level, even if only at a subtle level, with the volume always constantly increasing or decreasing, so that there is always a sense of either going somewhere or coming from somewhere. 

This is something that, more often than not, you won’t find indicated in the sheet music.  The number of markings required to indicate it all would make the notes illegible.  Musicians simply have to decide for themselves, within the context of the score’s currently marked dynamics, where the melody’s volume is going to rise and where it’s going to descend.  Minnesota’s rendering is somewhat minimal and understated since Vänskä has them doing only what’s on the page.  Amsterdam is more overt about playing through the notes and sustaining the melodic line’s coherency and sense of direction.

As this middle section begins winding down, elements of the Burleske begin creeping back in.  Minnesota dutifully renders it all 100% correctly, exactly as indicated in the score, but Amsterdam has the daring to blare these interjections right to the fore, rudely interrupting the reverie. During the ending section’s chaos, Minnesota makes sure we can properly hear everything on the page, but in Amsterdam, even the percussion are toasting their ale with the winds now, snare drum rolls “badly” out-balancing the whole string section. 

Finally, the last movement.  Mahler has spent three movements fighting a war between death and life (and losing), spotlighting the pointlessness of the gentry in the drunken dance movement, turning loose the chaos of life without purpose in the Rondo-Burleske, and now he finally gives in and takes on a resigned acceptance of the inevitability of it all, with the hope that death will, in the end, provide some answers and give some meaning.

My earlier point about the melodic line in the Rondo-Burleske’s middle section continues throughout the whole of the fourth movement, unceasingly.  Incidentally, the fourth movement’s primary melody is based on the Wagner Turn figure from the Rondo-Burleske’s middle section, now transformed by Mahler into something considerably more profound. In this movement, Minnesota is playing a tune, whereas the Amsterdam is singing a hymn – a hymn to life and to the release of death.

My overall points about the Vänskä recording should be abundantly clear by this point, so I am going to discuss only three specific spots in the last movement: first, at 9:30 in the Vänskä recording, the hymn tune rises to its biggest climax yet, fortississimo with a sforzando on top of that, accentuated with a crescendoing bass drum roll beneath the whole orchestra.  This is all followed by a subito pp on the very next note, clearly a very dramatic change.  Vänskä does the volume change perfectly, but Chailly adds one little detail that makes a huge difference.  At the end of the last fff note, Chailly inserts a split-second pause, barely long enough for the note to fade from the Grote Zaal’s reverb, leaving a silence that is every bit as “loud” as a full orchestra playing at full power.  It’s effect in rendering Mahler’s anticlimax here is heartbreaking.  Vänskä could argue quite correctly that there’s no such indication in the score – but I don’t think Chailly has any need to defend himself at this spot.

Secondly, at 14:21 in the Vänskä recording, we have the final all-or-nothing climax, not only of this symphony, but of all of Mahler’s completed symphonies, of his entire life.  The trumpets scream out the hymn tune’s Wagner Turn, but now reduced to semitones, their chromaticism making Mahler’s angst here searingly obvious.  This is followed by the violins’ emphatic unison descent, leading into the hymn tune’s most impassioned statement. As in the first movement’s climaxes, Vänskä refuses to turn the trumpets loose here, keeping them constrained within the overall balance.  We can hear them, they’re there all right, but this should be Mahler standing on the rooftop, screaming to the world at the top of his voice for the last time ever – but Vänskä is more concerned with keeping the textures balanced.  Amsterdam’s trumpets scream all the way up to the heavens with shattering force.

The violins, during their unison descent, are marked Wieder Zurückhaltend, (“holding back”) in reference to the tempo.  On top of that, Mahler also wrote ritardando and then after that, molto ritardando, yet, Vänskä hardly slows down at all, just playing right through it, nearly at tempo.  It’s dumbfounding; after spending so much effort throughout the whole symphony up to this point being so emphatic about doing exactly what it says in the score and not one iota more than that, now at this point when we’ve reached the climax of the whole symphony, he ignores what the score says? Chailly-Amsterdam, of course, give it everything Mahler asked for, and maybe then some more.  The effect is absolutely magnificent, powerful beyond words.

One final detail to discuss: the ending of the symphony.  Most Mahler devotees are quite familiar with the long, drawn-out ending, containing numerous silences between fading and diminishing whispers of the hymn tune as it fades away from life.  Bernstein’s Amsterdam recording is (in)famous for drawing out this section beyond the breaking point, to where it loses all sense of melodic coherence.  Fortunately, Chailly holds things together more successfully with the same orchestra.

Mahler wrote it out, being very specific how long he wanted the silences to be, giving each one a specific number of beats in the score.  The fact that he made them of differing lengths should be a conclusive indicator that he did not intend them simply as general pauses, but rather, as specific, individually measured lengths of silence. All the conductor has to do is patiently and methodically count through the beats (silently, of course) before bringing in the next melody fragment.  (There’s much more to a successful performance than just this one isolated detail, of course, but this is the specific aspect that needs to be pointed out.)

Vänskä short-changes the silences.  He does not count through them, and does not give them the time Mahler allotted them in his score.  It gives the impression that he is, perhaps, awkwardly uncomfortable with silence in a performance.  Sadly, for the listener, it cheapens the ending and detracts from its poignancy – and again, after so much previous emphasis on following the score so precisely, this makes no sense.

The Vänskä-Minnesota Mahler Ninth is a good recording, reproducing nearly everything on the pages of the score very faithfully (with two exceptions noted above); but in a Mahler Symphony, that’s not enough.  Mahler himself said that, to properly understand a score, it’s important for the performer to “read between the lines”. We might have the most clear and accurate recording ever made and yet it can still sound like it’s just going through the motions.

What concerns me, especially, is the thought that new Mahler listeners might acquire this recording and think this is what Mahler is supposed to sound like, that this is all there is to it – in which case, they would be missing out on so much more that is in there, as we discover when listening to performers with a better understanding of the Mahler idiom.

David Phipps

Help us financially by purchasing from

Presto Music
Arkiv Music