The original review

Brahms Piano Concerto No 2, Op 83

Leon Fleisher pf Cleveland Orchestra / George Szell

Sony

It is immediately evident that Fleisher and Szell conceive the music in terms of great strength and spaciousness from the way in which they build through from the opening horn-call to what follows, but what is perhaps less evident is their scrupulous care about details of rhythm and phrasing – details that all contribute to the performance’s ardour and vitality. My one complaint is that there is not enough really quiet playing from the orchestra: Szell, like so many conductors working in America, seems not to cultivate the intense piano cantabile that is one of the glories of the best British and central European orchestras. This tendency is felt again in the Andante; Brahms is careful to mark its opening paragraph on the strings no louder than mf even at its climax, but here it works up to something distinctly more than that. This means, inevitably, that a certain inwardness and intimacy is missing, or at any rate is only occasionally suggested; but there is a great deal to make up for that – above all, the combination of ardour with what I can only call dignity. Nothing sounds scamped or hurried, everything is given time to tell, though without letting it hold up the course of the music. Thus although one may occasionally feel that Fleisher and Szell are not quite getting to the heart of the music, there is no sense of perfunctoriness such as we get in more conventionally ‘high-powered’ performances; theirs is a consistent and responsible conception of the music, presented with the greatest skill and care. I thought I detected a trace of distortion in the orchestral climaxes of the second movement on my stereo copy, but this may well be confined to the test pressings and is in any case not enough to prevent my recommending this latest version of the Brahms B flat as the best available. Fleisher may not have Rubinstein’s impeccable keyboard elegance, and he may not yet dig quite so deeply into the music as Serkin or Arrau sometimes do; but what he does have is a sort of grand seriousness that is absolutely appropriate to Brahms, especially combined with Szell’s dynamic but no less serious account of the orchestral part and recorded sound quality that stands comparison with any other version. Jeremy Noble (8/64)

Rob Cowan Some of the finest recorded Brahms B flats have emerged from America: think of various versions with Rudolf Serkin (three with Eugene Ormandy, one with George Szell), Emil Gilels or Van Cliburn with Fritz Reiner, Sviatoslav Richter with Erich Leinsdorf and Arthur Rubinstein with Josef Krips. Gilels and Serkin for me were always among the best (the incendiary pairing of Gilels with Reiner has no equal, especially in the Allegro appassionato second movement), but when Leon Fleisher arrived on the scene in 1964 in league with Szell and his Clevelanders, their ‘scrupulous care about details of rhythm and phrasing – details that all contribute to the performance’s ardour and vitality’ (to quote Jeremy Noble), their blend of refinement, strength and tempered virtuosity, revealed aspects of this massive score that were hitherto uncharted, at least on disc. Noble mentions a lack of ‘intense piano cantabile’ that we in Europe tend to take for granted, but remember he was listening before the advent of CD, which has facilitated a marked improvement on the dynamic curve of American orchestral recordings from this period.

Jeremy Nicholas I confess I hadn’t previously put the two together: Brahms No 2 and America – but you are quite right, of course. And not only do all the classic recordings you cite come from that side of the pond but they were (almost all) made within the period 1958-67. By that time, everyone seems to have agreed more or less on the optimum tempo for each movement, so before addressing the merits of Fleisher–Szell, I’d like to raise a question. All of the recordings you mention are a world away from a recording of the Second which I have always been drawn to if only because it is said to reflect more closely than others the tempos that Brahms intended: Rubinstein with Albert Coates in October 1929 gets through the opening Allegro non troppo (no allegro maestoso for them!) in a mere 14'33", the Andante in just 9'10"; a far cry from Gilels and Eugen Jochum in the 1970s clocking in at an enervating 18'17" and 14'04" for these same movements. The middle way (those recordings you quote above) seems preferable, but I wonder if we have not lost something of the original spirit of this work with these more leisurely iterations.

RC Interesting point. ‘Leisurely iterations’ are no good for this most epic of piano concertos, I agree. I’d imagine that Brahms himself would have rolled out the solo part with rugged impetuousness, varying the music’s pulse and volume according to his mood and what he picked up from his audience. Tempo, I feel, is less of an issue than weight, attack and a strong bass line. For me the problem with the Gilels–Jochum Berlin version is less about tempo than about its relative luxuriance; it’s quite different from its faster, more spontaneous-sounding 1950s Gilels–Reiner Chicago predecessor. As to Rubinstein under Coates (of which I too am fond): it predates a period when the pianist took himself off to hone his technique and temper his youthful ardour, a process that had the baby on the brink of following the bathwater – judging by his later recordings of the work. Rubinstein would ease into the phrase, whereas with the likes of Robert Casadesus (1958) and Wilhelm Backhaus (1952), both under Carl Schuricht, the general idea was to throw caution to the winds. I feel that that was Brahms’s intention with the B flat Concerto, quite unlike the calmer, more measured D minor First Concerto. Fleisher and Szell followed suit in the B flat, as did Serkin and Szell not long afterwards.

JN I completely agree – spirit above tempo (not that they are mutually exclusive). But I’m glad you reference Backhaus, because he is another I returned to at the start of this project to refresh my memory – not the Schuricht version, but the 1939 account with Karl Böhm. Its opening, up to the first tutti, is to my mind the best of all introductions. This made me think. Whenever anyone has asked me which is the best version to have of Brahms’s Second, I have always said Fleisher and Szell; but then I reflected, ‘Is this my own received wisdom that’s talking? It’s some time since you last listened to it. What are the qualities that led you to form that opinion way back?’ Well, there’s no throwing caution to the wind as in the early Rubinstein and Backhaus accounts. Instead, we are presented with a lofty view of glowing perfection: Szell and the Cleveland at the peak of their considerable powers; and – with Brahms No 1, a Beethoven cycle, Mozart K503 and Rachmaninov’s Paganini Rhapsody already under their belt – the inspirational partnership of Szell and Fleisher.

RC There’s certainly a glow about their Brahms Second – it’s more polished than the Szell version with Serkin, less ‘widescreen’ than Serkin’s stereo account with Ormandy (1960). It’s a big performance, hefty yet rich in detail. Take the soloist’s thoughtful opening ascent, then the beautifully articulated cadenza and Szell’s sweeping first tutti – such a bounce to the rhythm; and tight, immaculately dispatched pizzicatos. Once it’s up and running, you sense there’s real dialogue happening between the two, the sort of Romantic ‘conversation’ that marks the B flat as being so different from the more marmoreal D minor. At 6'46", there’s that wonderful solo ‘confessional’ (at least that’s how it sounds to me, basically the tutti’s second subject), so perennially unexpected, though here lacking the sense of shock that springs from versions involving Backhaus and Casadesus. The qualities that mark out Fleisher as so special are, in my view, a combination of strength and control, clarity too: you don’t miss a note; in fact, you could virtually tabulate the score from listening to it.

JN Yes, clarity is the watchword here – from both parties. Fleisher manages to combine tremendous power with the utmost delicacy. I don’t know how he plays the triads in both hands at around 9'50" (first movement) and the unnecessarily heavy left-hand notation at 0'54" (second movement) with such a nonchalant light touch. Marvellous! The Cleveland’s contribution is why I enjoy this version so much. Szell (nicknamed ‘Old Flypaper’ because he could stick to any soloist) had created a musical instrument that for homogeneity of sound, phrasing and the players’ ability to listen to each other had few, if any, rivals at the time (1962). Then there are the individual contributions: listen to the duetting between the solo cello (Jules Eskin) and the oboe towards the end of the cello’s opening solo in the third movement. It’s beautiful to hear (amazing recording quality for the time). My respect and admiration is total. But is it all a bit too polite, too perfect? Does it send a tingle down the spine?

RC That’s the problem: does having all the boxes ticked necessarily add up to the right total? Demonstrate this superb Fleisher–Szell production to anyone and they will likely fall over themselves in search of adequate compliments. Flaws are conspicuous by their absence, but flaws (tumbles from perfection, clinging to the cliff’s edge and suchlike) are what make this most physically hazardous – or should I say, dangerous – concerto so involving. Take the second movement, the sudden, heroic central tutti and the knotty solo writing that comes afterwards (which, as Stephen Kovacevich once reminded me, is rarely played with all its notes intact); or the finale’s bodacious swagger, and you have music that invites scuffs and duelling scars. Attack, passion and a devil-may-care element, all cast on the grandest scale, are what make this fabulous work so compelling. So when Noble refers to the performers’ ‘scrupulous care about details of rhythm and phrasing’, do you think that he too had the score’s audible musculature in mind?

JN Indeed – I was going to use that Noble quote to make exactly that point. Yes, I do think that’s what he was getting at. To sum up, it seems we are in broad agreement. We are never going to get a Brahms Piano Concerto No 2 that has everything, but this Fleisher–Szell account gets pretty close. Personally, I don’t think the orchestral playing in either Richter–Leinsdorf or Gilels–Reiner, terrific as it is, quite matches the Cleveland’s, and later versions, for all their excellence, have not persuaded me to alter my allegiance – for instance, Nelson Freire with Riccardo Chailly (an enervating first movement); Stephen Hough with Mark Wigglesworth (unappealing sound). Still, I do wish our team sometimes had injected a bit more of the early Rubinstein zing and even elements of the much-maligned 1940 studio recording by Vladimir Horowitz and Arturo Toscanini. Or, as you would say, ‘bodacious swagger’!