Pyotr Il'yich TCHAIKOVSKY (1840-1893)
The Sleeping Beauty - ballet in three Acts with prologue, op.66 (1890) [150:28]
Raphael Druian (violin)
Minneapolis Symphony Orchestra/Antal Doráti
rec. Northrop Auditorium, Minneapolis, 10-11 April 1955
THE DORÁTI EDITION ADE019-020 [73:13 + 77:15]

The Nutcracker - fairy ballet in two Acts, op.71 (1892) [79:28]
Raphael Druian (violin)
University of Minnesota Singers
Minneapolis Symphony Orchestra/Antal Doráti
rec. Northrop Auditorium, Minneapolis, 12-14 December 1953

Until recently I had always shared my colleague Paul Corfield Godfrey's belief that the Hungarian-American conductor Antal Doráti (1906-1988) had been the first to record the complete Haydn symphonies. It took a brief flurry of e-mails on MusicWeb International's message board to alert us - and, I suspect, many others - to the fact that those particular laurels should rightly go to the rather less well known Ernst Maerzendorfer (see here).

Nonetheless, a recent spate of re-releases has reminded us that, even if Doráti wasn't quite the Haydn pioneer that we'd thought, he does have other claims on our attention. Most notably, perhaps, the contents of three comprehensive box sets documenting the Mercury Living Presence catalogue have demonstrated what a versatile musician he was, effortlessly turning his hand with almost invariably successful results to repertoire ranging from Handel, via a gamut of 19th century warhorses and lollipops, to Alban Berg.

At the beginning of his career, however, Doráti had been something of a specialist in ballet. For almost a decade from 1933, he was the Ballet Russe de Monte Carlo's principal conductor, touring with it throughout Europe, North America and Australia. He subsequently went on to lead the performances of the rising New York company Ballet Theatre - later to become American Ballet Theatre - until 1945. Even though he rarely conducted from the pit for dancers after that date, his later activity in the recording studio certainly confirmed the importance of that earlier association.

Commissioned by the Mercury label to record all three Tchaikovsky ballets in full, though in mono sound, Doráti and his own Minneapolis Symphony Orchestra produced, in quick succession, accounts of The nutcracker (1953), Swan lake (1954 – available as Mercury Living Presence 462 950-2 and included in the second of the three Mercury jumbo box sets) and The sleeping beauty (1955). Thereafter, he recorded a stereo version of the complete The nutcracker with the London Symphony Orchestra for Mercury in 1962 (Mercury Living Presence 432 750-2, see here), and yet another with the luxury casting of the Concertgebouw Orchestra for the Philips label in 1975 (now on Decca 442 562-2). His swan-song in this repertoire was a stereo recording of The sleeping beauty, made in studio sessions with the Concertgebouw over a three year period, 1979-1981, and currently to be found on Decca 478 3103.

Taped within a window of just 16 months - and marking the first time in recorded history that the same conductor and orchestra had recorded all three in full - Doráti's accounts of the Tchaikovsky ballets are pretty consistent in their approach. Equally consistent was the critical praise that they received at the time. Their release, however, came in the twilight years of mono sound and they were to be relatively quickly overshadowed. Indeed, within just eleven years of its first appearance, The sleeping beauty could be found listed in the pages of the Penguin guide to bargain records, where Doráti joined a veritable concert hall of real-life conductors whose older recordings had been relegated to cut-price labels such as Marble Arch and Wing - as well as a host of pseudonymous others on such bargain-basement outfits as Allegro and Fidelity.

Now, however, the enterprising Antal Doráti Centenary Society ( has reissued two of those 1950s mono Tchaikovsky ballet sets, with The nutcracker squeezing economically onto a single CD by the skin of its teeth. They are appearing for the first time in Europe on CD, with transfers engineered by Haydn House ( who, ironically enough, also offer those Maerzendorfer Haydn recordings on their website.

Perhaps the most important point worth noting about any recording of ballet music is its perceived rationale. We should keep in mind that its composer had been commissioned to write music to which dancers could perform on stage. Even though they or others might later create suites of individual numbers for the concert hall, they had still conceived the original works as danceable. That fundamental concept inevitably constrained, in particular, the tempi that they specified. Everything in the score had to be within the realistic physical capabilities of a dancer's body - and especially his or her feet - in action over a long, full evening on stage.

From the twentieth century onwards, however, more people have become familiar with ballet music via recorded or broadcast accounts than live in the theatre. For many of them, Tchaikovsky's craftsmanship and his gift for unforgettable melody have made him ballet’s poster-boy. Extracts from, and latterly complete performances of, Swan lake, The sleeping beauty and The nutcracker have become a staple of the recording industry. That confronts the conductor with a dilemma. Should he, in a dancer-less studio, still follow the composer's original intention and direct the music as if it were for a real, practical staged performance? On the other hand should he consider that the realisation of the score's fullest musical potential is thereby prevented. Rather more cynically should he take the view, that, in the absence of anything to look at, listeners at home need a more excitingly varied aural experience in order to hold their attention. All in all, ought he to ignore the "practical dancing" requirement altogether and adopt what is sometimes called a "symphonic" approach to the scores?

Perhaps surprisingly, given his early career history in the pit, Doráti apparently came to believe over time that performed ballet was essentially "non-musical" (see John L. Holmes Conductors: a record collector’s guide [London, 1988], p.74). In so doing he appeared to be positioning himself somewhat in the second camp.

That's not my own favoured approach. For me, a ballet is both a visual and an aural experience. I watch so much of it that, if I'm restricted to listening on CD, I want the recording to allow me to "see" believable dancing steps in my head. Not everyone shares that view. There are many who find staged ballets very hard going - especially Tchaikovsky's, overflowing as they are with tutus, twinkles, magic spells and fairy dust. They, I suspect, will appreciate Doráti's muscular, driven style of conducting – wickedly characterised by Sir Neville Marriner (see here) as “somebody fighting to get out of a sack” - rather more than I do.

In Tchaikovsky's ballets, however, that approach can mean pressing on relentlessly at the expense of local colour and atmosphere. As Tony Flynn’s perceptive MusicWeb International article, to which a link was provided in the previous paragraph, points out, Dorati himself later admitted to a tendency to inflexible “rigidity” at this point in his career. The nutcracker's no.12a Coffee (Arabian dance), for instance, seriously lacks much in the way of sinuous exoticism. Similarly, the romantic ardour expected in no.14a Pas de deux: intrada is sacrificed to an inflexible tempo that brings it in at just 4:08, far less atmospherically than such other conductors on my shelves as Alexandr Kopilov (4:55), Valery Gergiev (5:01), Mark Ermler (5:05), Evgeni Svetlanov (5:22) and Ernest Ansermet (a gloriously romantic 5:33).

There are also several instances where Doráti, as already suggested, adopts virtually undanceable tempi. Listen, for instance, to the closing bars of The sleeping beauty's Act 3 no.30 finale (track 14, 4:33-5:29). While Tchaikovsky specifies presto, this account makes it, in practice, more of a completely impractical prestissimo for dancers who would, at that point, have been nearing the end of more than two and a half hours on stage. It's worth noting, in passing, that the version that Doráti recorded 25 years later with the Concertgebouw is somewhat less frenetic and actually allows the mind's eye to picture assorted courtiers and fairy tale characters cavorting around the stage while not necessarily collapsing with heart attacks.

A little earlier in the 1955 performance, however, Doráti's very considerable strengths in this repertoire are revealed just as clearly. A good sampling point for any recording of The sleeping beauty is the second scene of Act 2, comprising no.19 Entr'acte symphonique (le sommeil) et scčne and no.20 finale. In those nine or ten minutes, a conductor needs to encompass the widest range of moods and atmospheres, from the creepy spookiness of the demon-infested forest and the revelation of King Florestan's slumbering court to the ecstasy of the prince's redeeming kiss and the lifting of wicked fairy Carabosse's curse. Here Doráti's control of orchestral colour and dynamics - and the performance of his Minneapolis musicians - is superb. The forest has rarely been more weirdly oppressive and the emancipating lovers' kiss - capped by a splendidly cathartic bash on the Minneapolis tam-tam - is clearly positioned as the ballet's emotional core. That sequence was, for me, one of the highlights of these discs.

The sound quality of these two releases is generally pretty good, given the age of the source material. While a major renovation of the Northrop Auditorium in recent years has apparently brought a vastly improved quality of sound, its acoustic in the 1950s was evidently quite dry - even a little reminiscent of New York's notorious Studio 8H. That, however, suits Doráti's direct, non-nonsense approach rather well. Some of the quieter passages expose a slight but acceptable degree of background hiss or, in quieter passages, rumble. There is, too, just one odd moment when the sound suddenly becomes very slightly clearer, almost as if a veil had been whipped off the microphone at that point (The sleeping beauty, disc 2, track 5, 6:34). It's the sort of effect that is sometimes associated with side-breaks, but because of its position that's unlikely to be the cause in this particular case.

The presentation of these releases is pretty minimal. There are listings of individual tracks, but no notes about the ballets themselves. I suspect though, that as these discs are unlikely to be anyone’s first purchase of these works, that really won’t matter too much.

In truth, while these are exciting, involving accounts and of undoubted historical interest, they will probably hold most appeal for Doráti completists - as one imagines many members of the Antal Doráti Centenary Society may well be. The enterprising ADCS certainly deserves the thanks of us all for bringing these - and many other hard-to-find - recordings to life once again and it's good to have them remastered, at long last, to CD. They are best seen, perhaps, as valuable augmentation of the widely available later stereo recordings, rather than as accounts that require those who already own the London Symphony Orchestra or Concertgebouw accounts to replace them wholesale.

Rob Maynard