Arnold SCHOENBERG (1874-1951)
Stokowski discuss Gurrelieder [4:43]
Gurrelieder [112:13]
Arkady DUBENSKY (1890-1966)
The Raven [12:03]
Paul HINDEMITH (1895-1963)
Kammermusik No. 2, Op. 36 No. 1 [18:17]
Paul Althouse - Waldemar
Jeanette Vreeland - Tove
Rose Bampton - Die Waldtaube
Abrasha Robofsky - Bauer
Robert Betts - Klaus-Narr
Benjamin de Loache - Sprecher
Princeton Glee Club, Fortnightly Club, Mendelssohn Club
Eunice Norton (piano: Hindemith)
The Philadelphia Orchestra/Leopold Stokowski
Rec. live 11 April 1932 ((Schoenberg); 9/10 December 1932 (Dubensky); 17 December 1932 (Hindemith)
Transfer by Mark Obert-Thorn
PRISTINE CLASSICS PASC612 [67:25 + 79:51]
The main work on this set, Schoenberg’s Gurrelieder, is a quite remarkable survival. While a number of European companies had begun recording substantial orchestral works and works by modern composers a good ten years before this recording (perhaps the most incredible being Oscar Fried’s acoustic recording of Mahler’s Second in 1923 - now that’s what you call chutzpah!), the American Victor company showed little real interest in either of these categories. Probably the closest thing that Victor had managed before Gurrelieder was the Mengelberg set of Strauss’s Heldenleben in 1928, which was hardly musically cutting edge even in 1928. How and by whom the suits in charge of Victor were persuaded to devote 28 sides at the height of the Depression to a composer whose name, even today, can empty a concert hall is something I have never seen explained but would love to know.
Although he had begun it in 1900, Schoenberg did not complete Gurrelieder for a decade, and its first performance was not until 1913 under Franz Schreker. Its sheer size made it inevitable that it should be infrequently performed. The three performances that Stokowski conducted in Philadelphia on 8, 9 and 11 April 1932, plus another in New York on the 20th, were the work’s US premieres. The Philadelphia Orchestra was augmented to 122 players, and the total number of performers was 532 - a challenge to record successfully even today. Not only did Victor record it, but they recorded all three Philadelphia performances, though the large number of performance problems on the first night led to its being aborted before the end. Eighteen sides survive from that first performance, and apparently at one point the performance had got into such a state that Stokowski can be heard shouting out a rehearsal number in an attempt to get everyone back together! The second performance was recorded for and issued on Victor’s newly introduced Long Playing discs. Although one can admire Victor’s courage in introducing this format, it is hardly surprising that a system which required a complete new playing system to be purchased was a failure. Victor did not help themselves by issuing only a small number of LP records, all but one other of which (Stokowski’s Beethoven 5) were dubbings of 78 rpm issues edited together. It is odd that Edison had made exactly the same mistake with his similarly short-lived Long Play Diamond Discs a few years earlier. This performance was issued on CD by Pearl in 1993 in an excellent transfer by Ward Marston (CDS9066 - review), but the LP format meant that the recording had be much less wide-ranging in dynamic than the 78 format. The present issue is taken from the third performance, which was the one used for the standard 78 format, and issued on 14 records. In his notes, Mark Obert-Thorn estimates that the $28 the set cost in 1932 is the equivalent of $531 in 2020, so it is hardly surprising that sales were minimal and the set is extremely rare. The 78s were transferred by RCA Victor onto 3 LPs in 1954 (LCT-6012) and I believe that this transfer was also used for the 2 LPs issued in 1977 (AVM2-2017), though I do not have access to the 1954 issue, so cannot be certain.
Fortunately, by the second and third performances, the performance problems had been corrected bar a few minor things which could happen in a performance even today. The work could almost have been written with Stokowski in mind; its huge palette of colour, dynamic variety and positively decadent romanticism coupled with its hints at the Sprechgesang and disintegration of tonality which would be seen in the composer’s subsequent pieces were meat and drink to him. It is therefore really very depressing to discover how utterly ungrateful Schoenberg was for these performances. The problem started when, through the Philadelphia Orchestra’s manager Arthur Judson, Stokowski gained from the publishers the right to give the first US performance of the piece. This was entirely above board, there was absolutely no skulduggery involved, but Schoenberg thought that they should have been awarded to him, despite the fact that he had minimal experience or talent as a conductor, and that Stokowski had already given in 1929 the US premiere of Variations for Orchestra, Op.31 (a much harder sell than the romanticism of most of Gurrelieder). One shudders to think what utter chaos could have ensued if the composer had attempted to control the 500 performers that Stokowski employed. However, Schoenberg never forgave him, and continued to make dismissive and abusive comments about the conductor right up to his death almost 20 years later. Stokowski, however, continued his advocacy of Schoenberg’s music and made no response. I’m not sure that I could have been so gracious.
Listening today to the orchestral side of this performance, it is quite difficult to see what there is that could be improved upon in any substantive way - and it is salutary to remember that this was one single live performance; there was no patching session or editing-in of rehearsal fragments. Stokowski has the momentum and sweep that the piece needs; he never dawdles or wallows, but his sensuality is rarely equalled in modern performances (partly simply a result of the more appropriate orchestral style of the performances - for example, the use of string portamento adds immeasurably to the effect in the prelude). But it is not only the more languorous parts in which Stokowski excels, the Zwischenspiel after the Wood Dove’s lament and the Wild Hunt of the Summer Wind are also superlatively done.
The singers are not outstanding in anywhere near the same way. The Waldemar, Paul Althouse (1889-1954), had a very respectable career at the Met between 1912 and 1940, including Grigory in the first US performances of Boris Godunov in 1913 with Didur under Toscanini, the title role in Weber’s Oberon in 1918 and Menotti’s Amelia goes to the Ball in 1938. After a visit to Bayreuth in 1925 he decided he wanted to move into heldentenor roles, and between 1934 and 1940 he sang at the Met the principal tenor roles in all the Ring operas plus Tannhäuser, Lohengrin, Meistersinger and Tristan. Clearly we are not dealing with some minor provincial-oratorio-circuit singer. The voice is a fine one with none of the spreading or strain that afflicts so many present day singers of these roles. Where he falls down is in the imaginative aspect. The style of the music is very similar to that of Wagner, but in 1932 he had not yet gained experience in the Wagner roles, and often seems to be just singing from one note to the next; there is little feel of someone who knows where the music is going. Had the recording been made six or eight years later it may have been much better, but as it stands it is a blueprint for a performance rather than a fully realised one, but at least it is not the severe trial of some more recently recorded Waldemars.
Jeanette Vreeland (1896-1939) was a concert singer who, as far as I know, never sang in opera. She was highly regarded on the concert and oratorio stage, but her early death has meant that she is almost entirely forgotten today. That she was engaged for Koussevitsky’s 1937 live recording of Bach’s St Matthew Passion and his 1938 Beethoven Missa Solemnis indicates the esteem in which she was held. Gurrelieder, however, is a long way from the sort of repertoire that was her staple, and there is a politeness and decorum about her singing which would have been fine in the music she usually sang, but isn’t really what one wants in this piece. Incidentally, in April 1932 Lauritz Melchior and Gertrud Kappel were singing in the Ring at the Met. Now wouldn’t that have been something if they’d nipped over to Philadelphia for a few days...
The other singer who remains at least a name to listeners with an interest in historic singers is Rose Bampton (1907-2007), who sings the Wood Dove. She also sang for many years at the Met, though only rarely in big parts, beginning as Mercedes in Carmen in 1929 until the arrival of Rudolf Bing, with whom she did not get on, in 1950. In the late 1930’s she made a possibly unwise decision to move from mezzo to soprano, and is probably most famous for being the Leonore in the Toscanini Fidelio in 1944. She continued to sing the occasional recital right up to the early 1970s. The scene-stealing Wood Dove gets the most famous number in the score, and Bampton sings this lament with musicality and a lovely tone, though in all honesty she cannot stand comparison with Janet Baker, Anne Sofie von Otter, Jennifer Larmore or Brigitte Fassbaender. The three smaller parts are all perfectly well done, with Benjamin de Loache, despite being an actor rather than a singer, singing rather than speaking much of his part. De Loache (1905-1994) must have made an impression on Stokowski as they appeared together over 30 times, firstly in the concert performance of Berg’s Wozzeck in Philadelphia in 1928 which was its US premiere.
De Loache appears again in one of the other two works in this set. These are really fillers which are interesting to hear, but unlikely to be played very often. De Loache is the reciter in Arkady Dubensky’s setting of Edgar Allan Poe’s The Raven. Dubensky was a Russian-born violinist and composer who moved to the US in 1921 (presumably to escape the Revolution) and almost immediately became a member of the New York Symphony Orchestra, continuing after its merger with the Phiharmonic to become the NYPSO and then the NYPO, and remaining with it until 1952. He died in 1966. Pieces consisting of spoken texts with musical accompaniments are technically known as “Melodramas” and have never really been successful. Richard Strauss made an attempt with a setting of Tennyson’s Enoch Arden for reciter and piano, which was first recorded by Claude Rains with Glenn Gould in 1962 and a surprising number of times since by all sorts of famous names, but such pieces rarely have a life outside the recording studio because they don’t really fit into any standard concert format. Dubensky’s music is very much in the style of a contemporary horror film score, (though it must be noted that neither of the two great early horror films, Dracula and Frankenstein from 1931, had any musical underscoring at all, so Dubensky was not just riding on their coat-tails). In its way it is really quite effective. The performance doesn’t always give it its best chance, though. De Loache’s performance is from that time of transition in acting between the fully rhetorical, almost sung approach that we can hear in, for example, the 1906 recordings of Herbert Beerbohm Tree, and the beginnings of the modern, naturalistic approach after World War 2. De Loache still uses sustained tone and vibrato at moments of particular emotional import, and actually sings the final “Nevermore”. He would have been even more effective if Stokowki’s tempi had not so often been just that bit too fast to allow de Loache to convey the narrator’s depth of melancholy. There is a sense of rush to many of the earlier verses that does the piece no favours. If ever there was a piece where the conductor should be completely subservient to the soloist, this is it - but that was hardly Stokowski’s way. The original issue of this is another of the great Stokowski rarities. I have seen the two 10” picture discs fetch well over £200.
The final piece, Hindemith’s Kammermusik No. 2, is one of those dry-as-dust neo-classical pieces that were fashionable for a short time in the 1920s and 30s (Stravinsky is about the only composer whose attempts have survived in any real sense). The pianist is Eunice Norton (1908-2005 - extremes of life-span seem to be a bit a theme with these performers). She studied with Schnabel in the 1930s, was highly regarded in the US and regularly performed new music. This is the sort of piece where all you need do is play it cleanly and with some verve, which both Norton and Stokowski do. You may respond more positively to it than I, but the monotonously unremitting tread, narrow dynamic range and emotional blankness of much of the music mean that I doubt that I will ever play it again.
The present transfer is another triumph for Mark Obert-Thorn and Pristine. I have never heard the Andante reissue of Gurrelieder (review), but the Pristine is in an altogether different league from the 1977 RCA LPs (which is hardly surprisingly if they do stem from the 1954 transfer). The sound here has a depth, immediacy and presence which are simply astonishing for the circumstances, and tribute must be paid to the Victor recording engineers for capturing so much of an occasion which could have ended up being a complete write-off. Obert-Thorn had the advantage of being able to use vinyl pressings of the great majority of the sides and excellent shellac copies for the others, so surface noise is almost non-existent. The detail that one can hear is sometimes simply amazing. The only parts which disappoint are those which include the full chorus, but this is only to be expected. To record effectively a huge chorus such as this in full flow would have been impossible in 1932, and the great danger was that a sudden fortissimo could have sent the cutting stylus ploughing into adjacent grooves, rendering that side completely unusable - which would have meant that the whole recording was unusable, as no retakes were possible. It is hardly surprising, then, that the engineers put safety first and kept the microphones a good distance from the chorus. Fortunately, the amount of chorus in Gurrelieder is quite small, probably not much more than 10 to 15 minutes in total, so this is not at all fatal. This is a wonderful issue, and one which anyone who loves Gurrelieder should have as an adjunct to a modern recording.
Paul Steinson
