FORGOTTEN ARTISTS - An occasional series by Christopher Howell
18. JONEL PERLEA (1900-1970)

Forgotten Artists index page

When I had been living in Italy just about long enough to make sense of the local record magazines, I was brought up short by a review that began something like this: “The great Mozart conductors of the twentieth century were Walter, Beecham, Karajan, Markevich, Perlea and Zecchi”. Regrettably, I don’t seem to have kept the magazine – we’re talking of 1976 or thereabouts – so I hope I’ve remembered the list correctly. The mix seems clear: two conductors from the mid-century who, in a pre-HIP age, might have topped anybody’s list, the most adulated living conductor, a major figure usually associated with other repertoire and two dark horses – Perlea and an elusive Italian pianist-conductor.

Having thrown down the gauntlet, the writer then got down to his main business of discussing the latest Karajan offerings. The list reverberated in my mind over the years but only now have I finally tracked down enough material to examine Perlea’s claims.

Perlea is nevertheless a conductor with at least a subliminal presence in the discographic world. If you are a fan of post-war opera stars, you will have collected a good many recordings directed by conductors who apparently seem to have been eternally destined for a supporting role. Perlea may, at first sight, seem one of these. RCA engaged him only for operatic work and, specifically, for three sets featuring the tenor Jussi Björling – Manon Lescaut (1954), Aida (1955) and Rigoletto (1956). If you look up reviews of these sets, including some on our own site, you will find that Perlea’s work is often singled out in a way that of Cellini, Votto, Erede or Molinari-Pradelli tends not to be. Nevertheless, RCA did not engage Perlea again – for reasons that will be examined later – until 1965, when he conducted a classic set of Lucrezia Borgia with Caballé in the title role.

There is another way in which Perlea subliminally entered many a semi-musical household. In the late 1960s, Fratelli Fabbri initiated a series of lavishly illustrated coffee-table magazines called “The Great Musicians”. The articles and the illustrations were actually quite good. Popped into the back, almost as an after-thought, was a 10-inch LP. Most of these were licensed from old Vox recordings, grittily pressed. It was generally agreed that the record itself was the Achilles’ heel of the enterprise, and quite a few were conducted by Perlea. It would not have occurred to many at the time that, one day, we might be listening seriously to these old Vox recordings – those by Perlea no less than those by Horenstein – for the evidence they give of a distinctive interpretative point of view.

As ever with this series, I am infinitely grateful to the blogs and YouTube channels, as well as a fellow enthusiast, that have made it possible for me to hear a fairly representative selection from a recorded legacy that proved far larger than I had imagined. If, for the most part, I do not specifically name the bloggers and YouTube posters, this is because these sites have a way of coming and going. Just in the few years since I began accumulating this material, several blogs and YouTube channels have disappeared, while others remain visible but with the links dead. Please assume, therefore, that I am discussing the performances and recordings, not specific transfers of them.

As for the information on Perlea’s life and career, I list in footnotes the websites I have found useful, again with the proviso that they have a way of appearing and disappearing. Information is sometimes contradictory. I have tried to make sense of it. If this article induces others to provide information, corrections and memories, these will be gratefully appended.

Early years and career 1900-1945
Jonel Perlea was born on 13 December 1900 in Ograda, a village on the Romanian plain (1) (birthplace - see right). His father was Romanian, his mother was German. Perlea senior was apparently a good amateur musician. He died when Jonel was only ten and his widow returned to her family in Munich. Thus, a Romanian-German dualism characterized the future conductor, in his education as well as in his family background. His heart surely remained in Romania; at the age of twelve he wrote an “Ograda Waltz” and his first public appearance – as performer and composer – was in Bucharest on 17 October 1919. Nevertheless, his musical studies were in Munich and Leipzig, where Paul Graener was his principal teacher. Some sources suggest he also studied with Max Reger. Reger certainly taught at the Leipzig Conservatoire, but he died in 1916, so any contact would have been minimal.

In 1922 Perlea began to work as a répétiteur in Leipzig, moving to Rostock in 1925. Also in 1922, his fellow Romanian Georges Enescu advised him never to give up composing, and in 1926 Perlea won the Georges Enescu composition prize with a string quartet. The quartet was published and achieved a number of performances.

Perlea then returned to Romania, and, had it not been for the war, would surely have maintained his base in that country. After a period with the Romanian Opera of Cluj-Napoca in the late 1920s – different accounts give different dates – he appeared with the Bucharest National Opera, of which he was General Director from 1932 to 1936. During this period he gave several local premières, including “Die Meistersinger” and “Der Rosenkavalier”. He also conducted some performances of “Boris Godunov” with Chaliapin in the title role. Meanwhile, he was making a reputation outside the opera house, and was conductor of the Bucharest Radio Orchestra from 1936 to 1944. Teaching also played a major role throughout his career, and in 1941 he was appointed professor of conducting at the Bucharest Royal Conservatoire.

What looked like an upward success story came to a halt just as the war was ending. In 1944, Perlea was travelling with his wife to Paris for a conducting engagement. They were stopped in Vienna. King Michael of Romania had switched his support to the allies, depriving Nazi Germany of a valuable source of petroleum. The Nazi authorities demanded that Perlea should broadcast to the Romanian people, urging them to keep faith with the Nazis. Perlea refused and was placed under house arrest. Accounts vary as to the ultimate consequence of this refusal. Some say he remained under house arrest till the end of the war, others claim he was sent to the concentration camp of Mariapfarr. Strangely, I find no trace of the existence of a concentration camp at Mariapfarr, an Austrian village near Salzburg, except in biographies of Perlea and one or two other Romanians. Maybe some reader can shed light on this?

Perlea was not to return to Romania till 1969.

Perlea the composer
Our earliest aural glimpse of Perlea is, somewhat surprisingly, as a composer. On May 5^th 1939, George Enescu conducted the New York Philharmonic in what appears to have been a slimmed down version of Perlea’s “Variations on an Original Theme”. Perlea took up composition again in his last years and must have conducted some of his works, enabling Gunther Schuller to discover that he was “a fine composer, more than just a conductor-composer” (4). Nevertheless, he was mainly active in this role during the pre-war years, so this seems a good moment to discuss such of his works that I have heard.

It is not clear how much of the “Variations” Enescu actually conducted. The fellow enthusiast who supplied me with this recording – which has occasionally been available in Enescu anthologies – says it has variations 3, 4 and 7. To my ears, there are four sections, with the first sounding as if it ought to be the theme itself. The recording is pretty obviously cut off, suggesting that, if Enescu omitted variations 1, 2, 5 and 6, he did play at least one more.

It is clear from the torso that Perlea had a wide and colourful orchestral palette, with considerable use of the vibraphone. His language is tonal, but pushing tonality to the limit – “Gurrelieder” filtered through Szymanowski and Enescu himself might give an idea. Much of the music is langorous, but folksy elements emerge in variation 7.

The recording is dim, but anything conducted by Enescu is of interest. He consistently clarifies the teeming textures, leading the listener’s ear towards the part where the principal melodic interest lies and ensuring that an overall line emerges.

These interpretative qualities are less evident in a 1998 recording of the “Don Quixote” Symphonic Scherzo in which Paul Popescu conducts the Romanian Radio Symphony Orchestra. The impression is of a capable conductor and orchestra wading their way gamely through a work they are learning as they go. This is what we usually get with complicated out-of-the-repertoire works and, as ever, gratitude for hearing the music at all is tempered with doubts as to whether it was really worthwhile. It would be interesting to know what aspect of “Don Quixote” Perlea is illustrating. The language and palette – minus the vibraphone – are much as before, with an injection of what seems to be Debussy recollected in a nostalgic vein. It’s so far from a Scherzo that I can’t help wondering if it is really meant to be like that. The 3-minute section starting around 11 minutes sounds as if it ought to be mercurial, but it isn’t here. I couldn’t help wishing someone like Munch or Markevich could have come along to fire things up a bit. The concluding 14 minutes are luxuriantly slow, with a long solo for bass clarinet against shimmering strings. It seems a bit indulgent, but maybe it wouldn’t if things had sizzled more earlier on. The last couple of minutes are undeniably beautiful.

“Don Quixote” was included on a disc dedicated to Perlea, composer and conductor, by Romanian Radio. The same CD also had four Lieder sung in 1984 by the soprano Georgeta Stoleriu, accompanied on the piano by Marta Joja. The disc gives the titles in Romanian and in English but they are sung – and were presumably composed – in German. It would be a real treat to know what they are about, beyond the bare titles.

The meat of the offering is the 3 Lieder op.10. In the first of these, “Vigil”, the lack of any explanation is particularly felt. It seems indulgently doleful, but perhaps there is a point to it. The second, “Longing”, is such a strikingly beautiful song that it makes an impact despite the odds stacked against it. Perlea proves himself here a master of soaring soprano lines to rival Richard Strauss – which is saying something. The last song, “Departure”, is much briefer, and makes an energetic conclusion. It is nevertheless one-upped here by the separate song “On your name day”, again brief but a “good sing”. Stoleriu manages the songs very well, and Joja is a fine accompanist as far as can be heard – she seems to have been recorded in the next room. The case for Perlea as a composer of Lieder in a rich post-romantic idiom seems a strong one – I hope he wrote more than just these.

Perlea in Italy - La Scala
We left Perlea in the hands of the Nazis. He was not able to resume his work in Bucharest. His anti-Nazi credentials would seem unassailable, but the Communist government installed in Romania soon after the war evidently felt he would be an awkward customer to have around. He was refused a visa to return home.

In the first place he established himself in Italy and soon found work coming his way. Toscanini heard him conduct the Santa Cecilia Academy Orchestra and immediately offered him an engagement at La Scala. There followed a brief but intensive period of collaboration with Italy’s foremost opera house. At the time of writing, La Scala’s website provides full details of all performances – opera and concerts – given from 1951 onwards. This just postdates most of Perlea’s work. From another source, I have found details – I hope complete – of his opera performances (2). I suspect he may have conducted more concerts than I am able to trace. Perlea’s opera performances at La Scala were as follows – the dates are those of the first night.

1947 (29 January):
Saint-Saëns: Sanson et Dalila
Ebe Stignani (Dalila), Fiorenzo Tasso (Sansone), Abelardo Martelli (messaggero), Erminio Benatti (1° filisteo), Ugo Savarese (Sommo sacerdote), Carlo Forti (Abimelecco), Cesare Siepi (vecchio ebreo), Giuseppe Menni (2° filisteo), ballerinas: Luciana Novaro, Gino Pessina, producer: Mario Frigerio, choreographer: Rosa Piovella Ansaldo: scenes: Camillo Parravicini

1947 (17 Aprile)
Mozart: Così fan tutte
Suzanne Danco (Fiordiligi), Tatiana Menotti (Despina), Giulietta Simionato (Dorabella), Marcello Cortis (Guglielmo), Pietre Munteanu (Ferrando), Fritz Ollendorff (Don Alfonso)

1947 (11 June)
Gluck: Orfeo ed Euridice (Berlioz version)
Suzanne Danco (Euridice), Loretta Di Lelio (according to the archives, but replaced by Lia Origoni) (Amore), Ebe Stignani (Orfeo), scenes: Gio Ponti

1947 (10 September)
Mussorgsky: Boris Godunov (Rimsky-Korsakov version)
Bianca Baessato (Xenia), Giuseppina Sani (nutrice), Rina Corsi (Marina), Dora Minarchi (Teodoro), Ebe Ticozzi (ostessa), Aleksandr Vesselovskij [Alessandro Wesselovsky] (Vassili Sciuiskij), Tomaso Spataro (Grigori-Dimitri), Giuseppe Nessi (Missail), Giacinto Sgaravato (Kruscev), Giuseppe Zazzetta (boiardo), Cesare Masini-Sperti (innocente), Attilio Barbesi (Andrea Celkalov), Aristide Baracchi (Lavitzki), Aldo Bacci (Cernikovskij/Nikitich), Tancredi Pasero (Boris Godunov), Boris Christoff (Pimen/Rangoni), Nicola Rossi-Lemeni (Varlaam), Eraldo Coda (guardia frontiera/voce interna/Mitjuscia)

1947 (7 October)
Giordano: Siberia
Adriana Guerrini (Stephana), Disma De Cecco (fanciulla), Ebe Ticozzi (Nikona), Antonio Annaloro (Vassilij), Gino Del Signore (principe Alexis), Cesare Masini-Sperti (Ivan), Abelardo Martelli (sergente/Ipranivick), Erminio Benatti (cosacco), Giovanni Inghilleri (Gléby), Eraldo Coda (Walinoff/capitano), Aldo Bacci (governatore/Walitzin), Michele Cazzato (Miskinski/starosta), Aristide Baracchi (invalido), Attilio Barbesi (commissario/ispettore)

1948 (7 maggio) Double bill:
Renzo Bianchi: Gli Incatenati
Clara Petrella (figlia), Wanda Madonna (madre pazza), Cesare Masini-Sperti (giovane), Vasco Campagnano (capo), Piero Campolonghi (padre), Eraldo Coda (sbirro), Enrico Campi (vecchio), producer: Alessandro Brissoni, scenes: Antonio Molinari
Strauss R: Salomé
Lili Djanel (Salome), Ebe Ticozzi (paggio), Maria Benedetti (Erodiade), Silvana Zanolli (schiava), Fiorenzo Tasso (Erode), Gino Del Signore (Narraboth), Giuseppe Nessi (1° giudeo), Cesare Masini-Sperti (2° giudeo), Erminio Benatti (3° giudeo), Mario Carlin (4° giudeo), Gino Penno (1° nazareno), Piero Guelfi (Jochanaan), Aldo Bacci (cappadoce), Eraldo Coda (5° giudeo), Dario Caselli (1° soldato), Carlo Forti (2° soldato), Enrico Campi (2° nazareno), producer: Hans Zimmermann, scenes: Isaia Colombo

1948 (7 October)
Massenet: Werther
Dora Gatta (Sofia), Giulietta Simionato (Carlotta), Silvana Zanolli (Kathchen), Giacinto Prandelli (Werther), Cesare Masini-Sperti (Schmidt), Piero Campolonghi (Alberto), Enrico Campi (Johann), Eraldo Coda (podestà), Aldo Bacci (Brulmann)

1949 (31 January)
Beethoven: Fidelio
Delia Rigal (Leonora-Fidelio), Hilde Güden (Marcellina), Mirto Picchi (Florestano), Angelo Mercuriali (Giacchino), Giuseppe Taddei (Don Pizarro), Boris Christoff (Rocco), Dario Caselli (Don Fernando), producer: Oscar Fritz Schuh, scenes and costumes: Felice Casorati.

1952 (2 April)
Mozart: Il Ratto dal Serraglio
Maria Callas (Konstanze), Tatiana Menotti (Blonde), Giacinto Prandelli (Belmonte), Petre Monteanu (Pedrillo), Salvatore Baccaloni (Osmin), Nerio Bernardi (Selim). Scenes by Nicola Benois.

According to Grigoriu, Perlea also conducted La Traviata, Turandot and Lohengrin at La Scala, but this seems to be an error. The annals clearly list performances of these operas during those years, but with other conductors. Possibly Grigoriu is confusing performances that Perlea conducted elsewhere in Italy. He conducted Turandot with Maria Callas at the Naples San Carlo Theatre in a run of four performance opening 12 February 1949, for instance.

The concert appearances by Perlea listed at La Scala’s site – limited to those from 1951 onwards – are listed below. Here too, the date may be the first of a run of two or three performances:

1951 (17 August)
Cherubini: Anacreonte Overture
Beethoven: Symphony no.3 “Eroica”
Wagner: Parsifal Prelude
Mussorgsky-Ravel: Pictures at an Exhibition

1951 (25 October)
Debussy: Nocturnes
Malipiero, Riccardo: Cantata Sacra (first performance)
Haydn: Symphony no.94 “Surprise”
Ravel: Valses Nobles et Sentimentales
Glinka: Russlan and Ludmila Overture
Irma Lucia Bozzi (soprano), Coro della Scala (chorus master Vittore Veneziani)

1957 (22 June)
Cherubini: Anacreon Overture
Debussy: La Mer
Liadov: 8 Russian Folk Songs
Tedeschi, Alberto Bruni: Messa per la Missione di Nyondo (first performance in Milan)
Mariella Angioletti (soprano), Aldo Bertocci (tenor), Alfredo Giacomotti (bass) Ottavio Fanfani (reciter), Corro della Scala (chorus master: Norberto Mola).

All the opera performances were sung in Italian – this was actually the last time “Fidelio” was heard at La Scala in the vernacular. Only two years later the Wunderkind Karajan conducted a production in German.

A story is attached to the “Orfeo ed Euridice” production. According to the annals, the part of Amore was sung by Loretta Di Lelio, but it seems that she pulled out just before the dress rehearsal. Her place was taken by a young singer called Lia Origoni who, in her innocence, failed to observe the standard practice that only Orfeo takes a curtain call at the end of Act I. The Orfeo, Ebe Stignani, rewarded her apparent impertinence with a hearty kick backstage. Perlea was witness to this and roundly castigated Stignani, declaring that “we should all be grateful to Lia Origoni for saving the show”. Fortunately, subsequent performances went harmoniously and this opera was among the events chosen to celebrate a visit to Italy by Eva Peron (3).

The pattern of Perlea’s collaboration with La Scala is clear. For two years, from January 1947 to January 1949, he conducted a wide repertoire and, in 1947 particularly, had a more consistent presence than any other conductor. Thereafter, he conducted a single opera, in 1952, and that only as a substitute for an ailing Issay Dobrowen. He was still conducting concerts in 1951 – I am assuming, maybe wrongly, that he also conducted concerts in the years 1947-1949 – and then returned just once, in 1957.

Most resumés of La Scala’s postwar history seem reluctant to mention Perlea at all. He does not appear to have held any specific post, such as Music Director, Artistic Director or Principal Conductor. In view of his recommendation by Toscanini, this rather looks like a variant on what happened in Berlin, where Perlea’s compatriot Sergiu Celibidache kept the seat warm for Furtwängler, but failed to reap the hoped-for reward.

Obviously, Toscanini’s situation was very different, politically, from Furtwängler’s. Given his enormous prestige, his freedom from all taint of collaborationism and the disgraceful treatment he had received from the Fascists, probably no one in Italy would have denied that he should return to his old post at La Scala if he wished. Hence the willingness to give full rein to a Toscanini protégé while there was still some hope that the great man himself would take over. It soon became clear, however, that, given his considerable age, Toscanini would be unable to do more than conduct the occasional concert. Thus the stage was set for new manoeuvres.

There has always existed a strong feeling in Italy that La Scala, of all Italian opera houses, should have an Italian as its Music Director. Even in recent years, the Barenboim period was largely seen as an unwelcome interlude. And after all, La Scala still had – theoretically – an Italian Music Director. Victor De Sabata had taken over following Toscanini’s forced exit in 1929. A very different type of conductor from Toscanini, those who felt the latter too inflexible believed De Sabata was the greatest living Italian conductor. Even for those who appreciated Toscanini’s rigour, the obvious candidate for La Scala, if it couldn’t have Toscanini, in 1947 as in 1929, was De Sabata. The only problem might have been a political one.

It was Toscanini’s unwavering opposition to the Fascist dictatorship that had led to his departure after the “schiaffo di Bologna” in 1929. Any conductor who took his place under such circumstances was implicitly associating himself with the regime. De Sabata, in any case, made no secret of his friendship with Mussolini, and even conducted private performances in the dictator’s own home. In spite of his Jewish origins, he managed, too, to be a welcome guest in Germany until at least 1939.

Germans who had collaborated with the Nazis had to be de-Nazified after the war. I have never heard tell of any similar de-Fascistification process in Italy and I presume that any such programme must have been a pretty mild affair. Many from the artistic world conveniently “discovered”, after 1945, that they had really been Communists all along. De Sabata never stooped to this, but there was no real obstacle, once it was clear that Toscanini would not return to the post, to easing De Sabata back into a position he had theoretically never relinquished. Most reference books, in fact, list him as Music Director of La Scala from 1929 to 1953 uninterruptedly, and Artistic Director till 1957. A few list brief tenures in the 1940s, notably by Mario Rossi and Franco Capuana. None mention Perlea.

Clearly Perlea, like Celibidache in Berlin, had become an uncomfortable presence. The best thing was to quietly let him go and pretend he’d never been. The question remains, whether he had helped his departure by lending hostage to fortune in some manner.

This might have come about in two ways. Firstly, he may have got uppity about the operas assigned to him. As we shall shortly see, he certainly did this at the Metropolitan a year or so later. So maybe he did this at La Scala too.

The other question is – how did his performances stand up compared with those that Toscanini or De Sabata might have given? Some of the operas he conducted were, in fact, taken up by De Sabata over the following few years. Even one lacklustre showing would have provided ample ammunition for those wishing to be rid of him.

A full-scale biographer of Perlea would doubtless find time to sift through contemporary press comment and to establish whether any of Perlea’s performances failed to reach the desired standard – or, at any rate, whether critical opinion thought they did. A rather curious discography by Grigoriu – rich in omissions and errors – lists live recordings that have circulated on obscure labels, “not all listed in the international discography”. According to this, recordings exist of Perlea’s La Scala performances of Così fan Tutte, Boris Godunov, Werther and Die Entführung aus dem Serail. I’m inclined to feel that I’ll believe in these performances when I hear them. It really does seem impossible that a recording of Maria Callas in Die Entführung would not have circulated widely if it existed, whatever the sound quality. Still, if any reader can throw light on these recordings, better still tell me where I can get them, I would be delighted to know. They should be able to demonstrate, for better or for worse, the quality of Perlea’s work at La Scala.

Apart from this, we have just two tantalizing souvenirs. These were set down by Decca in 1947. The most revealing is that of the Ballet Music from Saint-Saëns’ Samson et Dalilah. It was set down on 19 July with Victor Olof as producer. Samson et Dalilah had been Perlea’s first opera at La Scala. He conducts with real flair and a sense of sinuous oriental colour – more so, perhaps, than in his nevertheless very good 1956 Vox recording of it with the Würtemburg State Symphony Orchestra. He also obtains some brilliant articulation which Toscanini would surely have approved of – La Scala’s orchestra is shown to have been a fine band at that time. What differentiates this version completely from the Vox one, however, is the vocal, swooning strings in the central section. This is a type of playing, generally thought of as “pre-war”, which lingered on in Italy into the early 1950s. It is quite probably what Saint-Saëns would have expected, but we can hardly blame Perlea if the Würtemburg orchestra provided just good, clean playing at the same point.

Decca’s other La Scala offering was a souvenir of Suzanne Danco’s Fiordiligi in Così fan tutte – Come scoglio. Così was still a “problem opera” back then so, rather than couple it with Fiordiligi’s other aria – if it was sung at La Scala, for Perlea was quite a cutter, as we shall see – they had her sing Voi che sapete from Le nozze di Figaro. Both were set down on 30 July 1947. Danco was much associated with Fiordiligi – her Italian debut had been in this role (Genoa 1941). Her first appearance with the Glyndebourne Festival Opera was also as Fiordiligi, at the Edinburgh Festival under Gui in 1948, so this recording provided a neat visiting card.

I have never much warmed to Danco’s art. Anna Russell once referred, cruelly but pertinently, to the “Nymphs and Shepherds, or pure white” style of singing, meaning, I suppose, the kind of British soprano typified by Isobel Baillie. The Franco-Belgian “pure white” style was a little different but I find it similarly dated. Nevertheless, I cannot deny the security and cleanness of her singing here – though she lacks a trill. The orchestra is neat and crisp. Danco’s Cherubino is well known from the famous Erich Kleiber recording. The main difference here is Perlea’s big ritardando before the restatement of the principal theme.

Decca have made precious few recordings at La Scala over the years. Maybe they thought they would be onto a good thing by grabbing La Scala’s up-and-coming new conductor. History went differently.

Perlea in Italy – after La Scala
Though Perlea faded from view at La Scala, he was in considerable demand elsewhere in Italy for most of the 1950s. He gave the local premières of Tchaikovsky’s Mazeppa and The Maid of Orleans and Richard Strauss’s Capriccio (Genoa 1953), as well as championing Nino Rota’s I due timidi and, notably, Henze’s Boulevard Solitude (Naples 1954). I due timidi was originally a radio opera, premiered under Franco Ferrara in 1950. Rota’s reworking for the stage was heard under Perlea at Palermo in 1955.

At least four Italian opera performances under Perlea have been issued in some form or can be found on YouTube. The ever-hopeful Grigoriu lists several more but I will limit discussion to the four I have heard. The technical quality, I must say, does not greatly encourage the search for others.

Mozart’s Le Nozze di Figaro was given at the Teatro San Carlo di Napoli on 20 February 1954 with a cast consisting of Italo Tajo (Figaro), Alda Noni (Susanna), Giulietta Simionato (Cherubino), Renata Tebaldi (La Contessa d'Almaviva), Scipio Colombo (Il Conte di Almaviva), Giuliana Raimondi (Barbarina), Endré von Koreh (Bartolo), Agnese Dubbini (Marcellina), Piero De Palma (Don Basilio), Cristiano Dalamangas (Antonio) and Gianni Avolanti (Don Curzio).

This performance has been issued by Hardy Classics (HCA 6017-2). Up to a point, they play fair. Most of the Act II finale, and the whole of the Act IV finale, are declared missing. They admit to an “unevenness of pitch”. The sheer ghastliness of what this entails can be savoured in the “letter” duet between the Countess and Susanna. The piece starts with good sound at a nice, relaxed tempo, and Alda Noni sings her opening phrase very attractively. When Tebaldi enters the whole acoustic changes, seemingly coming from an adjacent bathroom, the pitch rises between a quarter and half a tone and, obviously, the tempo spurts ahead with it. Then Susanna sings another phrase and the picture is normalized. And so it goes on. We are therefore getting a good impression of Noni’s performance, a falsified one of Tebaldi’s. Ives-like, I rather wondered what it would sound like if, when they sang together, each continued at their different pitch and speed. Obviously, it didn’t happen. I’m racking my brains to think how the tape could have systematically gone faster for one singer than for the other but I can’t think of any explanation. Nor can I imagine how, in the course of a piece lasting about three minutes, the parts regarding one singer are so much more damaged than those regarding the other. My conclusion is that the recording had to be pieced together from at least two different tapes, one much better than the other – for when it actually is good, it’s remarkably good for its age – and, above all, recorded on tape recorders that went at different speeds. But can’t modern technology straighten these things out? It would be a terrible job for someone, so the first question would be, is it worth it?

Well, we have here the only surviving testimony, so far as I know, to either of the two Mozart roles sung by Renata Tebaldi – the other was Donna Elvira. Her two arias emerge fairly unscathed – just a few blips. Her steady production and sheer vocal beauty grace “Porgi, amor” and the first part of “Dove sono” and it’s much purer – with less scooping – than one might have supposed. The faster section of “Dove sono” show an ability to toss off light, agile music that she might have profitably developed further. In her recitatives, those that are at pitch, and her ensemble work, with the same proviso, she enters fully into the character. So yes, this is one reason for attempting further restoration of the recording.

Another reason, and the other Mozartian dark horse, is Giulietta Simionato as Cherubino. Unfortunately both her arias are above pitch. “Non so più” is made to go like the wind and “Voi che sapete” is similarly made to sound about the fastest version I’ve ever heard. Even at the right pitch, they must surely be on the urgent side, but here I must suspend judgement.

Alda Noni’s recitatives, in particular, well above pitch, are made to sound squeaky and breathless. But “Venite, inginocchiatevi” is at the right pitch and a relaxed tempo and she sounds lovely. So she does in “Deh, vieni”, taken fairly slowly. The Barbarina and Marcellina sound less happy.

Of the principal men, Italo Tajo, mostly known on disc in smaller roles, is fine when we hear him at the right pitch, and so is Scipio Colombo as the Count. Piero De Palma offers a very caricatured Don Basilio. Unusually, he gets his aria – the only major cut is Marcellina’s aria – but to little avail. Bits are at the right pitch and sound good, but they alternate with bits up about a tone. At the end the orchestra races still further and the final chord is up a third.

At this point, it is difficult to reach any conclusion about Perlea himself. When the pitch is right, the performance has a rococo grace and is even a tad too relaxed, though with light textures and well-sprung rhythms. Other bits of the score are jacked up in pitch and tempo to sound absolutely manic – compare the Act II trio (up in pitch) with the Act III sextet, which unfolds almost too patiently. I just wish the surviving material could be handed over to someone like Mark Obert-Thorn who might be able to sort the pitches out.
 
The following year, 1955, Perlea was back at San Carlo, Naples, with Massenet’s Manon. The cast, singing in Italian, was Clara Petrella (Manon), Ferruccio Tagliavini (Le Chevalier Des Grieux), Saturno Meletti (Lescaut) and Vito De Taranto (Le Compte des Grieux).

Given that Perlea could be a vicious cutter, the listener’s first reaction as the Prelude leads straight into the second scene – and the first act also lacks its final scene – will be “he’s at it again!” It’s not quite that simple. Go to Vittorio Gui’s 1952 Italian-sung performance (RAI Milano with Carteri, Prandelli, Poli and Clabassi) and you’ll find that the cuts are the same. So they are in an Italian-sung 1966 RAI film under Maag (from Modena with Freni, Garaventa and Bruson). So, presumably, are they in a 1969 La Scala performance in Italian under Maag (with Freni, Pavarotti and Panerai), which I haven’t heard – but the cuts are described in an MWI review. Another common feature is that the spoken dialogue is replaced with recitative.

All this is discussed very fully in a doctoral dissertation “The Impact of Jules Massenet’s Operas in Milan 1893-1903” by Matthew Martin Franke (Chapel Hill 2014), available on Internet. What we have is not an arbitrary piece of scissors-work by these three conductors – though Perlea makes a few more snips of his own – but an “Italian Manon”, an alternative version of the opera prepared for Italian use. The extent of Massenet’s own involvement in the omissions and reordering will probably never be known, since the archives of the publisher Sonzogno, which possibly conserved such information, were destroyed in the Second World War. It has been definitely established that Massenet himself wrote the recitatives to replace the spoken dialogue. He set them in French, but the general assumption is that he did so in the expectation that they would then be adapted for the Italian edition. They were never taken up in French performances.

It must be said that the “Italian Manon” has something to be said for it. Firstly the recitatives, which should please those who, like Tchaikovsky – and myself, if it matters – are not entirely happy with spoken dialogue in what is essentially a late-romantic opera with a tragic ending. Could anyone really not prefer singing voices soaring over the off-stage choral “Magnificat” in the Saint-Sulpice scene rather than spoken interjections? There is, of course, no reason why these recitatives could not be reinserted into uncut performances in French, but this would be a hybrid version and modern musicology does not approve of such things.

As for the cuts, the principal one, apart for the omissions in Act 1 already mentioned, is the total exclusion of Act III Scene 1, with Acts IV and V treated as one to make an opera in four acts. There are also fairly substantial snips in what now becomes Act IV Scene 1. All this makes for an opera lasting around two hours – slightly less under Gui and Perlea, slightly more under Maag. The famous Monteux set and the well-regarded modern version under Pappano last 40-45 minutes more. Dare I suggest that two hours of music, plus intervals, might be enough for an opera that, for all its beauties, hardly has the density of Tristan und Isolde?

Perplexities remain. The Italian vocal score, published by Heugel in France and Sonzogno in Italy (and currently downloadable from IMSLP) contains more than is actually heard under Gui, Perlea and Maag. In particular, Act III Scene 1 is present. Nevertheless, Franke has been unable to trace any Italian performance of the opera which actually included that scene. As far as can be told, the opera that was first heard in Italy at the Teatro Carcano of Milan in 1893 was the opera that Italians continued to hear until original-language performance became the norm.

Given that the “Italian Manon” is a freestanding version, prepared in some sort of collaboration with Massenet himself, and given that the 1969 La Scala production probably marked its swansong, it would be nice to have a decent recording of it for reference purposes.

The recording balance is the main reason why the 1955 Naples production won’t fill the bill. It seems to have been recorded by someone sitting close to the orchestra, and well to one side of the theatre. At best, the singers can just about be heard. For entire long sections, one strains one’s ears to hear Tagliavini, who frequently seems to be singing off-stage entirely.

This is a pity, since there is every indication that Perlea was just as much in sympathy with Massenet’s Manon as his RCA recording proved him to be with Puccini’s Manon Lescaut. He conducts with vitality and passion, but keeping within French bounds – the otherwise excellent Gui occasionally strays into Mascagni territory. Perlea’s pacing of the Saint-Sulpice scene is particularly inspired – flexible and atmospheric but also well-structured.

Clara Petrella was known as a “singing-actress”. This label often carries the implication – as in the case of Magda Olivero – that the voice was not inherently anything special. I’m not sure that this recording offers any basis for assessing her vocal beauty. She has a considerable, but well-controlled vibrato, her production is firm, her intonation is true. She is sometimes rather free rhythmically, after the verismo manner, and occasionally catches Perlea out. She is fully in the part and is moving in the last scene.

Tagliavini has the idiosyncratic way with rhythm common to many great Italian tenors of the past, but does not carry it to excess. He can produce a full, ringing sound but his particular speciality is a honeyed head-voice. “En fermant les yeux” is a miracle of control and is greeted by such a farmyard of applause – a herd of cows and at least two elephants appear to have been in the theatre that evening – that it has to be repeated. Tagliavini fans might seek out a performance he gave in Rome Opera Theatre in 1957 under Napoleone Annovazzi and with Victoria de los Angeles as Manon (yes, in Italian). Reports suggest you might hear the singer better.

Under the circumstances, the Gui performance, in fair radio sound for 1952, is probably a better bet if you want to hear the “Italian Manon”. Rosanna Carteri and Giacinto Prandelli are less individual but are fully in their parts. Your only chance of seeing an “Italian Manon” is the 1966 Modena film.

As stated above, the Italian première of Tchaikovsky’s Mazeppa was given by Perlea. This was at the Maggio Musicale Fiorentino on 6 June 1954. The singers were Ettore Bastianini (Mazeppa), Magda Olivero (Mariya), Boris Christoff (Kochubey), Mariana Radev (Liubov), David Poleri (Andrei), Jorge (Giorgio) Algorta (Orlik), Fausto Flamini (Iskra) and Piero De Palma (Drunken Cossack). The performance was issued on LP on Rococo RR 1016, Estro Armonico EA 046 and Cetra “Opera Live” LO 43. It can currently be found on YouTube.

The fact that this was the first performance in Italy of Mazeppa lends a certain interest to the recording, which I take to be somebody’s taping of a broadcast. For what it is, it doesn’t sound too bad, though the voices become faint when the producer has them sing at the back of the stage – or at least, that is how I interpret what I hear. Nonetheless, it is possible to gain a reasonable impression of the performance. It is also good enough to suggest that Mazeppa, however little performed, is a fine and effective opera.

Nowadays, obviously, we expect to hear the original language. Nevertheless, there are a number of reasons why opera buffs might still want to hear this performance.

The first of them is the Mariya of Magda Olivero (1910-2014). Renowned as an artist who carved out her own niche in an epoch dominated by Callas and Tebaldi, she was famously under-recorded. In a stage career stretching from 1932 to 1981, with occasional concert appearances even into her nineties, she took part in only two official complete opera sets: “Turandot” (as Liù) in 1938 and “Fedora” in 1969. Fortunately, Italian Radio (RAI) had her sing a number of major roles. These and other broadcasts have become collector’s pieces over the years.

This recording is certainly collectible, but it also explains why recording companies found her problematic. She is often referred to as a “singing actress”, as compared with other operatic stars who basically stand there singing, with the occasional waddle around the stage at the producer’s insistence. The definition carries with it the implication that, maybe, there was always something about her that you’d miss if you weren’t there in the theatre to hear her. This is probable, but it is likely to remain an article of faith. What is evident here is that her very detailed, deeply felt conception of the music led to a certain breaking up of phrases where we might prefer to hear the long line. If you compare her with the two “singing-singers” in this performance, Bastianini and Poleri, their words are clear but are always part of the long vocal line. Olivero’s line tends to come in bulges, a fact that is probably compounded by the way in which this recording loses the softer tones. Nevertheless, her voice was a beautiful one at this stage in her career – by the time of her Fedora it had become a little jaded. There is a golden gleam on it and the expressive bulges are under tight control. Contrast them with the Liubov, Mariana Radev, whose bulges are just squally. Altogether, this is a fine end moving portrayal, never more so than in the haunting lullaby with which the work closes.

Another reason for collectors to hear this is the Kochubey of Boris Christoff. The great Bulgarian bass was an Italian by adoption and much loved in Italy. For Italian opera houses and RAI he studied a long series of major Russian roles in Italian translation. Of course, one would like to hear him sing the role in Russian, but if you want to hear him sing the part – and what admirer of Russian opera would not? – there is only this, so far as I can make out. His unmistakable tones carry great authority, but his phrasing is also detailed and compassionate. Great singing.

A further reason, maybe, is the Andrei of David Poleri (1921-1967). In spite of his name, Poleri was American and is mainly remembered for his contribution to Charles Munch’s Boston recordings of Beethoven’s 9^th Symphony and Berlioz’s “La Damnation de Faust”. His career was brought short by a helicopter accident. He has a strong, even and ringing voice. This is not really a tenor’s opera, but he makes a good effect with his one important solo, at the beginning of Act 3.

Bastianini is a better documented singer, of course. He was never a particularly expressive artist, but he pours plenty of well-focussed tone into a role that by its nature can hardly engage our sympathy.

Which brings us to Perlea himself. He brings a fluid, conversational touch to the earlier scenes, and produces admirable vitality in the battle prelude to Act 3. He is fully participant in all the more dramatic moments. The orchestra plays well, but in a manner that sounds old-fashioned even compared with the RAI orchestras in the early 1950s, with a lot of string portamenti and a singing attack even in forte passages. The brass vibrato yields nothing to that of Russian orchestras of the same period. It is perfectly possible, of course, that this conserves a playing style that Tchaikovsky himself would have recognized. The chorus is sometimes a bit ragged, but the production evidently requires them to do quite a lot of dancing as well as singing – they seem to be wearing wooden clogs at one point.

A less prestigious venue, internationally speaking, the Festival Senese at Perugia, saw the Italian première of Tchaikovsky’s The Maid of Orleans on 30 September 1956. This was sung by Marcella Pobbe (Joan), David Poleri (King Charles), Ugo Benelli (Cardinal), Gianpiero Malaspina (Dunois), Enzo Mascherini (LJonel), Fernando Corena (Thibaut), Isidoro Antonioli (Raymond) and Belen Amparan (Agnes).

Theoretically, the recording is not much worse than that of Mazeppa, but in practice, the dividing line between something one can still listen to with a degree of pleasure, and something one will put up with only for the sake of hearing something very rare, seems to fall midway between these two recordings. Maybe, too, the dividing line between singing so individual one will make any reasonable sonic sacrifice, and singing that would be enjoyable enough if such allowances didn’t have to be made, falls midway between Olivero and Christoff, on the one hand, and Pobbe, Corena et al on the other. Indeed, I can trace no issue of this performance, whether on LP or CD. It can be found on YouTube.

On the whole, the Perugia public got a good introduction to the opera. Marcella Pobbe has a vibrant voice and is thoroughly in the part. Occasional intonation problems are countered by many more moments of gleaming security and she has a believably Slavonic sound, even though she is singing in Italian. Poleri can be enjoyed again, as can Corena and, indeed, most of the cast, though Pobbe is not alone in having the occasional intonation problem. The name of Ugo Benelli is to be approached with caution. The Cardinal is a bass role. The celebrated tenore di grazia Ugo Benelli made his debut in Montevideo two years after the present performance and I find no suggestion that he had previously trained as a bass. Assuming the name is correct, therefore, I presume this is another singer of the same name, though I can find no other mention of an Ugo Benelli who sang bass.

The orchestra is decidedly inferior to that of the Florence Maggio Musicale but Perlea conducts with a fiery passion and a dose of what one would call slancio if this were Verdi. Rather more problematic, though, is the question of what he is actually conducting. I must confess, at this point, that, to hear Mazeppa, I found an Italian libretto on Internet and followed with this – though the translation sung was a different one – rather than a score. I got the impression that a few cuts were made but I did not investigate more fully. In the case of The Maid of Orleans, I found no libretto in a language I know and so downloaded the full score (in Russian and German) from IMSLP. Alarm bells started ringing when Tchaikovsky’s quite developed and dramatic prelude had a cut of 150 bars in the middle, reducing it to a mere 41 bars. The long flute cadenza at the end also went missing. The singing and dancing at the beginning of Act 2 is confined to just one song. But all through, following the score became a sort of obstacle race as snips and slashes, with occasional bits of rewriting to cover the patches, came thick and fast. Tchaikovsky’s operas are not my musicological patch but I do know that Tchaikovsky himself was asked, for the first performance, to recast the leading role for a mezzo-soprano and to make cuts. In 1882 he made a new version, reinstating the cuts. Both the full score and the vocal scores downloadable from IMSLP are of the 1882 revision, always supposing that the cut version of the first performance was published at all. If Perlea is making the cuts sanctioned by Tchaikovsky for the first performance, he nevertheless grafts them onto a version with a soprano heroine so, to a greater or lesser extent, what we hear is a “Perlea version” of the score. It lasts about half an hour less than the recording under Rozhdestvensky. Quite frankly, I am a little surprised the difference is not greater still – assuming Rozhdestvensky himself plays the score complete.

To modern ways of thinking this is pretty shocking. On the other hand, just to play the devil’s advocate for a moment, the usual complaint is that the opera is dramatically flawed and ineffective in the theatre. So far as I can tell by listening only, Perlea’s version is taut and dramatically well-shaped. So why insist on philological purity and then complain that the opera is no good?

Idomeneo in Greece
Belonging to this period in Europe is a production of Mozart’s Idomeneo given on 17 September 1955 at the Herod Atticus Theatre of Athens. The cast were Constantino Ego (Idomeneo), David Lloyd (Idamante), Kostas Paskalis (Arbace), Eleanor Steber (Ilia), Maria Kerestedji (Elettra) and Petros Hoidas (La Voce). Perlea conducted the Athens State Symphony Orchestra. This was issued in 2006 by Walhall (WLCD 0166).

Back in 1955, Idomeneo was sufficiently little-known for the larger part of the Athens audience to have been blissfully unaware that they were hearing a very odd version of it. We will not say too much about the tenor Idamante, since this continued to be normal till some time later – though it must be said that the first studio recording of all, the 1950 Haydn Society issue under Zallinger, used a soprano. From a romantic standpoint one can even see the logic of a baritone Idomeneo, though the logic is not Mozart’s. One might even forgive the odd snip – after all, the Glyndebourne version under Pritchard, the first “complete” version to have wide circulation, sounds more like extended extracts today. If only it were the odd snip … If only, indeed, they were big snips, but following the order of the music in the score. This, at least, is what happens in the first act and the last part of the third. In the second act and most of the third, one can only conclude that Perlea had an unbound score with the pages unnumbered, which got scattered by the famous Greek winds, leaving him to re-assemble it as best he could. Numbers are not just shuffled around within the acts, they drift back and forth between acts 2 and 3.

Always supposing one does not reject such a practice out of hand – and some might – I daresay the very worst person to assess the validity of what Perlea has done is a musician-listener with a score in front of him. Time after time, I had to stop the music and leaf backwards and forwards in the score to find where on earth they’d got to now. In truth, only the people who followed the performance in the theatre more than 60 years ago are in any position to say whether a theatrically viable experience was on offer. Failing that, I would need to do a scissors and paste job on my score, or at least on my libretto, to make it match what I am hearing. I’m not convinced that the performance would repay the very considerable effort needed to do this.

The principal justification for listening to the production today is Eleanor Steber. This is grand, secure and passionate singing. Occasionally it seems more on a Verdian than a Mozartian scale, but this is big Mozart and it’s no use selling it short. A magnificent assumption, then, and there is no other record of Steber in this role so far as I can see.

For the rest, Lloyd and Hoidas are unattractive, the others middling-decent.

Up to a point, in spite of his peculiar editorial decisions, Perlea is the other reason for hearing this performance. The Athens orchestra is a motley crew, but Perlea conveys a good deal of energy, grandeur and expressive warmth. He proves that he would have been capable of giving a notable interpretation of the score as Mozart wrote it, or would he have insisted on his strange “Perlea edition” under whatever circumstances? As it is, this is a curiosity for Steber fans and for Perlea fans prepared for rapture and frustration in equal measure.

America: The Metropolitan
It seemed logical to continue the story of Perlea’s Italian career, but in fact it developed in parallel with another that eventually came to predominate. Just as La Scala closed its doors, Perlea was summoned by the one American opera house of comparable international fame – the Metropolitan. During the 1949-1950 season, he conducted Tristan und Isolde, Rigoletto, La Traviata and Carmen. I leave Gunther Schuller to tell the tale:

One of that season’s happiest encounters for me — and I think for most of the orchestra — was the arrival of Jonel Perlea, one of the best conductors to grace the Met’s podium during my years there …
In his very first rehearsal [of “Tristan”] we could tell that we were in the hands of a superior musician. … He managed to bring to that ecstasy- and hysteria-laden score a wonderful calming restraint. With Fritz Stiedry the more frantic episodes in Tristan, especially in the third act, could easily spin out of control. It is incredibly intense music, sometimes more intense than it can readily tolerate. Perlea treated the music with an almost chamber music transparency — lyric, eloquent, even elegant — without diluting the drama and emotional excitement of Tristan, or for that matter of Carmen or any of the operas Perlea was given. … We really loved this man (4).

Some of the elderly Schuller’s memories may have become a little confused, however. He suggests that, even in 1949-50, Perlea was forced to conduct with his left hand only as the result of a stroke. This didn’t happen until 1959. He is also confusing on the reasons for Perlea leaving the Met after a single season.

All year long we kept hearing backstage rumors that certain conductors, especially Alberto Erede, also new at the Met in 1949, were agitating with the management to have Perlea retired. If true, it was but another typical example of what is known far and wide in the music world as “opera intrigue.”
 
Actually, Erede came to the Met for the 1950-1951 season, and was called because Perlea had virtually marched out. The issue was the operas the Rudolf Bing was willing to assign to him. It would appear that the likes of Traviata or Rigoletto weren’t good enough for Perlea, who wanted the heavier stuff like Wagner and late Verdi. But, although Perlea had somehow got to conduct that acclaimed Tristan, the heavy stuff at the Met in those years belonged to Fritz Stiedry, and there was no getting round that. Bing was only too happy to give Perlea the “lesser” Verdi, the Puccini and Mascagni, and so on, but Perlea refused it. This is where Bing remembered Erede, whose work he had admired at Glyndebourne, and called him in to fill the gap Perlea had created. Erede may or may not have been the Macchiavellian schemer Schuller believed he was, but he was not the villain in this particular piece. Rather, Perlea had chosen to cut off his nose to spite his face. During his one season he had conducted Traviata at the Met with Albanese, Di Stefano and Warren, Rigoletto with Berger, Di Stefano and Warren, and Carmen with Risë Stevens, so why not take Bing’s crumbs gratefully and reflect that Stiedry was not blessed with eternal life? And what conductor today would not sell his birthright to conduct Traviata or Rigoletto at the Met with casts half as good as those?

Perlea’s Met debut, in Wagner’s Tristan und Isolde, took place on 1 December 1949. It was taken to Philadelphia on 6 December and repeated at the Met on 17 December and 2 January. The 17 December performance was broadcast and has recently surfaced on Archipel ARPCD 0183-3. The full cast was Lauritz Melchior (Tristan), Helen Traubel (Isolde), Blanche Thebom (Brangäne), Mihály Székely (Marke), Herbert Janssen (Kurwenal), Emery Darcy (Melot), Peter Klein (Hirt), Philip Kinsman (Steuerman) and Leslie Chabay (Stimme eines jungen Seemanns). The variable sounds suggests that two sources, each with its attendant problems, have been used.

Having duly listened to these records, I find myself in the slightly embarrassing position that I really cannot think of better words to describe Perlea’s conducting than those of Schüller. In certain respects it seems a very modern concept, anticipating Boulez’s Wagnerian manner in its coolly analytical approach. But, whereas Boulez could sometimes throw the baby out with the bathwater, with Perlea, it is a case of “less is more”. For sheer riveting tension this performance can match practically anything, except maybe Furtwängler. Regrettably, the notorious “Bodanzky cuts”, removing large slices of Acts 2 and 3, were still operating at the Met. Given Perlea’s track record as a cutter, it would be risky to suppose he disagreed with them.

Melchior had first sung Tristan at the Met in 1929 and this run of performances was his last. Twenty years of singing roles like that is heavy going. Nonetheless, his voice rings out strongly and steadily in the upper register and above mezzo forte, especially in Act 2 – Act 3 shows some strain here and there. Even ten years earlier, critics were noting that his softer notes in the lower register could sound foggy. No doubt the recording has exaggerated this defect, but as presented here he sounds downright croaky, at the beginning of the Love Duet for example. Herbert Janssen is a magnificent Kurwenal, so this is one of those Tristans where the Kurwenal is better than the Tristan.

Blanche Thebom is a splendid Brangäne, but this is not one of those Tristans where the Brangäne upstages her mistress, for Helen Traubel is a gleamingly secure, fully involved Isolde. Székely is an unimposing Marke and the smaller parts are not particularly well taken.

Perlea’s Met performances of Carmen and Rigoletto – but not Traviata – were also broadcast, so may still survive somewhere.

America: San Francisco, San Antonio, Connecticut, Manhattan
The Met was followed by a season (1950) at San Francisco Opera and then a season (1951) at the San Antonio Grand Opera Festival in Texas. Perlea seems to have found it difficult to carve out a permanent niche for himself on the American operatic season.

With purely orchestral work he did better. He still enjoyed Toscanini’s support and this gained him several engagements with the NBC Symphony Orchestra. Some reports say that Perlea’s appointment to the Connecticut Symphony Orchestra was at Toscanini’s suggestion, but the site of the Greater Bridgeport Symphony, as it was later renamed, tells a different story.

The Connecticut Symphony's 10th anniversary concert at the Klein in 1955, also featured the debut of the orchestra's Music Director, Jonel Perlea, a widely known Romanian, who made his American debut at the Metropolitan Opera in 1949. Mr. Perlea came to Bridgeport because of an unusual sequence of events beginning in 1952. That year a conductor backed out of a contract over a disagreement about program selections. Another Maestro was substituted at the last minute.  Elizabeth Hughes [Program Chairman] complained to her friend, impresario Arthur Judson, that she didn't want to be left without a conductor after an agreement with his office and the Symphony had been signed, and off-handedly asked if Perlea was available for the following summer.  She had recently met him at a party in New York. The following July, Perlea gave up two weeks of his European schedule to be at the Fairfield campus for a single appearance.  Two years later he accepted the Symphony's offer to be its Music Director (5).

Perlea remained with the Connecticut Symphony till 1965. So far as I can make out, the Connecticut Symphony didn’t make records and I have seen no objective analysis of what Perlea did or did not achieve in those years – repertoire, orchestral standards and so on. What is clear from the Greater Bridgeport Symphony site is that his tenure did nothing to ease the orchestra’s financial problems. Officially, his contract was not renewed in 1965 because the Connecticut Symphony as such ceased to operate. It was revamped under its new name, the Greater Bridgeport Symphony. Initially, as an expense-saving exercise, it worked only with guest conductors, till matinée idol José Iturbi was appointed in 1967, tweaking the orchestra’s image in a more popular direction to good economic effect.

Unofficially, though, there was some reason to suppose that Perlea was no longer the man to revitalize the orchestra’s image. In 1957 he suffered a heart attack while conducting, followed by a stroke in early 1959 – different accounts give different dates, I will come to this later. The right side of his body remained paralyzed, but he reinvented himself as a left-hand only conductor and resumed work in late 1960. He remained active for almost another decade, but it would not be entirely surprising if some Connecticut board members felt that their Music Director was living on borrowed time.

Perlea’s other long-standing American appointment was with the Manhattan School of Music, where he trained the student orchestra, apparently to a very high standard, and taught conducting. Various testimonies to his work in Manhattan can be found. A cellist, Evangeline Benedetti recalls:

Manhattan School had a good orchestra then, with a conductor who’s not well known — a masterful conductor, Jonel Perlea — who worked the orchestra unbelievably hard. So I’d had that discipline, and then auditioned for Stokowski and was accepted (6).

As usual, different accounts give different dates. Manhattan School’s own site has a photo of Perlea taken in 1952 at a reception given shortly after his appointment. This would seem to settle the date conclusively (7). He withdrew in 1959 following his stroke but was then reinstated and remained in harness till his death.

The 1950s – recordings in Europe
Perlea would have remained an elusive figure indeed had he not been booked extensively as a recording artist during the 1950s. All three companies concerned – Remington, Vox and RCA – were American but all the recordings were made, probably for economic reasons, in Europe.

Remington
The history of the Remington label is documented in detail at the Remington site(8). Many of their recordings were made with the Berlin RIAS Symphony orchestra – their recordings with Anatole Fistoulari have already been mentioned in this series of articles. Perlea’s recordings, made from 1953 to 1955, were:

Debussy: La boîte à joujoux, Berlin RIAS SO (R-199-159)
Saint-Saëns: Le Carnival des Animaux/Tchaikovsky: Swan Lake (excerpts), Berlin RIAS SO (R-199-160)
Brahms: Piano Concerto no.2, Edward Kilenyi (piano), Berlin RIAS SO (R-199-164)
Liszt: Piano Concerto no.1/Totentanz, Edward Kilenyi (piano), Berlin RIAS SO (R-199-166)
Ulysses Kay: Concerto for Orchestra, Teatro La Fenice Orchestra, Venice (R-199-173)
Dane Rudhyar: Sinfonietta/Peggy Glanville-Hicks: Gymnopedies, Berlin RIAS SO (R-199-188)
Donizetti: Lucia di Lammeroor (R-199-200/3) (details below)

I’ve been able to hear most of these.

Those who claim that Debussy’s La Boite à Joujoux is Perlea’s greatest single recording have a point. The music itself is normally thought to need special pleading – it doesn’t sound so here. For once, Perlea has a really fine orchestra to work with. As a result, he hones in on the weight, timbre and character of each phrase and episode with Boulez-like precision, revealing the work as one of extraordinary originality and modernity.

Perlea is in particularly fiery form in the first movement of Brahms’s Second piano Concerto – the orchestra really bursts in after the preliminaries. The sound is lean but with ample Brahmsian phrasing. The American pianist Edward Kilenyi gives a strong, clear account, with less pedal than we often hear. It’s a forward-moving, naturally musical performance.

Hopes that this will be an outstanding dark horse on the lines of Mrazek/Swarowsky or Jenner/Dixon – both discussed in this “Forgotten Artists” series – are not wholly realized. The second movement is slowish, though slower tempi have been brought off successfully. There is a shortage of bite from both parties, Kilenyi is sometimes mannered in his phrasing and the music takes wing only in its later stages.

The third movement is kept on the move. The impression is that Perlea is trying to bring off as musically as possible a tempo he would not have chosen himself. It sounds uneasy and Kilenyi tends to move forward even so, suggesting he would rather be playing Rachmaninov. A telling moment occurs after the central climax. Perlea brings in the orchestra at a noticeably slower tempo, creating a mood of hushed mystery. Kilenyi responds to this and the whole passage is beautifully done. Then, with the re-entry of the solo cello, the faster original tempo is resumed, as I suppose it had to be, and the mood is lost.

The last movement starts blandly. Not a matter of tempo but of getting bite and snap into the dotted rhythms – this comes here and there from Perlea. Certain changes of tempo suggest they had different ideas about the music and not enough rehearsal time to sort them out.

Liszt’s First Piano Concerto (I haven’t heard Totentanz) is much more satisfactory. The first section of the Concerto had me thinking that Kilenyi might be too wayward and, at the same time, too peremptory to engage the listener’s feelings. In the slow second section a real poet emerges. Most remarkable is the scherzo section. The tempo is sufficiently slow for every single note to be heard, yet played with such precision, lightness and evenness as to sound truly dazzling. From there to the end the performance blazes and thrills. Perlea is a highly positive partner, even slightly upstaging Kilenyi in the first part before the pianist has fully warmed to his task.

Ulysses Kay (1917-1995) was an Afro-American and the nephew of a jazz musician. After receiving encouragement from fellow-Afro-American William Grant Still, he studied with Howard Hanson and Bernard Rogers at the Eastman School and then spent a year (1941-2) working with Hindemith at Yale. He studied in Rome from 1949 to 1953, which perhaps explains why an Italian orchestra was playing his Concerto for Orchestra, which he brought to Italy with him, having written it in 1948.

I must say I looked up Kay’s biography only after listening, but “neo-classical” and “late Hindemith” were the labels that immediately came to mind. It is surprisingly unshowy for a Concerto for Orchestra, ending with a mostly slow Passacaglia. Did Kay feel over-conscious of the need to prove that an Afro-American can write serious music?

It’s a well-wrought work with a spacious “Arioso” second movement and a feeling of nobility both here and in the Passacaglia. Perlea builds it up well and obtains considerable dynamic shading from the orchestra. A certain Italianate quality to the orchestral sound, particularly the trumpet vibrato, had me thinking of Casella as much as of Hindemith – though Casella’s own Concerto for Orchestra is a punchier affair. I will not avoid other works by Kay if they come my way, but I’m not stimulated to seek them out at all costs.

Dane Rudhyar (1895-1985) was born in Paris as Daniel Chennevière. In 1916 he came to New York for a performance of one of his works and remained in the USA, becoming an American citizen in 1926. He was drawn into the worlds of astrology and theosophy, about which he wrote copiously. He also wrote novels and painted. As a composer, he associated with those like Ruggles and Cowell who were more for throwing brickbats at the system than for setting up rigorous alternative systems of their own. An off-beat figure for most of his life, he was rediscovered as a New Age guru in the 1970s.

In line with his theosophical beliefs, Rudhyar believed that the “new composer was no longer a ‘composer’ but a “medium, an evoker, a magician” (9). The danger with this sort of thing is that unreasonable expectations may be aroused. We might remember how Cyril Scott, another exponent of the esoteric, published a piano suite in 1913 called Egypt, supposedly inspired by memories of “my past Egyptian lives”. Stylistically, it is no different from his other music written at the same time.

Rudhyar’s Sinfonietta was reshaped in 1928, according to the site dedicated to his work, from a Sonatina for piano (10). A drastic reworking surely, since only the quiet third section sounds as if it could be effective on the piano as it stands. It was revised in 1979, but Perlea obviously played the first version. Oddly enough, Rudhyar’s reliance on short, gritty motives, deliberate avoidance of logical continuity and an alternation between quiet meandering and colossal climaxes – the conclusion is awe-inspiring – reminded me of Havergal Brian as much as anyone. Another musical outsider, but one whose standpoint could not have been more different. Perlea conducts with power and conviction, the early vinyl sometimes buckles under the strain.

Perea’s earliest “official” opera recording was his only one for Remington – Donizetti’s Lucia di Lammermoor. The cast consisted of Renata Ferrari-Ongaro (Lucia), Giancinto Prandelli (Edgardo), Philip (Filippo) Maero (Enrico), Norman Scott (Raimondo), Luigi Pontiggia (Arturo), Tosca Di Leo (Alisa) and Uberto Scaglione (Normanno). The orchestra and chorus were those of the Teatro La Fenice, Venice.

Operadis gives the date of this performance as 15 March 1951, which does not tally with the statement by the Remington Site that Perlea recorded for Remington from 1953 to 1955. The “Sound Fountain” is definitely wrong in describing it as a live performance. Operadis gives it as a studio recording and it quite clearly is so. Listening on headphones reveals no trace of audience rustle. It would also have been impossible in Italy, in those days, for a tenor to sing his major arias even half as well as Giacinto Prandelli does without roars of applause breaking in before the orchestra has finished. Furthermore, the precision and fine intonation of the horn chording at various points would have been difficult to achieve live with a much finer orchestra than this. One further oddity: on some issues, the conductor was described as Laszlo Halasz, who acted as producer for Remington and certainly conducted some of their recordings. The general view is that Perlea was the conductor.

Back to him shortly. The heart of this opera, the justification for doing it, is Lucia. You won’t find many references to Renata Ferrari-Ongaro. For Remington, she sang Liù in a “Turandot” made at about the same time under Capuana. As just Renata Ongaro, there’s a live “L’Italiana in Algeri” from 1960, also under Capuana, and she sang in Ghedini’s “La Pulce d’Oro” under Abbado at the Teatro Comunale of Florence in 1963. As a teacher, she is listed in a good many singers’ curriculums.

On this showing, she has a very nice voice, pure, steady and virginal, inherently suited to Lucia. It is a very even voice throughout the register, remaining beautiful even on the unscripted top Es. She is technically well prepared, unfazed by the agility of the traditional cadenzas. So far, so very nice. Unfortunately, I don’t think I’ve ever heard a singer so emotionally inert. It really does not seem to have occurred to her that there is anything more to the music than pretty pattern-making. In her first scene, she us upstaged expressively by the Alisa. In the closing duet of Act 1, in the famous chromatic theme Prandelli shows her time and again how, without exaggeration and tricks, just a few subtle inflections and a bit of colouring, this music can come to life. But all to no avail. She goes her placid way. I realize this is pre-Callas, pre-Sutherland, at least in concept. But, well before this, Lina Pagliughi had shown that a light, agile voice, such as Donizetti probably had in mind, can perfectly well express emotion too.

Could Perlea have done something about it? Well, in a low budget production like this, we just don’t know how much time he had to work with the singers. He begins well, with a fine appreciation of the orchestral timbres. There is plenty of energy, but also plenty of atmosphere. He seems to want to relate Donizetti back to Cherubini rather than forward to Verdi. This in itself is no bad thing, and there are passages in Lucia’s faster music where the conductor seems to be trying to push the soprano on – but the lady’s not for pushing. By the end, the prima donna’s apathy impregnates the whole enterprise. After her very sane mad scene has dragged to its end, one welcomes the arrival of Prandelli, anticipating that from here to the end, at least, the performance will be worth hearing. He certainly does raise the temperature, but the suspicion is that Perlea himself has given up by then. He accompanies precisely and correctly but without apparent involvement. I was reminded of certain performances under Erich Leinsdorf where everyone seems on tenterhooks to get everything right, while missing out on the one thing that matters.

In truth, it doesn’t help that Perlea’s orchestra is very backwardly recorded. The voices are clear, but when two or more of them are singing, I struggled to grasp the harmonic structure of the music, even with the score in front of me.

A set for Prandelli admirers, then. The other singers are mainly good, and Norman Scott, as Raimondo, has a finely resonant timbre. Only the Arturo is woodenly inexpressive, suggesting that he might actually have been the right match for this particular Lucia.

Vox
Billboard of 5 June 1954 described Perlea as “now an exclusive Vox artist”. As we have seen, he continued to record for Remington until 1955 and his first RCA opera set was made in 1954, others following in 1955 and 1956, so any exclusivity must have been limited to the repertoire Vox wanted from him. Gray’s catalogue shows that he recorded regularly for Vox from April 1954 till September 1958. The repertoire ranged from Mozart to the early 20^th century. There were notable concerto collaborations as well as purely orchestral works. It would be useful to give a complete, chronological list but I am not sure that my information is sufficient as yet. Most of the records are listed in Gray, but not quite all, and the information does not always match that given elsewhere. I shall therefore discuss the records I have been able to hear in a musically chronological order.

Vox – Orchestral 1 – Mozart
Gray’s catalogue lists versions of Symphony no.25 in G minor K.183 made by Perlea on both 11 April 1954 and 16 May 1956. For the earlier one, the orchestra is not given, for the later it is said to be the Vienna Symphony Orchestra. The disc I have heard is said to be played by the Vienna State Opera orchestra. Confusingly, Gray gives the same catalogue number for both dates – PL 8750. Were two performances really issued under one number?

Whatever, the performance is a fine one. The première recording of this work, under Alfred Wallenstein in 1938, took a very brisk view of the first movement. Perlea finds a tempo that allows the music to breathe, ideally encompassing both drama and poignancy. The second movement has considerable depth of expression, though without attempting to read into the music more than is there. The Minuet is slow but not heavy. The finale goes well but here I rather miss Wallenstein’s sheer drive. Perlea is not especially generous over repeats; none in the first movement, first halves in the second movement and finale.

Symphony no.29 in A K.201 (Vienna State Opera Orchestra, 16 May 1956, PL 8750) gets a majestically flowing interpretation. Rhythmically, it is sufficiently well sprung not to suggest slackness or staleness. On the other hand, a spot more sheer dash in the finale would have done no harm.

To describe Perlea as ungenerous with repeats would be putting it mildly – apart from the Minuet and Trio, he ignores them all. It should perhaps be added that this symphony is very rich in repeats and a complete performance, at these tempi, would have taken over half an hour – beyond the capacity of an LP side. While I won’t say this is not a good performance, for a forgotten interpretation from the 1950s, my vote would go to Keilberth’s Bamberg SO recording on Telefunken.

Symphony no.32 in G K.318 (Vienna Pro Musica Orchestra, c.1957, PL 10140) shows Perlea’s Mozart at its best. It is relaxed yet pointed and spirited, with a heartfelt central section. Another version from the mid-1950s, by Leinsdorf and the RPO on Westminster, is a bit punchier, more obviously driven, though not insensitively. Leinsdorf’s breaking down of the music into taut thematic cells is closer to modern HIP performances. Marginally, I’ll go with Perlea’s longer-breathed manner.

Perlea gives Symphony no.33 in B flat K.319 (Vienna Symphony orchestra, 16 May 1956, PL 8750) unusual stature. Not by pumping it with grandeur but by finding a serenity combined with elegance and vitality that touches the sublime, for example in the Jupiter premonitions in the central part of the first movement and the principal theme of the second.

A comparison with Rudolf Kempe – in a live performance from 9 April 1960 with the Rome RAI Orchestra – finds the German conductor chunkier, with clucking staccatos and an over-fulsome second movement. It’s attractively bucolic, but fails to achieve the suggestions of sublimity that we hear from Perlea. The comparison shows that it’s Perlea’s long-breathed phrasing that gives breadth without heaviness to the music. No repeats in the finale in either version (none are marked for the first and second movements).

The opening of Symphony no.35 in D K.385 – “Haffner” – (Vienna Pro Musica Orchestra, c.1957, PL 10140) may seem a little low-key but we soon find that the tempo allows the music to spin along naturally, with vitality and also gravity where necessary. The lead back to the recapitulation in the first movement is memorable. The ultimate impression is of justness and fullness of expression. That might sound as if I’m talking about Klemperer, but there’s a gentler touch to Perlea’s Mozart and he never slogs.

A broadcast performance under Ansermet (24 October 1955), available from René Gagnaux’s site, points to the dangers of veering too far one way or the other. Ansermet’s outer movements are hectic, his second restless, his Minuet pompous. Ansermet was justly famous for other things, but I expected more insight than this. Neither performance plays the repeats in the second movement (none are marked in the outer movements).

The opening bars of Symphony no.36 in C K.425 – “Linz” – (Vienna Pro Musica SO [Vienna SO according to Gray], 16 May 1956, Vox PL 8750) may not impress you very much – the dotted rhythms are a little lax. From bar 4 the music is setting up an inevitable trajectory and thereafter I can find nothing to fault. Once again, Perlea finds tempi where vitality can give way to gravity without loss of tension. It all unfolds as if it were a single page. The first movement repeat is played, not those of the second and fourth movements.

A comparison with Walter Goehr’s MMS recording with the Winterthur SO is instructive. No one could say Goehr’s opening bars are slack. Indeed, he is concerned throughout to justify his slightly slower tempi with sharp accents, bristling staccatos and maximum dynamic contrast. It’s all very tense and dramatic, and indeed very fine in spite of a suspicion of tub-thumping. What it seems to lack is the sense – which we get from Perlea – that it is illumined from within.

The opening of Symphony no.40 in G minor K.550 (Bamberg SO, 8 June 1955, PL 13100) steals in perfectly. The cells of the famous theme are well separated but the tempo is swift enough for them to sing as a single line. Thereafter everything unfolds perfectly. The outer movements are fairly speedy and have considerable drive, but a drive that seems to come from within the music not something the conductor is pushing. The second movement is broad and deeply intimate in feeling, the Minuet unexpectedly, and refreshingly, swift. Repeat in the first movement, none in the second and fourth.

You realize how good Perlea is when you turn to Keilberth’s version with the same orchestra, the more so when the Keilberth is an excellent performance too. Keilberth takes slightly more than a minute longer, with the same selection of repeats. At the beginning of the first movement you feel he is spelling it out a bit too much, but later the music flows more naturally and sounds remarkably similar to Perlea’s. In the second movement both conductors take a similar tempo. Just occasionally, here too, Keilberth spells things out a little, but generally this is another intimate, natural reading. Keilberth is slower in the other two movements. He makes a good case for his tempo in the Minuet, and he makes a lovely thing of the Trio. In the finale he seems to be holding back, and in places the orchestra seems to want to move on – and the performance sounds better when they do.

Perlea recorded a “Jupiter” the previous day, which I haven’t heard.
 
Perlea also set down a set of ten Mozart Overtures: Bastien et Bastienne, Così fan Tutte; Die Entführung aus dem Serail, Die Zauberflöte, Don Giovanni, Idomeneo, Il Re Pastore, La Clemenza di Tito, Le Nozze di Figaro and Der Schauspieldirektor (Vienna Pro Musica Orchestra, 10 March 1956, PL 8720). This was presumably the most comprehensive selection to date, though Josef Krips soon followed with a set of 8 (Zurich Tonhalle Orchestra. December 1960, MMS 2218) that added La Finta Giardiniera, but omitted Bastien, Idomeneo and Il Re Pastore.

It is here, more than in the symphonies, fine as they mostly are, that I would find justification for the Italian critic, mentioned at the beginning, who numbered Perlea among the greatest of Mozart conductors. Throughout there is a sense of fiery, passionate conviction remaining just, but only just, within the bounds of the music’s surface elegance. Generally, this means slowish tempi, though in Così fan tutte the wind soloists have a job to keep up. He does not condescend the early Bastien et Bastienne, given unusual stature, nor the earlyish Re Pastore and Schauspieldirektor, which get tough, stingingly accented performances. I’ve never heard Die Entührung sound so dramatic and Die Zauberflöte has mystery, unease, even menace. Figaro, at 4’ 19”, would give you a fairly digestible egg, but its jabbing accents might not lead you to suppose a comedy is on the cards. Less striking, perhaps, are Don Giovanni, Idomeneo and Clemenza, but only because they differ less from other conductors’ treatments of them.

Vox – Orchestral 2 – Beethoven
A slight rhythmic unsteadiness at the beginning of Symphony no.1 in C op.21 (Bamberg Symphony Orchestra, 11 April 1954, PL 9120) – the conductor seemingly pushing ahead of the beat – is deceptive. Things soon settle down into an energetic, well-pointed performance. We will suppose the small dynamic range to be of the recording’s making. The balancing of the woodwind well to the fore, sometimes dominating the texture, seems likely to have been a choice of the conductor.

The second movement is slow but with relaxed phrasing filling the phrases. This comes closer to the sublime than usually seems possible in this movement. The third movement has splendid articulation and real fire. There is a pause before the trio that some may disapprove of but the trio itself, fractionally slower, is wonderfully perky, perhaps the best I’ve heard. The finale goes with real ebullience. No repeats in the first two movements. The finale gets its repeat and all are made in the third movement. In spite of a first movement that makes less of a mark than the rest, this is a performance which vividly portrays the real Beethoven breaking out of Mozartian and Haydnesque moulds.

This performance was issued both separately and in tandem with the Seventh. In the case of the separate issue, I hope the jacket notes explained the connection between George Caleb Bingham’s iconic American image “Fur Traders Descending the Missouri” and a Viennese classical symphony. See below for the more suitable, if less memorable, cover for the First and Seventh together.

I have no date for Symphony no.4 in B flat op.60, stated on the original issue to have been played by the “Vienna State Philharmonia”, changed to the Vienna State Opera Orchestra on later issues PL 8740). It was out in time to be reviewed in Billboard of 6 November 1954.

The outer movements are on the slow side but with the sort of lean sound and sizzling attack we usually associate with faster performances. The result is one of the most dramatic versions I’ve heard of a symphony that sometimes chugs along amiably like a preview of the “Pastoral”. The slow movement is broadly sung but the dotted rhythms are made to threaten the calm. The scherzo is swift and driven. In many ways this has as much to offer as any version, certainly it has its own particular insights. However, it has its downsides too. The orchestra, while playing with intense conviction, is not always immaculate, suggesting these low budget operations allowed few retakes. The recording is raw. This at least might be corrected with sympathetic remastering. Lastly, the performance is very short on repeats – none in the outer movements and even the scherzo is not unscathed.

Symphony no.7 in A op.92 (Vienna Pro Musica Orchestra [Vienna Symphony orchestra according to Gray], 11 April 1954, PL 9120), came out in tandem with the First Symphony – quite a long disc for those days.

A spacious but lean, forward-moving introduction is notable for the conductor’s total lack of hurry in the isolated phrases before the Allegro breaks in. The pauses get their full value. The Allegro itself is swift but not very swift. The dotted rhythms are properly articulated all through – this is a notorious trap and some celebrated performances come a cropper somewhere along the line. Trenchant articulation and high energy levels at speeds that allow for clear enunciation also characterize the third and fourth movements. The trio to the former is majestic without dragging. The second movement is broadly paced and allowed its full tragic power. While the abiding impression is one of concentrated energy, there is some lovely wind phrasing in all movements.

On the debit side, the performance is short on repeats, though an editing glitch at the start of the development in the first movement leaves me wondering if that repeat, at least, was actually played but then removed, none too tidily. The recording is top-heavy. Nevertheless the conviction and sheer justness of this performance mark it out, even with its flaws, as one of the sevenths that count.

Symphony no.8 in F op.93 was paired with the Fourth – details as above. It is a trenchantly articulated performance. Tempi are broad, but not really slow, in all four movements. There is a pleasing pastoral quality to the woodwinds’ softer interjections and the trio to the minuet has both warmth and depth. The ultimate effect, though, is of strength and vigour. The first movement repeat is played. A very fine performance.

Vox – Orchestral 3 - Earlier Romantics
Schubert’s Symphony no.9 in C was set down with the Bamberg Symphony Orchestra on 19 September 1956 (PL 10200).

The introduction is extremely broad. Towards the end, Perlea allows no more than a minimal tightening of the tempo – in those days an accelerando into the Allegro was the norm. Perlea’s tempo for the Allegro is sufficiently moderate for the second subject to fall naturally into place without any slackening. The whole movement is propelled along vigorously with some baying trombones. At the end, Perlea at first holds the tempo when the theme from the introduction appears. Then, when it is played by the strings in bare octaves, he allows just a tiny broadening. Not many versions up to that time had played this movement so straight – even the usually faithful Boult had treated it more flexibly. The second movement struts along rather briskly. This means that Perlea can give power to the grinding climax without either collapsing under his own weight or having to move on. I’m afraid I find more logic than magic here, though. The remaining movements go with plenty of fire and attack at quite brisk tempi. However, the Trio to the Scherzo sounds hustled – it invariably does when the conductor chooses a brisk tempo for the Scherzo and insists on holding it through the Trio. The finale, too, lacks poetry in the gentler moments. It’s all a well thought-out but rather one-sided view. The orchestra is sometimes messy, though they play with conviction. As for repeats, it’s quicker to say that a couple of shorter ones in the Scherzo and Trio are played and that’s it. That was the norm for this symphony in those days. A few conductors played a fuller Scherzo and Trio, but a full clutch of repeats had to wait another two decades – and the technology to put so much music on a single disc.

Perlea also recorded the Rosamunde Overture and some overtures by Mendelssohn and Weber.
 
Berlioz’s Symphonie Fantastique, again with the Bamberg Symphony Orchestra, was recorded on 8 June 1958 (PL 11090).

With his smallish-sounding orchestra, Perlea emphasizes on the one hand the classical, Gluckian roots of this music. At the same time he gives full measure to every unusual harmonic turn and brings out many disturbing counter-melodies I do not remember having ever heard so clearly – and this in a recording that is fairly limited, at least in the pressing I’ve heard.

This series of “Forgotten Artists” is revealing many unusual facets of this symphony. We’ve had psycho-drama from Vladimir Delman and lithe classicism from Vladimir Golschmann. With Perlea one is immediately struck by a trough of despondency that sets the tone throughout. Using slowish tempi in every movement, Perlea gives us Byronic melancholy, the romantic wanderer. Whereas the idée fixe often seems a parenthesis in other performances, here it seems an inescapable destiny. The parentheses – long ones – are the other passages in which the protagonist tries to get away from it, or where he dreams he has found a way of escape.

I have seen this performance trashed as slow and dull (11). For me it is truly hypnotic, one of the great ones on record. I only hope that the original stereo masters can yield up a sound that does fuller justice to the colour and power that one senses is present, and that I shall one day hear the result.

Rossini is not really a romantic, but Perlea’s disc of Overtures belongs here at least chronologically. Included are Guglielmo Tell, Il Barbiere di Siviglia, La Cenerentola; La Gazza Ladra, La Scala di Seta and Semiramide. The orchestra is the Bamberg Symphony Orchestra (STPL 511.180). It was issued in 1959 so was presumably among the last prior to Perlea’s stroke.

It’s as good a Rossini overture disc as I’ve heard. The two slow sections of Guglielmo Tell are beautifully phrased and intensely felt. In the storm Perlea really lets fly with proto-Wagnerian drama while the Galop goes with immense zest. In Il Barbiere di Siviglia, Perlea relishes the colours of Rossini’s scoring in a rather cocky, upfront approach. In the coda, he drives the orchestra almost beyond their capacities. La Gazza Ladra is perhaps a little on the steady side, though deliciously pointed and with fireworks in the coda. The most testing pieces here, though, are the other three, which all depend on a perfect timing of their mock seriousness and quirky about-turns. Perlea gets this all right. On this showing, it may be pity he didn’t record any of Rossini’s operas complete. The original LP, by the way did not contain La Scala di Seta, though I’m not quite sure of the history of this.

Vox – Orchestral 4 – later German Romantics
Of Brahms, we have only a selection of Hungarian Dances – nos. 1, 3, 5, 6, 7, 10, 17, 19 and 21. Once again, the orchestra is the Bamberg Symphony (9 September 1958, PL 11240).

These are strict-time dances. In no.1, for example – perhaps the second most famous of them all – the three chords in slower tempo are given their due, but Perlea does not then rush on as fast as possible, he returns to his original tempo. This may sound straight-laced, and yet it all sounds so merry and carefree. How is it done? Partly with acutely flexible phrasing within the strict time. Staying with no.1, the strings are encouraged to a degree of portamento that suggests a gypsy-like abandon rather than classical decorum. Sui generis yet authentic-sounding.

A selection of Wagner was played by the Stuttgart State Opera orchestra. The items were: Tannhauser: Bacchanale, Die Walküre: Ride of the Valkyries, Siegfried: Forest Murmurs, Götterdämmerung: Dawn and Rhine Journey, Funeral March, and Parsifal: Good Friday Music.
The opening of the Bacchanale suggests this might be low-key Wagner. Not at all. Perlea is soon giving weight, colour and tension to notes and details often scuffed over. It’s slowish Wagner, yes, but it crosses the terrain with grand inevitability and, whatever it touches, a conflagration occurs. In the Ride of the Valkyries we see, not just horse-bound Amazons, but a whole civilization passing. Far more than picturesque tone painting is at stake in the Forest Murmurs and the Rhine Journey while the Funeral March is properly shattering. The Good Friday Music is slower than I would want to hear it every time but Perlea’s particular brand of nervous tension sees that it never flags.

Vox – Orchestral 5 – Russians
Perlea’s Italian premières of Mazeppa and The Maid of Orleans show him to have been a convinced supporter of Tchaikovsky’s cause. On record, he got to do only shorter pieces. The most nearly “major” offering in a group set down on 11 April 1954 with the “Vienna State Philharmonia Orchestra” (PL 8700) was Romeo and Juliet. Perlea sacrifices nothing here to formal logic and structure, yet he shows that such an approach need not be at the expense of seething tension and passion. In the last resort, maybe the sheer colossal impact one is occasionally lucky enough to hear as the love theme returns and tragedy approaches is just missed. But this comes rarely in the studio.
 
A steady-as-she-goes 1812 Overture has clean articulation and as much dynamic shading as the piece will allow. There is nevertheless a nervous tension that ensures that musicality does not exclude excitement or conviction. The recording hardly encompasses the final stages but the music comes up remarkably freshly nevertheless.

Perlea extracts a louring passion and inner intensity from the opening stages of Capriccio Italien, expressed in darkly glinting orchestral colours that equate the music to the greater Tchaikovsky. He manages to carry these qualities into the more rollicking sections to a greater degree than one could imagine. Unlike many conductors with a tendency towards slow tempi, he does not object to firing things up towards the end.

He begins Marche slave with an almost desperate, Mahlerian intensity. As the music turns banal – recalling “Nick-nack-Paddy-whack, give a dog a bone” to English ears – Perlea gives an ironic leer to the proceedings. Even he cannot save the final tub-thumping.

A little later, Perlea returned to Tchaikovsky for a Nutcracker Suite (Bamberg SO, 17 September 1956, PL 11390). The Miniature Overture is extraordinarily slow, but beautifully poised, not at all heavy. Most of the other pieces are on the slow side but not remarkably so. Nevertheless, Perlea succeeds in casting a veil of gentle melancholy over the whole suite. By careful articulation and by honing on many piquant details of the orchestration he had me hearing several of the pieces with new ears. The Bamberg SO is not immaculate. There is the suspicion, in the overture, that they had never before been expected to really play every note and were a bit nonplussed by the experience.

Gray’s catalogue mentions an unissued version of the Polonaise from Eugène Onegin dated 10 September 1958.

Perlea’s other recordings of Russian music reveal him to have been a masterly interpreter of that aspect of the Russian psyche that is drawn towards fairy tales, folklore and pictorialism.

A Glinka collection contains Russlan and Ludmila: Overture, Kamarinskaya, Jota Aragonesa, Valse Fantaisie, A Night in Madrid and Life for the Tsar: Overture (Bamberg Symphony Orchestra, 15 September 1957, PL 10600/STPL 510600). Unfortunately, I have only heard the first two of these. In Russlan and Ludmila Overture, Perlea chooses a tempo that allows proper articulation and, with pounding rhythms, it ends up sounding faster than it is. If this is exemplary rather than thrilling, Perlea’s insights come into their own in the evolving colour-variations of Kamarinskaya. His un-hectic but vital approach gets maximum value out of a piece that makes evolving repetitions its creed.
 
Perlea’s warm, atmospheric and detailed approach to Balakirev’s Tamar (Bamberg Symphony Orchestra, 3 June 1955, PL 9530) does not disguise the fact that the music is somewhat repetitive. I have not made comparisons so cannot say whether a more flamboyant approach would improve matters. Those who love the piece will surely love this unhurried, noble interpretation.

Perlea also gaves an exuberant, colourful and, where required, tender, even nostalgic reading of Balakirev’s Islamey in Alfredo Casella’s orchestration (Bamberg SO, STPL 510280). The orchestral version does not seem able to recreate the stunning effect of a pianist pulling out all his stops in what is notoriously one of the hardest pieces in the repertoire. Here it sounds a bit too easy-going, even cosy, but I’m not sure Perlea is to blame.
 
As Perlea begins the Polovtsian Dances from Borodin’s Prince Igor (Bamberg SO, 3 June 1955, PL 9530), one is struck by the open, pastoral quality of the playing. When the famous theme arrives, it is beautifully fresh and tender, without a trace of sultriness. Unfortunately, the solo oboe is not a great player and struggles to meet the conductor’s phrasing. When the cor anglais enters the effect is immediately transformed, quite gorgeous. So, too, is the result when the strings take up the melody – Perlea obtains marvellously detailed phrasing. In the lively music the follows, Perlea sees the music as leaning forwards to Ravel rather than Stravinsky. Tempi are steady and no greater care over balancing and refinement of texture could have been taken if he had been directing “Daphnis et Chloe”. I found this revelatory and right now this is the version I would go to. Only time will tell if I ultimately find myself missing the barbaric drive of most other performances.

With steady tempi, Mussorgsky’s Night on a bald mountain (Bamberg Symphony Orchestra, 3 June 1955, PL 9530) is no less scary for being non-hectic. Strangely, while Perlea related Borodin to Ravel, this Rimskyfied Mussorgsky has much of the hard-hitting austerity of real Mussorgsky. The end is highly poetic though, as in the Borodin, there is a disparity between the quality of the wind soloists. The clarinet is only adequate. The flute has marvellous tone and phrasing but does not always seem in tune.

Recordings under Perlea of Rimsky-Korsakov’s Scheherazade variously claim to be played by the Orchestra Filarmonica Triestina and the Bamberg Symphony Orchestra. The illustrations show that both came out with the same cover illustration, the Vox naming the Bamberg, the Club du Disque naming the Trieste orchestra. The one I have heard says the former and I cannot say whether the two versions are actually different. The Orchestra Filarmonica Triestina is not a pseudonym, in any case. Now known as the Orchestra del Teatro Verdi di Trieste, it was so named from 1944 till 1964. It made a few recordings around the late 1950s under Heinrich Hollreiser, Francesco Mander and, perhaps, this Scheherazade under Perlea. I listened to the recording before I discovered this discrepancy and noted that it was a proficient band, comparable with the Bamberg and Vienna orchestras with which Perlea regularly recorded. This inclines me to think that I have actually been listening to the Bamberg Symphony.

Some will find Perlea too laid back here, an effect compounded by the backward balance given to the percussion, which at times scarcely seem to be present at all. This apart, the recording is decent mono. In truth, Perlea is always very close to Rimsky-Korsakov’s own directions. Speaking as one whose teenage passion for this work – and for Beecham’s recording of it – has irreparably staled, I found this rather refreshing. Perlea is by no means literal when freedom is specifically asked for – as it often is in the wind solos. Moreover, the final section sees the ship powerfully and inexorably driven onto the rocks while the epilogue is played with very great tenderness, so the cumulative effect is greater than the sum of the parts. A Scheherazade to be borne in mind.

It is not often that more than lip-service is paid to the “other” member of the Mighty Handful, César Cui. I’m not sure that a brief, lively but not especially memorable Tarantella (Bamberg Symphony Orchestra, 3 June 1955, PL 9530) amounts to more than lip-service either. Just possibly, the conductor might have added a shade more panache to his well-textured but only moderately vivacious reading, though he does fire things up at the end.

A particularly attractive Perlea disc is his Liadov collection: 8 Russian Folksongs, Baba-Yaga, Kikimora and The Enchanted Lake (Bamberg SO, STPL 510280, the mono version was reviewed in Billboard, May 1957). Perlea makes the best possible case for finding Liadov a major rather than a minor “petit maitre”. His traversals are unhurried and without the hectic quality some feel necessary in Russian music, but this does not preclude finely-sprung rhythmic vitality and drama. Above all, the conductor proves a master of colouring and dynamic shading. The highest point is perhaps the 8 Russian Folksongs, each a tiny masterpiece that makes its point in very little space, reaching real depth and eloquence in the slow pieces. But, from the mock-dramatics of Baba Yaga to the doleful wastes with which Kikimora opens and the almost pointillist tone-painting of The Enchanted Lake, Perlea unfailingly characterizes each piece.

It should be said that Perlea’s concept of Russian music is at the opposite end to the blatant primary colours of an interpreter like Svetlanov. I find it very attractive, but I realise it will seem underpowered to some.

Vox – Orchestra 6 – Slavonic and other national schools
Vltava, from Smetana’s Ma Vlast (Bamberg Symphony Orchestra, 31 May 1955, PL 9500), is portrayed as a lazily flowing, even sluggish river. Unlike some of Perlea’s performances that seem low-key at the beginning, this doesn’t gather all that much steam as it proceeds. If you like an unhurried view, a little-remembered version by the Vienna Philharmonic under André Cluytens finds more bite and contrast in the various episodes.

If Smetana does not seem to have engaged Perlea much, his Dvořák recordings take a high position in both his own discography and that of the composer.

His complete set of the Slavonic Dances opp. 46 and 72 was set down with the Bamberg Symphony orchestra on 3 September 1958 (PL 11240). The first three dances go splendidly, the rhythms and colours pointed yet elastic. At the end of no.3 Perlea inspires the orchestra to something I can only describe as collective euphoria. Thereafter this is not merely excellent, it is great conducting as Perlea, at one with the orchestra, shows how these outwardly homely pieces contain a gamut of human emotion as great as you’ll find in any Mahler symphony. This is not to deny that great conducting is also to be heard from the several famed Czech conductors who have recorded this music; just to say that Perlea is very special and up there with any of them.

It is little short of amazing, by the way, that the conductor who caresses op.72/2 with such loving rubato could have been so straight-laced in the second movement of Schubert 9.

In Symphony no.8 in G op.88 (Bamberg Symphony Orchestra, 6 June 1958, PL 11050), Perlea registers Dvořák’s mood-swings with extraordinary acumen. The usual pastoral poetry is not lacking, but nor are dark shadows. The conductor takes some unusual tempo turns but to my ears he penetrates Dvořák’s psyche as I have never experienced before, revealing an insecurity and instability that make a powerful artistic statement. Even the finale, often taken as straightforward exaltation, proves to have its grotesque elements. None of this would have been possible if Perlea did not exert a remarkable control over phrasing, dynamic shading and colour. The orchestra, if not always immaculate, is with him in every way that matters.

A transfer of this can be downloaded from David Gideon of ReDiscovery. He has extracted some remarkably fine sound for most of the time. Only, since he is working from an LP pressing, there is some end-of-side deterioration. If the original tapes of this are in good condition, maybe we will one day hear this performance in its full splendour. It would be worth it. It’s one of the great Dvořák records.

In the Scherzo capriccioso op.66 (Bamberg SO, 1 June 1955, PL 9500), the outer sections go with plenty of colour and lilt. What really makes this special is the central section. With the help of a very fine cor anglais player, Perlea distils some wonderful poetry here. He also obtains a fine burst of euphoria at the end.

Grieg’s Peer Gynt Suites 1 and 2 were recorded with the Bamberg Symphony Orchestra on 10 September 1957 (PL 121410). Perlea draws considerable expression from “Morning”, “Aase’s Death” and, especially, “Solveig’s Song”, which is expansively but tenderly portrayed. The dances are nicely turned, but “Peer Gynt’s Return”, a tame sea-piece in any hands, is no more than dutiful. The CD transfer I heard didn’t include the first piece of Suite 2, though the original issue had it.

Kodaly’s Dances of Galanta (Bamberg Symphony Orchestra, PL 9500) are notable for their dark passion. The strings really dig into their melodies and there’s a flexibility of phrasing against a pulse that gently sways but never loses its shape. That’s what rubato is all about. Notable, too, is the way Perlea builds the piece up so each climax caps the one before. He has the orchestra practically reach exploding point by the end. It is this sense of an overall shape that eludes an admirable musician like Walter Susskind, who recorded the piece with the LPO in the 1970s. It’s all nicely done, rather sunnier in mood than Perlea’s. But Susskind does not succeed in making an emotional statement out of it.

On home territory, in Enescu’s Romanian Rhapsody no.1 (Bamberg Symphony Orchestra, PL 9500), Perlea goes for the musical values of this piece. Tempi grow steadily, inexorably, generating a powerful impetus, glinting with colour but never flashy. Only, this is early Enesco, not mature Sibelius. Turn to another 1950s performance, by the Los Angeles Philharmonic under Alfred Wallenstein (1952), and every phrase is wilfully twisted and tweaked, milked for effect. There something of the circus about it, but Wallenstein has you reflecting that the music maybe needs this if the listener is to be held to the end. Theoretically Perlea is preferable but …

I mention here, if only to avoid opening a new section for just one disc, a collection of operatic ballet music (PL 9550), performed partly by the Württemberg State Orchestra and partly by the Vienna Symphony. To expected items from Saint-Saëns’s Samson et Dalilah, Gounod’s Faust, Verdi’s Aida and Ponchielli’s La Gioconda, Perlea added, more enterprisingly, a piece from Bizet’s Djamileh. I’ve been able to hear only the first two items. As I mentioned earlier, the Saint-Saëns Bacchanale is perfectly OK but not a patch on the version Perlea made with La Scala Orchestra in 1947. In the Gounod, Perlea stresses the Belle Époque elegance almost to excess, stretching the orchestra up only occasionally. Fistoulari showed that a little more emotion can be drawn from this music without sacrificing its balletic elements.

Vox – Concertos 1 – Cassadò
Vox generally seem to have paired their soloists with whichever conductor was handy. In the case of the Spanish cellist Gaspar Cassadò (1897-1966), something of a stable relationship was set up with Perlea. Indeed, two of Cassadò’s records, the Boccherini/Haydn and the Saint-Saëns/Lalo/Fauré, having been first recorded in Vienna with Rudolph Moralt, were promptly remade in stereo under Perlea. This implies a preference expressed by the cellist, though I haven’t heard the Moralt versions and so cannot comment on any possible lack of rapport. Cassadò’s dark, burnished tone and measured, impassioned style must surely have struck a sympathetic chord in Perlea. It must be said, though, that the recordings sometimes catch a petulant edge to Cassadò’s upper tone. This is not always present and maybe it was not evident at all when you heard him live.

One work where I noticed it was the Boccherini-Grützmacher Cello Concerto in B flat (Bamberg Symphony Orchestra, STPL 510790). Still, with full but transparent textures from Perlea and a strongly felt slow movement, the performance makes a fair case for this outmoded confection.

Sharing the same disc, Cassadò’s and Perlea’s Haydn: Cello Concerto in D will seem gloriously spacious or unduly drawn out according to taste, or even one’s own mood of the moment. Careful rhythmic pacing just about keeps it afloat and one couldn’t describe it as heavy. All the same, the impression that all three movements flow gently in about the same tempo detracts even from the impression that the warmly played middle movement might make if heard on its own. Having written this, a little research shows that, on the strength of their overall timing, Cassadò and Perlea are not, in fact, unusually slow. Evidently the very gently, relaxed phrasing, applied equally in all three movements, creates a deceptive impression.

Cassadò was a long-term promoter of an odd-ball “cello concerto” he arranged himself from Schubert’s Arpeggione Sonata. He recorded it with Barbirolli and the Hallé in about 1927, a live performance from 1940 under Mengelberg has survived, and lastly came this Vox with the Bamberg Symphony orchestra, of which I don’t have the exact date.

I had heard of this “concerto”, but had never actually listened to it. I supposed, innocently, that it would be a straightforward version of the “Arpeggione” as we know it with the accompaniment orchestrated. Cassadò teases us by starting the first movement as if he is going to do just that. Before long he’s off on a jaunt of his own. Schubert’s material is rewritten and reordered and large sections, if by Schubert at all, come from some other work. At the end of the movement he teases us again. Just as you think he’s finished, he comes back with Schubert’s own last few bars.

The second movement, on the other hand, is as Schubert wrote it with an orchestrated accompaniment. I doubt if such a slow performance would come off with just a piano accompaniment. Basically, the last movement is also as Schubert wrote it, though here the orchestral accompaniment is more inventive and colourful. On the last page, Cassadò suddenly breaks loose with some music of his own. His playing of the rondo theme changes the phrasing and substitutes a dogged gruffness for Schubert’s heart-easing songfulness. In truth, I find that Cassadò has succeeded in making the work sound not like Schubert without creating a defined alternative character such as might justify the exercise. It would be interesting to hear how he played Schubert’s original, if he ever did. On this showing, he was no Schubertian. Perlea, whose own Schubertian credentials weren’t entirely proved by his “Great” C major Symphony, accompanies loyally.

I always approach a new performance of Schumann’s Cello Concerto (Bamberg Symphony Orchestra, 15 September 1956, PL 10210) hoping to change my view that it’s a tired and disappointing work by a much-loved composer. Cassadò and Perlea try hard. They give it a sharp rhythmic profile and struggle to find drama to offset the mood of wan introspection. To begin with I thought they might succeed. But this is one of the recordings where there is something unduly doleful in Cassadò’s timbre itself that turns pleading into whining. As it is, I enjoyed mainly the brief central intermezzo, where Schumann’s inspiration momentarily recovers its erstwhile glory – and that’s how I usually end up when I listen to this work.

Cassadò’s and Perlea’s performance of the Dvořák Cello Concerto in B minor op.104 (Vienna Symphony Orchestra [also described as “Vienna Pro Musica Orchestra”], 25 May 1956, PL 9360 can take its place among Perlea’s other Dvořák recordings.

This isn’t the first time I’ve heard the opening of the concerto played slowly and dolefully. Normally, though, the conductor soon has the accelerator down. This is actually the tempo Cassadò and Perlea intend to take. The other movements are also slower than the norm. It might have been risky, but I can only say that I listened entranced as passage after passage was revealed in all its details and colouring. I felt as if I had never fully heard the concerto before. I never felt the music was dragging. As with other revelatory performances at slow tempi, one would not want to hear the music this way always. But I really feel that, if you never hear this version at all, your knowledge of this concerto is incomplete. My own perception of the work, and maybe of Dvořák himself, has been changed for ever.

Cassadò gives a quite extraordinarily free and flexible account Tchaikovsky’s Variations on a Rococo Theme (Bamberg Symphony Orchestra, PL 9360), illuminating corner after corner and almost elevating the music to the level of the Elgar concerto that is the one major omission from his discography – did he really never play it? Whether it is a question of microphone placing or whether Cassadò himself is in a state of exceptional grace I cannot say, but that slightly nasal timbre that can sound petulant elsewhere sounds touchingly wistful here. Maybe it always did sound like that when you heard him live. Perlea identifies completely with the soloist’s view – a task that would have stumped a lesser man.

The outstanding performance in Cassadò’s French coupling (Bamberg Symphony orchestra, STPL 510920) is the Lalo Cello Concerto. One sometimes reads words to the effect that this gets played only because it’s a cello concerto and there aren’t many of them. Cassadò and Perlea make it sound one of the better ones. Cassadò himself, his tone well caught without the nasal edge noted in some of his other recordings, is very free and romantic. He provides passionate rhetoric where called for but in the many gentler passages he enters into a mood of gentle self-communion that removes all doubt as to the genuine inspiration of the music. Perlea is with him all the way, drawing moments of piquant colouring and wistful poetry from the wind instruments in particular. They avoid lapsing into easy jollity in the finale.

In the Saint-Saëns Cello Concerto no.1, Cassadò certainly grabs the attention, pitching in almost feverishly at a rather faster tempo than usual. His combination of fervid passion and inmost communing seem less suitable here than in Lalo, though there’s no gainsaying his dramatic intensity. Perlea follows him loyally but it is only in the minuet-style second section that the performance seems really attuned to Saint-Saëns’s neo-classical muse.

The Fauré Elegie, however deeply felt, underlines points in a way that is never to Fauré’s advantage. Compare this with the performance by Samuel Mayes and the Boston SO under Erich Leinsdorf to hear the music float sublimely onward. Their performance is shorter by less than a minute, yet seems half the length.
 
Vox – Concertos 2 – Pianists
Vox’s mix’n-not-necessarily-match policy came into full play with their pianists. Friedrich Wührer recorded all five Beethoven concertos for them with four conductors. Piano Concerto no. 4 in G op.58 was set down with Perlea (Vienna Pro Musica Orchestra, 12-13 September 1957, PL 10640).

Wührer’s opening statement is crisply rapped out where most others prefer a poetic reverie. Perlea seems in full agreement and the performance is notable for its full-steam-ahead regal grandeur. Wührer does not apparently seek out the intimate poetry of the music, but nor can he be said to negate it. If his point is that with splendid, effortless playing and fidelity to the text – no one could deny he provides these – all the poetry will emerge anyway, he may be said to have proved it. An unassertively magnificent reading.

Guiomar Novaes was another senior artist who recorded extensively for Vox. With Perlea, she recorded Beethoven’s Piano Concerto no.5 in E flat op.73 – “Emperor” (Bamberg Symphony Orchestra, 1957, PL 511930 and Chopin’s Piano Concerto no.1 in E minor (Bamberg Symphony Orchestra, 18 May 1957, PL 10710).

The middle movement of the Beethoven is nicely handled but elsewhere Novaes’s unruffled moderation seems unaware of the greater matters at stake in this concerto. Perlea manages to inject some grandeur into his tutti during the first movement but in the finale the prevailing mood of matronly comfort reduces his contribution to the merely humdrum.
 
Novaes was particularly noted for her Chopin and this performance of the First Concerto impresses by its unforced, natural fluency. A generally strict-time approach is no inhibition to poetry, especially in the second movement. Many passages in the outer movements that are often played more assertively continue in the same gentle manner. However, when something stronger is called for, and also in the left hand during the second subject of the first movement, a hint of lumpiness slightly detracts from one’s pleasure. The unhurried tempo for the last movement will be a sticking-point for some. It enables Novaes to give melodic value to passage-work that in other hands emerges as merely virtuosic. The Novaes approach tends to relate the concerto back to Hummel rather than forward to Liszt. Ultimately, I found it a bit too comfortable. It’s not as if the panache provided by that dashing young man, Artur Rubinstein, is exactly unmusical. Perlea accompanies with warmth and evident feeling, though the orchestra is not always immaculate.

A performance of the Second Piano Concerto which I found on Internet – I forget where – and purporting to be also by Novaes and Perlea with the Bamberg SO, aroused my suspicions since I could find no other trace of their having recorded this work. The performance is clearly different from the Novaes/Klemperer recording, also on Vox. A little research showed it to be identical to another Vox offering, by Orazio Frugoni with the Vienna Volksoper Orchestra under Michael Gielen. Somewhat embarrassingly, it’s a better performance than the Novaes/Perlea version of no.1, even a great performance. Maybe I’ll come back to this in another article – I didn’t know Frugoni was that good.

Monique de la Bruchollerie (1915-1972) was a French pianist noted for a no-holds-barred manner at radical odds with the general perception of French pianism. With Perlea, she set down Franck’s Variations Symphoniques and Rachmaninov’s Rhapsody on a Theme by Paganini (Orchestre de l’Association des Arts Colonne, 1955, PL 9750).

Old Pater Seraphicus this Franck ain’t. From the beginning it is clear we are getting a dramatic interpretation from both artists, with every contrast underlined. Bruchollerie provides some exquisitely shaded answers to Perlea’s peremptory opening phrases, and as the variations proceed she explores Franck’s many troubled inner lines. It’s a remarkable display, even if it seems to be Franck played by someone who would rather be playing Scriabin. I part company with the operation in the finale, however. The hard-hitting approach sounds crude.

The Rachmaninov is remarkably different. It’s a scintillating, mercurial performance that, like Rachmaninov’s own, stresses this work’s unique combination of Mendelssohnian lineage and post-modernity. That said, the 18^th variation is straight to a fault – a tad more indulgence might have done no harm here. Conversely, the straight reading of the final pay-off suggests that humour didn’t come easily to Bruchollerie. The full marks she earned up until at least half-way through had to be slightly modified by the end. Perlea is with her all the way as far as can be heard – the orchestra is sometimes so backward I wondered if it was playing at all.

Bruchollerie and Perlea also set down a Chopin Second Concerto that remained unissued until very recently. I haven’t heard this, but my colleague Stephen Greenbank has and was not at all impressed by Perlea’s contribution.

Vox bagged Maria Tipo to record a pair of Mozart concertos – no. 21 in C K.467 and no. 25 in C K.503 (Vienna Symphony Orchestra, PL 10060), the Schumann Concerto (PL 8540), all with Perlea, and a selection of Scarlatti in 1955, in the wake of her North American debut. This was far more of a sensation than you would think from her subsequent few appearances on disc. I attended a Mozart concerto performance by her in Milan in the late 1970s and went away with no particular impression – I don’t even remember which concerto it was or who conducted, though I do remember that she played Schumann’s “Warum” as an encore.

A stronger impression emerges here. Though the VSO is not immaculate, Perlea opens K.467 with an engaging middle way between majesty and perkiness. Tipo’s playing is crisp, clearly phrased and well shaded. Everything fits into the initial tempo without suggestions of rigidity. No liberties are taken, but none appear to be needed. The only drawback to an otherwise perfect rendition is that Tipo’s tone, as recorded, becomes brittle above mezzo forte. This may not be her fault, since the orchestral strings are also shrill in the higher range, but I do remember feeling, when I heard her live, that she displayed no great tonal depth.

The second movement floats along beautifully and the finale attains operatic bustle without pushing the tempo. I’d dearly love to know who wrote the cadenzas for the outer movements, which veer towards post-impressionism in a rather endearing way. Conversely, places which call for an improvised flourish, such as at the piano’s first entry, before the trill, are left bare. All the same, if the recording could be remastered to tame the harshness, this would be a classic version.

In K.503 the hardness of the piano is scarcely noticeable. Maybe the engineers repositioned the mikes between one piece and the other. A further pointer to the improvised nature of these sessions is the lacklustre opening. Within a few bars things pick up and Perlea offers a steady, insightful introduction. Tipo’s first entry is memorable, seemingly improvised but without waywardness. It sets the tone for a performance of this movement in which every detail falls into place. This is really a remarkable display of unforced, natural musicianship.

Perlea’s introduction to the second movement is sublime, deeply felt without becoming sticky. Tipo is so completely in accord with this view that is impossible to say, in the end which of the two artists it stems from. Hopes that the finale would set the seal on a great performance are not quite fulfilled. We need a joyful release from the inner communing of the middle movement. Instead, Perlea’s conducting of the rondo theme is almost humdrum, Tipo’s playing of it a little prissy. The episodes are better and the central part truly takes wing, after which prissiness returns. This, I have to say, is more the Maria Tipo I remember hearing two decades later, so perhaps this movement provides a clue as to why the 1955 performances were not the first in a glorious series of Mozart concerto recordings. Another drawback is the first movement cadenza. As unidiomatic as those in K.467, it has none of their endearing qualities. It’s rather unpleasant, in fact.

I haven’t heard the Schumann.

RCA – Opera
RCA took to recording in Rome in summer 1954 for economic reasons. Why they chose to have a basically Met cast for Puccini’s Manon Lescaut led by a conductor who had parted company with the Met after a single season is a mystery. We may be glad they did. The full cast was Licia Albanese (Manon Lescaut), Jussi Björling (Des Grieux), Robert Merrill (Lescaut), Franco Calabrese (Geronte/Un sergente degli arcieri), Mario Carlin (Edmondo/Il maestro di ballo/Lampionaio), Anna Maria Rota (Una voce sola), Plinio Clabassi (L'oste/Un comandante di Marina), the chorus and orchestra were those of the Rome Opera House (RCA Victor LM 6116 [USA]/HMV ALP 1326-28 [UK]).

This Manon Lescaut is still a favourite with many opera buffs. Some reservations have to be made over Licia Albanese’s heroine, however. Though 45 hardly seems an advanced age, her voice was beginning to wear. As can happen with a certain type of spinto production, this is more noticeable in the middle and lower registers. It means that she can start an important scene sounding jaded, and resorts to chesty barking for her lower notes, but soars easily upwards as the climaxes arrive. The result is that she sounds slightly elderly in the first two acts but comes more into her own in the last two, giving the best of herself in the final scene.

Björling is a little ungainly in his first entry but even so, there is an obvious difference between the better schooled and extremely good Edmondo, Mario Carlin, and Björling’s tangible vocal personality. For the rest of the opera, Puccini basically requires him to pour forth golden tone, ardent, languid or finally broken-hearted, and he certainly does so. Vocal personality, I suppose, is something Albanese never quite had to that degree – though see my comments below on her four separate Puccini arias. Robert Merrill has it and gives a strong performance, while the rest of the cast, which includes some well-remembered names, in Italy at least, never disappoint.

The singers are greatly helped by Perlea, who belonged to that school of conductors who give precedence to the words and the vocal line rather than the bar-lines. Indeed, such is his fluidity and flexibility that tempi change even within the bar, space is given at the slightest sign of a tenuto. Yet the effect is not wilful; the music always expresses itself naturally. Nor does he lack impetus where needed – indeed, the Intermezzo becomes a little feverish towards its climax. There are no hard edges – long waves of sound support the singers as if on a cushion. This is Puccini conducting at its finest.

1955 was the turn of Verdi’s Aida, with Zinka Milanov (Aida), Jussi Björling (Radames), Fedora Barbieri (Amneris), Leonard Warren (Amonasro), Boris Christoff (Ramfis), Plinio Clabassi (Il Re d’Egitto), Bruna Rizzoli (Una Sacerdotessa), Mario Carlin (Messaggero) and the Rome Opera Chorus and orchestra (RCA Victor LM 6122 [USA], HMV ALP 1388-1390 [UK]).
 
Back in the fifties Zinka Milanov was considered the third – for some the first – of the leading ladies then on the international scene, alongside Callas and Tebaldi. Somehow the public imagination has conserved memories of the Callas-Tebaldi rivalry to this day while allowing the name of Milanov to recede. Maybe in the USA, where she reigned supreme at the Met for so long, things are different. This Aida certainly shows she could stand her own alongside Callas or Tebaldi, and she had certain qualities for which some might legitimately prefer her.

By 1955 she was 49 and had been on the stage for 28 years. There is a momentary impression that this is a rather old voice for Aida, but this impression is swiftly dispelled. It is, perhaps, a voice that is older in its manner and training – is this why her memory has faded more? It’s totally straight, with never a wobble and, though vibrant, it is practically without a trace of vibrato. This enables her to soar up to high pianissimos with a virginal purity and a total control. She does not lack strength, however. More importantly, this is a voice with a timbre that, in some inexplicable way, reaches the heart and moves it.

In 1955, Fedora Barbieri was 35 and in her finest form. It is a more earthy, elemental timbre than Milanov’s, well suited to the more scheming, passionate Amneris. She rises to particularly eloquent heights in the last act as she listens, distraught, to the priests’ condemnation of Radamès.

Jussi Björling must be one of the best-mannered tenors to have ever sung Radamès. Perhaps at the conductor’s insistence, he is the only tenor I have ever heard who comes near to the written pianissimo at the end of “Celeste Aida”. Surrounded by vocal giants, one almost wishes he would thrash out with a few typical tenor mannerisms. On the other hand, he successfully portrays a nice young man caught up in events beyond his control. His singing in the last duet is glorious.

The unmistakeable tones of Boris Christoff lend exceptional authority to Ramfis’s pronouncements. Leonard Warren is an impassioned Amonasro, Plinio Clabassi a strong King. Quite a trio. But every role is cast from strength – back in Italy, Bruna Rizzoli and Mario Carlin carried many an evening on their own, and not just in the provinces.

A cast without weaknesses, but it still might not have come together if the conductor had failed. Perlea is magnificent. At times he is extraordinarily flexible. Try the last duet. The singers have all the time they want to dwell on single notes – Verdi’s “tenuto” markings call for this, but it’s not easy to do it convincingly. The interesting thing is that Verdi’s plink-plonk pizzicatos underneath should lose all rhythmic shape when they’re made to fit such vocal liberty, and yet we still feel the basic pulse underpinning the music. This is true Verdi conducting.

But Perlea can also drive things onward excitingly, while never compelling his singers to leave their words behind, and always finding those moments, even within a strict allegro, where a little extra space is needed to project an important phrase or to introduce a new idea. His very detailed phrasing somehow pervades the entire cast. The big moments, the choruses and the marches, are not underplayed, but textures are kept lean and the picturesque elements are made to seem a backdrop for the intensely human drama unfolding around them. I’m not surprised that some still find this the most satisfying Aida ever recorded.

Before passing on to Perlea’s third RCA opera, it is worth mentioning that on 16 and 21 December 1955 he accompanied Licia Albanese in four Puccini arias. The orchestra was described as the “RCA Italiana Orchestra”. This probably means the recordings were made in New York. The arias were Vissi d’arte from Tosca, Signore ascolta and Tu che di gel from Turandot and, slightly off the beaten track, Ch’il bel sogno di Doretta from La Rondine.

If I complained that Licia Albanese sometimes displayed a slightly old-sounding voice, for all her artistry, in Manon Lescaut, I can only say she is in wonderful form here. Not only is the sound rich and firm, interpretatively these are quite exemplary performances. True, there is a little more gulping and sobbing than you might get from a modern singer, but it’s done with such artistry that I never felt it was over the top. Albanese is mostly known to us from the recordings she made with Toscanini. Away from his controlling presence, she proves to be a much freer and more communicative artist than we might have supposed. Particularly valuable is the relatively little-known aria from La Rondine.

This seems the best place to mention a few more Perlea-conducted arias that are floating around the YouTube.

Firstly, on 28 February 1955, in a live concert with the NBC Symphony Orchestra, Jan Peerce gave an aristocratic and detailed rendering of the Lamento di Federico from Cilea’s L’Arlesiana. Also with the NBC Symphony Orchestra, undated, is a performance by Leonard Warren of Nemico della Patria from Giordano’s Andrea Chénier, which I haven’t heard, and a slightly odd performance of Casta Diva, from Bellini’s Norma, by Zinka Milanov. The small notes are smartly aspirated, in defiance of the more usual sweet legato. I cannot say I prefer it that way.

I suspect these items are all part of a single concert. 28 February 1955 was a significant date for what was still theoretically Toscanini’s orchestra – he had retired the previous year – for in that day the Maestro returned to New York from his final visit to Italy. What would be more likely than that the orchestra, a group of singers who had worked with him, and a conductor he was known to admire, should give a concert in his honour, maybe even hoping – vainly – that he might conduct one or two items himself?

Lastly, on 8 December 1957, in the Carnegie Hall, Perlea conducted an unidentified orchestra in a recital by Licia Albanese. She began with Or sai chi l’onore from Mozart’s Don Giovanni. She does not pretend to be other than the full-voiced lyric soprano we know her to be. The tempo is unusually slow and she allows considerable, but not ungainly, portamenti that would be frowned on today. Unfortunately, the orchestra is recorded so backwardly and distantly as to sound frankly comic, so it is difficult to judge how well the performance worked. Fortunately, in the other items the orchestra, if not as present as one would wish, is reasonably audible.

In a quite extended excerpt from Verdi’s La Traviata, Pura siccome un angelo, the part of Germont is sung by Seymor Schwartzmann. He has a good, even voice, but sings with absolutely no expression or participation. Albanese sings as well as she can with nothing to react to, and Perlea backs her splendidly, disconcertingly lapsing into humdrum indifference whenever the baritone is on his own.

In the taxing L’altra note in fondo al mare from Boito’s Mefistofele, Albanese shows that time has not dimmed her technical command. Her last two arias, Senza mamma from Puccini’s Suor Angelica and Io son l’umile ancella from Cilea’s Adriana Lecouvreur, are frankly glorious. She won me over to an aria that has never seemed to me among Puccini’s most involving, and the Cilea is a classic performance, equal to any I’ve heard.

Back to Perlea’s third RCA opera. Verdi’s Rigoletto, was set down in 1956 with Robert Merrill (Rigoletto), Roberta Peters (Gilda), Jussi Björling (Duca di Mantova), Giorgio Tozzi (Sparafucile), Anna Maria Rota (Maddalena), Silvana Celli (Giovanna), Vittorio Tatozzi (Monterone), Arturo La Porta (Marullo), Tommaso Frascati (Borsa), Leonardo Monreale (Conte di Ceprano), Lidia Grandi (Contessa di Ceprano), Santa Chissari (un Paggio) and, once more, the Rome Opera House Chorus and Orchestra (RCA Victor LM 6051 [USA], RCA RB 16031-16032 [UK]).

This set is a little more problematic. We must try not to be influenced by the stories that have emerged over the years, but it is difficult to discount them entirely. Göran Forsling’s detailed MusicWeb review is well worth reading, but for convenience I repeat his account here.

According to Stephen M. Stroff’s Björling biography there were quarrels between the tenor and the conductor during the recording sessions. This came to a head during the first take of the quartet. Björling was in brilliant form, as can be heard on the recording, but Perlea wasn’t satisfied. He wanted more sotto voce. Björling bristled. ‘This is my solo’, he said, ‘I sing the way I feel’. ‘Softer’, Perlea said. ‘Read the score!’ On the second take Björling whispered his part. Perlea was furious, but the producer Richard Mohr decided in Björling’s favour. It was the first take that was used on the finished recording. Björling was however so furious that as soon as the ensemble scenes were finished he left Rome and returned to Stockholm. There he recorded his solos to the pre-recorded tape.
 
I’ll come back to Björling later, but my first doubts concern the Rigoletto himself, Robert Merrill. It is an excellent account, firmly and clearly sung and not without emotive participation. Maybe on stage he could convince that he was a wronged man with all the world against him, and cracking up under the strain. Just with his voice, he can’t. Obviously, I don’t require the singer’s voice to crack up any more than I expect a Violetta to sing as if she’s tubercular. But I do expect to hear Rigoletto’s plight in the vocal colouring and acting.

One singer who could truly do this was Sesto Bruscantini. When he sang “Cortigiani, vil razza dannata!” you could hear all his accumulated bitterness, rancour and desperation. When he sang that his daughter was all the world to him, he really sounded like that and it was heartrending to hear him. The excellent Merrill, compared to this, is looking at the character from outside. The trouble is that Bruscantini never made an official recording of the role. My knowledge of his assumption comes from five extracts – practically a slightly cut second act – taped from an Italian radio (RAI) programme about Bruscantini. They actually derive from a complete RAI performance of 1963. This can be heard entire on YouTube, but in shallow sound that tires the ear. I see that Bongiovanni have issued this performance. Brief samples can be heard on their website and the sound seems to be very good. This, then, sounds a very serious proposition, especially since the conducting, by Carlo Franci, is notably imaginative, at least in the extracts I have.

Since I’ve mentioned the conducting, perhaps the time has come to say that Perlea is vital, alive to detail and atmosphere. Perhaps because Björling had put him in a grumpy mood, he seems unyielding in places, especially when the tenor is on the scene, and seems less at home in this more intimate canvas than amid the grandeurs of ancient Egypt.

Another issue is the Gilda. Roberta Peters was the Met’s reigning specialist in brilliant, coloratura roles, just as Milanov was their queen of tragedy. Her clear placing of every note in “Caro Nome”, culminating in a totally clean top E held longer than you’d think humanly possible, is a pleasure in itself. But it’s not quite a voice that also carries internal emotion. The Bruscantini/Franci performance had Emilia Ravaglia who, to judge from the Act 2 extracts, conveyed more emotion in her voice, but that is a very partial judgement without hearing her “Caro Nome”. I must say, too, that Peters seems more emotionally involved in the last act, where she is very fine.

And so to Björling. First of all, I don’t hear him in “brilliant form”, let alone “glorious” (pace Forsling) in the solo section that opens the quintet. I hear a singer forcing a voice that’s flagging. I wonder if the real issue was that Björling knew that the “sotto voce” Perlea wanted just wouldn’t come, at least not that day. But maybe the problem is mine. I know there are people who think Björling the greatest tenor that ever lived, but voices and our reaction to them are very personal and I often find Björling emotionally inert. Obviously this is better than being hectored by Del Monaco, and Di Stefano was floundering by the mid-fifties. The Bruscantini performance had Aldo Bottion. The parts I’ve heard suggest a decent provincial tenor just about coping, but I’ll reserve judgement on that. One tenor who had all the nonchalant ease, and the sun in his voice to portray the sort of smiling villain who could deceive poor naïve Gilda, was the very young Pavarotti. He can be heard on a 1967 RAI performance under Mario Rossi. Indeed, considering that RAI also have an old performance – intermittently on Cetra – with Pagliughi as Gilda, it seems they have the ideal singers for each of the three principal roles, but spread across three different performances.

Regarding Björling’s tiff with Perlea, I wonder if the account has not got exaggerated in the telling. The quartet is in Act 3, after all. I can believe that “La donna è mobile”, which comes immediately before, was done as a karaoke, because here the orchestra bulldozes ahead regardless, which I suppose it would if nobody was actually singing. A proper rapport, however sour the two parties felt, can be heard when it is reprised softly after the quartet. I also seem to hear – but perhaps I’m just trying to be wise after the event – Björling’s voice in a slightly different acoustic. I don’t get this impression with “Questa o quella”. Perlea is a bit unyielding, but no more than he was in the ensemble scene that opened the opera. Maybe most of the solo work had been set down before the tiff?

It seems only right to point out that this Rigoletto had another very detailed review in MusicWeb, from Robert Farr. Apparently unaware of the back-story to the recording, he felt Perlea supported his singers well. So maybe it’s best not to try to be too clever – the various parties involved have covered their tracks pretty well. Who knows, after all, how many more such temperamental scenes lie unrevealed behind much-loved opera sets. The smaller parts are well taken – Giorgio Tozzi’s Sparafucile rather more than that, the not very pleasant-voiced Page a bit less.

It would seem, then, that Perlea had shot himself in the foot once more. Unless you are Arturo Toscanini, you don’t tell famous tenors to read the score. In 1957 and 1958 Björling and an RCA team assembled once more in Rome with an assortment of other famous voices to record Tosca and Turandot. The conductor was the rigorous and exacting, but more diplomatic, Erich Leinsdorf. It’s difficult to conclude otherwise than that Björling had vetoed future collaborations with Perlea. Or maybe the refusal came from Perlea himself. Or perhaps those powers that be in RCA just thought it better not to risk putting them together again. Either way, what looked like a nice, ongoing, once-a-year opera recording schedule came to an end, apart from the much later Lucrezia Borgia, discussed below.

Heart attack and stroke
There is, however, another reason why Perlea may have been unable to record a further opera in 1957 – he suffered a heart attack that year. Unfortunately, chronology becomes elusive and several accounts, including the widely-consulted Wikipedia, have lumped his heart attack and his stroke together. I’ll try to sort this out, but I would be very interested to hear from anyone able to fill in the gaps.

According to Grigoriu, some time during 1957 Perlea had a heart attack while conducting Beethoven’s Ninth Symphony in the Carnegie Hall, struggled to the end of the work and then “collapsed after the last chord, like a soldier felled by bullets on the front line”. Grigoriu became a personal friend of Perlea in later years but could not have been an eyewitness in New York in 1957. We will take the account as dramatized but substantially correct. Grigoriu adds that Perlea returned to work as soon as he could, then suffered a stroke on an unspecified date, resulting in the paralysis of the right side of his body and the subsequent reinvention of himself as a left-hand-only conductor.

More helpfully, the Mason City Globe-Gazette, of Iowa, October 28, 1960 reported that “Conductor Jonel Perlea raised his baton for an operatic performance only 19 months after a crippling stroke...”. This puts the stroke around March 1959.

If we assemble such dates as we can find, we see that Perlea recorded for Vox in May and September 1957, and in June and September 1958. This was consistent with his pattern since his Vox recordings began – sessions in late spring and early autumn but not in summer. The nearest to an exact date for the RCA opera sessions I can find is a statement that Manon Lescaut was made in “summer 1954”. It seems likely, therefore, that the Vox sessions bookended the RCA opera sessions in 1954, 1955 and 1956. This is also logical, since, apart from special festivals, opera houses are closed in July and August, so that is a good time to get an international cast together. This being so, there seems no reason why Perlea could not have recorded operas in the summer of 1957 and 1958, since he managed the Vox sessions – and they include some of his best work. Health reasons do not, therefore, account for the termination of his RCA venture.

In June 1957 moreover, Perlea conducted his last concert performance at La Scala, listed above, while in November 1957 he conducted the première of Nicolas Flagello’s Missa Sinfonica with the Manhattan Orchestra. So, in the latter half of 1957 he seems to have been working regularly.

Moreover, in August 1958 Perlea stood in a very short notice for Georg Solti, who had met with an accident, for performances of Mozart’s Die Zauberflöte at Aix-en-Provence. According to the Manchester Guardian of 4 August 1958, Perlea “was all at sea in his tempi, many of which were deadly slow, and seemed doubly so in contrast to Jean-Pierre Grenier’s very skilful production”. Taking over a production at the last moment is never easy, and slowing down what one imagines would have been Solti’s nervously brilliant interpretation without seeming to put a wet blanket on the proceedings is harder still. Furthermore, Perlea obviously would not have had time to work out his interpretation in terms of the producer’s.

The most likely chronology, then, is that he had a heart attack while conducting fairly early in 1957, was back at work, perhaps unwisely, by late spring and soldiered on through 1958, suffering a major stroke in March 1959. In 1959, in fact, he resigned from the Manhattan School, but resumed his work there in late 1960 or early 1961. No Vox recordings are listed in Gray after September 1958. Or rather, some Wagner is listed for 30 October 1959, but on investigation this record – PL 11550 – proves to have been conducted by Hans Swarowsky. Possibly, Perlea was originally booked for the sessions.

After the stroke – New York
Perlea returned to work, then, in late 1960. His activities seem to have concentrated more and more on his Manhattan post. As described above, his period with the Connecticut Symphony orchestra ended in 1965. According to Wikipedia, after his stroke he preferred to concentrate on concerts and recordings rather than opera. If there were any European opera performances after his stroke, the bootleg industry has not yet caught up with them. It would be interesting to know which recordings Wikipedia had in mind, though, since Lucrezia Borgia appears to be the only one in nearly nine years. For whatever reason, no further Vox recordings appear to have been made after September 1958.

Various press reports show, however, that he was far from idle, at least in Manhattan. In November 1966, for instance, he gave the first performance of Joseph Wheeler’s fourth and final completion of Mahler’s Tenth Symphony. Though Deryck Cooke’s completion looks like remaining the standard one, if by any chance a recording survives of Perlea’s performance, it would be very interesting to hear what it sounded like.

A Manhattan performance that has had a fringe existence on record – beyond my reach so far – is that of Vaughan Williams’s Flos Campi, with Lillian Fuchs as viola soloist. This comes from a 1965 concert. One Perlea Manhattan performance that I have heard is an undated live rendering of César Franck’s Symphony. It begins slowly and broodingly, with a considerable impression of suppressed power as Perlea refuses to move ahead. The Allegro is strong and forthright while avoiding any sense of easy gushing. The slow movement has a restrained, doleful quality and the inserted scherzo is kept shadowy. The finale is once again powerful and notably well structured. Though the student orchestra is not of the finest, there is no doubting the intensity Perlea obtains from them, justifying an approach that sets aside the more usual Catholic fervour in favour of something much more darkly troubled.

Perlea did, however, conduct a few operas in concert for the American Opera Society. The most celebrated of these was Donizetti’s Lucrezia Borgia, given in the Carnegie Hall in 1965 by Montserrat Caballé (Lucrezia Borgia), Alain Vanzo (Gennaro), Jane Berbié (Maffio Orsini, Kostas Paskalis (Don Alfonso), Jerold Siena (Jacopo Liveretto), Vern Shinall (Don Apostolo Gazella), William Wiederanders (Ascanio Petrucci), L.D. Clements (Oloferno Vitellozzo), Adib Fazah (Gubetta), Mauro Lampi (Rustighello), Arnold Voketatis (Astolfo) and the American Opera Society Orchestra and Chorus. It has been issued, I’m not sure how officially, by Voce 7 (2LPS) and Standing Room Only SRO 801-2.

My colleague Robert Farr has related the history of the Carnegie Hall performance in some detail. I will only repeat here that it was intended as a star vehicle for Marilyn Horne, who was at that time still undecided between soprano and mezzo-soprano repertoire. She was also pregnant and problems related to this compelled her to withdraw with just a few weeks’ notice. Frantic overtures were made to Sutherland and Gencer, the two internationally established artists who might have managed it. Both were busy elsewhere. Then somebody mentioned a young Spanish soprano who might be able to cope with the part.

The young Spanish soprano was Montserrat Caballé and she did more than cope with the part. She became a star overnight, showered with international engagements and booked by RCA to set down the opera in Rome the next year with the same conductor, but a different cast: Montserrat Caballé (Lucrezia Borgia), Alfredo Kraus (Gennaro), Shirley Verrett (Maffio Orsini), Ezio Flagello (Don Alfonso), Franco Ricciardi (Jacopo Liveretto), Franco Romano (Don Apostolo Gazella), Ferruccio Mazzoli (Ascanio Petrucci), Fernando Iacopucci (Oloferno Vitellozzo), Vito Maria Brunetti (Gubetta), Giuseppe Baratti (Rustighello), Robert Amis El Hage (Astolfo) and the “RCA Italiana Chorus and Orchestra”, presumably the Rome Opera House Chorus and Orchestra. This was first issued on RCA Victor RE/SER 5553-5555 (3LPS, UK no.)

The sound on the original Carnegie Hall performance is pretty good, much better than these lucky survivals tend to be – so much so that I wouldn’t make the recording quality a factor if I preferred the performance.

No doubt RCA thought they were upgrading the performance by replacing the other three principals. I’m not so sure. I yield to no one in my admiration for Alfredo Kraus’s technical security, unfailing musicianship and excellent style. Maybe I’m alone in finding the timbre as such a bit thin and the manner somewhat cold and aloof. If so, I suppose I will be alone in finding that Alain Vanzo combines all the excellent qualities of Kraus with a vibrancy that is more involving.

Much the same goes for Verrett versus Berbié. Maybe it’s a question of studio rather than live conditions. Verrett is undoubtedly excellent, but so is Berbié and she conveys more character. Rather less marginally, I prefer Kostas Paskalis to Ezio Flagello as Don Alfonso. Flagello sings well, with roundly sonorous tone, but that’s the sum of it. Paskalis was known as a character baritone and he brings the words to life to a far greater degree.

However, I don’t want to make too much of this. I think the RCA set would have been better still if they had taken over the Carnegie Hall cast as it stood – comprimari are good in both performances – but it’s not as if the RCA cast does not sing well. They don’t stand in the way of a preference for the RCA recording if there are other grounds for choosing it.

One ground would be that of completeness. The RCA is about twenty minutes longer. The Carnegie Hall version was mercilessly hacked down, presumably by Perlea himself. The RCA text is almost complete – so nearly complete that one wonders why the few remaining cuts, sometimes no more than a chord or two, could not have been opened up as well. Perlea conducts the nearly full version with wholehearted conviction, so had he been instructed by the American Opera Society to finish the evening in under two hours? Moreover, the RCA text adds Lucrezia’s cabaletta in the Prologue – a later addition to the opera – and uses the alternative ending which Donizetti rather reluctantly provided. Originally, Gennaro dies rather slowly and movingly. Dramatically I prefer this, especially with Vanzo to sing it. In the alternative, Gennaro is quickly got out of the way, leaving the stage free for a virtuoso display by the soprano. Caballé admirers will no doubt applaud the decision to use this ending. Donizetti admirers will be sent back to the Carnegie Hall, only to be repulsed by the many other cuts.

Perlea conducts effectively either way, with a strong feeling for colour and atmosphere. In his hands, Donizetti’s orchestration sounds original and piquant. He sets some coursing tempi in the “strettos” but avoids any suggestion that this is early Verdi. In a way, the formula is the same as with Lucia, but with an inspired leading lady instead of a dully correct one. It makes all the difference. At the beginning I thought he was maybe a bit more cautious in the studio but this impression soon passed. The live version has a few imprecisions – this can hardly be an easy opera to conduct with one hand only. As for Caballé, she has perhaps refined further her interpretation by 1966. There is a feeling of history in the making that gives the Carnegie Hall recording a special atmosphere – and plenty of applause which, however deserved, might be trying upon repetition. It hardly needs stating that Caballé admirers will need both – presumably they have both already. The status of the RCA set as a classic for those exploring serious Donizetti beyond Lucia seems fully deserved though, if I had both to hand, I would be tempted to burn a compilation which switches to Donizetti’s original ending – as heard at the Carnegie Hall – for the final few minutes.

On 4 April 1967 it was the turn of Gluck’s Orfeo ed Euridice, given in Carnegie Hall with Dietrich Fischer-Dieskau (Orfeo), Elisabeth Schwarzkopf (Euridice), Lucia Popp (Amor), Veronica Tyler (Blessed Spirit) and the American Opera Society Chorus and Orchestra.

Fischer-Dieskau made various attempts to establish his right to sing the part of Orfeo. In 1956 he recorded it for DG in German under Fricsay. In August 1967 he set it down again for the same company in Italian, with Gundula Janowitz and Edda Moser as Euridice and Amor and Karl Richter conducting the Munich Bach Orchestra. Earlier that same year, however, he took part in this concert performance, which has had some sort of unofficial circulation.

The recording was clearly made by a member of the public with good equipment – no distortion – but sitting well back in the hall. Anyone who wonders why singers record into individual mikes a few inches from their mouths, and who supposes that a single mike back in the hall will record what the public actually hear, could do worse than to ponder this document. The more generalized mix of the chorus and orchestra comes over well enough. Any thought that it might be a relief to hear Fischer-Dieskau from a safe distance is countered by the discovery that his – and the other soloists’ – voices come to us as if rolling down a long, corridor-shaped echo chamber, gathering acoustic muck as they travel. Just why it is that a member of the public hears the voices clearly while a mike sufficiently close to him to pick up his occasional coughs beautifully, registers this acoustic halo, is a phenomenon that an expert could doubtless explain. No expertise is needed to assess the results. I haven’t heard Fischer-Dieskau’s DG recording from the same year, but I assume that no one in their right mind would prefer the Carnegie Hall retrieval. The numerous shades of vocal colour the great baritone doubtless brought to the task are barely hinted at, though even these circumstances cannot dim the tender expression of dignified grief he brings to “Che farò”. The other singers fare no better, maybe worse – Schwarzkopf must have been the most distant of all from the mike.

However, our business is with Perlea and his contribution registers well enough. This is Gluck of yesteryear and if you think there should be a harpsichord continuo, you won’t hear one. Gluck’s recitatives, though, are harmonically complete in the strings, unlike Mozart’s, so let us allow the possibility that Perlea made a conscious decision that Gluck had taken a step towards the future and left behind the world of secco recitative. Perlea avoids injecting 19^th century drama into the score and, though generally stately, is not particularly slow and not at all heavy. In fact, he understands well the structure of the work with its alternating arioso and choral comment and evokes a considerable impression of classical dignity. The edition used is basically the Italian one, with a few additions to Act 2 – chiefly the famous flute solo, unsurprisingly – and minor cuts. No ballet at the end.

One further American Opera Society event was the first American performance of Verdi’s Alzira. Perlea conducted this in Carnegie Hall on 17 January 1968 with Elinor Ross (Alzira), Gianfranco Cecchele (Zamoro), Louis Quilico (Gusmano), Beverly Evans (Zuma), Sidney Johnson (Ovando), Vern Shinall (Ataliba), Frederic Mayer (Otumbo), Michael Devlin (Alvaro) and the American Opera Society Chorus and Orchestra. This has had some kind of bootleg circulation. As of now I’ve heard just one of Alzira’s arias, which can be found on YouTube. The American soprano Elinor Ross (b.1932) was developing a considerable career in the USA and also in Italy but had to retire early through illness. On the strength of this one aria, she had a strong, agile voice with the right Verdian ring. The recording has a similar echo-chamber effect to that of the Gluck, but to a lesser extent – if it’s all like this it should be reasonably tolerable.

Return to Romania
Post-war, Communist Romania was pretty isolated from the rest of the world. In an interview with Mihaela Marinescu, Theodor Grigoriu related how Perlea “was a legend to my generation, like Enescu, like Lipatti. Before my trip to the United States I didn't even know whether he was still alive. A friend I had gone to high school with, Serban Fotino, suggested I should meet Jonel. It was a very impressive meeting, a definitive friendship set in between me and the maestro, despite the age difference. I told him how much his former youth friends admired him, and how wonderfully happy they would be if he accepted to return to this country, as a visitor, or for some concerts” (12).

The seed was sown. Despite the discouragement of some of his American friends, who felt the gesture might be interpreted as supportive of the Communist regime, arrangements were soon made for a return visit, including three concerts. Another factor was the state of Perlea’s health. Though he did not advertise the fact – perhaps only his wife and doctors knew – he was by now suffering from severe lung disease and had not long to live. It was now or never.

Perlea also fitted in a visit to his home village of Ograda. The concert programmes are listed by Grigoriu as follows:

Mozart: Die Zauberflöte: Overture
Beethoven: Symphony no.1 (seemingly replaced by no.5)
Paul Constantinescu: Wedding in the Carpathians: Suite
Mussorgsky-Ravel: Pictures from an Exhibition
At the Athenaeum

Weber: Oberon: Overture
Theodor Grigoriu: Homage to Enescu
Wagner: Tristan und Isolde: Prelude and Liebestod
Enescu: Symphony no.1
At the Radio Studios

Brahms: Symphony no.1
Debussy: Prèlude à l’aprés-midi d’une faune
Richard Strauss: Till Eugenspiegel
Enescu: Romanian Rhapsody no.1
At the Palace Hall

The Romanian Electrecord company issued parts of all three concerts. Very likely, then, they survive complete somewhere. The recording quality is not exactly state-of-the-art for 1969 and the orchestras – variously described as the Bucharest Symphony orchestra, the Romanian Radio National orchestra and the George Enescu State Philharmonic Orchestra, I’m not sure if these are actually three different institutions – are none too secure. Perlea gets a wholehearted response from them all.

From the first concert, the performance of Beethoven’s Fifth Symphony has an almost desperate intensity. Orchestral precision is low and there are signs that the players had some difficulty in following the left-hand-only signals. Nevertheless, for a slow, weighty performance, it has a lot to be said for it.

The first movement is extremely rhetorical. The “motto” theme is not pounded out all that much slower than the main tempo but the unwritten pauses between the opening statements, pauses that even the straightest performance has to insert to some degree, are spelt out to almost Furtwänglerian effect. The repeat is played. The second movement is played with much intensity. Like Klemperer, Perlea has a common tempo for the last two movements and, like Klemperer, it’s a broad one. The effect, though, is not so much of grandeur as of fiery intensity. I keep using the word “intensity” but that’s the impression the performance leaves with me. What with the recording and the orchestral playing, this would be nobody’s first choice, but collectors of Beethoven performances should seek it out.

Rather more of the second concert has circulated. Grigoriu’s Homage to Enescu is an inoffensive but uninspiring middle-of-the-road contemporary work for strings. Perlea seems to be taking great care over nuance and dynamic shading – it’s clearly not his fault if nothing significant emerges.

The Prelude and Liebestod from Wagner’s Tristan und Isolde is not greatly different in concept from the 1949 Met performance. It shows that, even with an orchestra unfamiliar with him and only one hand, Perlea could obtain both detailed shading and fine overall control of the gradually all-engulfing waves of the music, unleashing massive climaxes to both pieces. In any representative compilation of Perlea performances, this would have to be included.

I am not the ideal person to comment on any recording of Enescu’s First Symphony, which was new to me. It certainly gets a free-flowing performance with textural clarity to the fore – I imagine it could sound very congested. Conviction and love shine through the music-making, producing a radiant glow in the middle movement which begins and ends mysteriously, while rising to climaxes of great eloquence.

From the last concert, the opening of Brahms’s First Symphony promises a poundingly tragic vision. On the whole, this is a broad performance, but broad performances generally go in for well-upholstered sonorities. The interesting thing is that Perlea combines a slowish gait with lean sonorities, taut, nervous phrasing and an analytical approach to the texture, insofar as the orchestra allows – the first oboe is a particular problem. Nevertheless, a certain gut conviction makes the first movement worthwhile. The second movement is not actually all that slow. The detailed phrasing gives it a sense of impermanence, like a moment set aside for love in full awareness of the turbulent events taking place outside the cloistered garden. The third movement succeeds in not being soggy at a slowish tempo and the central section, without an increase in pace, is ominously impressive. The finale begins broodingly rather than dramatically. Be warned that the horn player has a strong vibrato but, granted this, he is a good player and ushers in an ethereal vision. The body of the finale is freer in tempo than one might have expected, more of an operatic finale than a symphonic one. I even thought of Die Meistersinger at times. It all works up to a fine blaze. A fallible but thought-provoking performance.

Also from this concert, Debussy’s Prélude à l’après-midi d’un faune is, in some ways, gorgeous. It’s a decadent, sepia-coloured performance in the old French style, perfumed and drifting, perhaps sounding more like Delius than Debussy. It does, however, share another feature with the old-style French performances – ensemble is a relative thing meaning, more or less, that if things go awry, you just catch up when the next section starts. Still, if you don’t try to home in on the details – the recording doesn’t really allow this anyway – the general impression is atmospheric and often very passionate.

Finally, a bit of the concluding performance of Enescu’s First Romanian Rhapsody can be seen and heard on a newsreel film available on YouTube. The picture is black and white but quite clear. With a much wonkier picture, taken from Romanian television, about five-minutes’ worth of the piece has also circulated. The performance has a barbaric splendour that Enescu himself, let alone more sophisticated takes by the likes of Wallenstein or Golschmann, do not attempt. The camera wanders infuriatingly around the orchestra when one would like to see Perlea the whole time. There’s enough to show that, though seated and one-handed, he cut an authoritative figure. Indeed, the clarity of his gestures are such that one does not immediately note that he has the use of only his left hand.

As stated above, Perlea was by now seriously ill. He died in New York on 29 July 1970. He left his musical manuscripts to the Romanian Union of Composers and Musicologists. His parental home in Ograda has been renovated as a museum (see right). Their brochure includes tributes to him from several distinguished musicians, including the singers Tebaldi, Simionato, Stella, Silveri and Taddei. Looking around for suitable phrases to quote, I had to face the fact that they have been translated into such a strange English that I wouldn’t even trust myself to try to put it straight. Linguistically adventurous readers – plus anyone who can read Romanian – should give it a look (13). A Jonel Perlea International Lied Competition has also been instituted in Romania in his memory.

Some concluding thoughts
Jonel Perlea seems to have been posthumously gathered back to his native land. The events of his last years suggest he would have wished this. But what is his legacy to the musical world in general?

The “Forgotten Artists” has never attempted to claim greatness for all the musicians it has sought to remember. Some have done sterling work in a limited geographical area, some have enlarged the repertoire of recorded music, some have been good friends to their contemporary composers. Some, though, have shown signs of greatness, of a greatness that maybe didn’t quite work out. Parallels might be found between Perlea and Vladimir Delman, another conductor who had a knack of digging his own grave. By and large, we have to hear both of them through the filter of less than top-notch orchestras and poor sound. The case of Perlea is a little better in this respect – the Bamberg and Vienna Symphony Orchestras were not all that bad and the Vox recordings could yet be processed to yield reasonable results. And we do have the four RCA opera recordings.

The association of Perlea with budget-label recordings has induced a widespread critical assumption – which not many critics have actually checked – that Perlea was a passably competent hack. Anyone who has read thus far must surely have realized he was not that, or at least not to some ears. The personal nature of his art – a tendency towards a sort of subdued melancholy, coexisting at times with a more euphoric side – may not appeal to all. His Berlioz Symphonie Fantastique and his Dvořák 8^th Symphony – but surely not the Slavonic Dances – may strike some as doleful and lacking in fiery brilliance. I think it is worth trying to get on his wavelength – in a wider range of works than you might imagine, he has a story of his own to tell, he reveals aspects of the music that others have passed by.

Footnotes
(1) Several writings on Perlea by Theodor Grigoriu, a Romanian composer who became a personal friend of Perlea, can be found on Internet. Unfortunately, the one on which I have drawn particularly, dated Bucharest 1999, seems to have disappeared. Grigoriu’s accounts contain many obvious inaccuracies and I have tried to treat as critically as possible any information not corroborated elsewhere.
(2) http://www.amadeusonline.net/almanacco/pag-20?r=&alm_sett=&alm_giorno=&alm_mese=&alm_anno=&alm_testo=Teatro+alla+Scala
(3) http://www.lamaddalena.info/lia_origoni_-_intervista_a_cura_di_c_origoni.html#.V5cw6vmLSUl
(4) Gunther Schüller: A Life in Pursuit of Music and Beauty, University of Rochester press 1911.
(5) http://www.ctpost.com/local/article/New-maestro-for-Greater-Bridgeport-Symphony-5490128.php
(6) http://www.persimmontree.org/v2/fall-2012/the-true-story-of-a-symphony-player-and-the-time-she-broke-through-the-glass-ceiling/
(7) http://connect.msmnyc.edu/Archival_Treasures_Blasts_from_Our_Past
(8) http://www.soundfountain.org/
(9) Carol J. Oja: Making Music Modern: New York in the 1920s, OUP 2000, p.101.
(10) http://www.khaldea.com/rudhyar/listofworks.shtml
(11) For example in Gramophone 8/1969 p.307.
(12) http://icr.ro/bucuresti/dialogue-with-composer-theodor-grigoriu-about-Jonel-perlea
(13) The site of the Perlea memorial House is here: http://www.perlea.ro/index.php?util.index.utl. Unfortunately, I find no trace of the bilingual PDF, describing Perlea’s career, which was available for download on my previous visit to the site.