TOSCANINI Philharmonic-Symphony Orchestra of New York Complete Recordings ∙ Volume 1 (1929-36) - PASC575

The present series brings together, over two volumes, all of Arturo Toscanini’s studio recordings with the Philharmonic-Symphony Orchestra of New York for the Victor and Brunswick labels, along with the two recordings of Beethoven’s Fifth Symphony that Victor made during live performances in 1931 and 1933. All issued takes are presented, including several which were substituted for the original recordings in the early 1940s, as well as one alternate take unpublished on 78 rpm. As a bonus, the first volume concludes with a previously unissued studio recording which Toscanini made with the NBC Symphony in 1941.

The Philharmonic years (1926 – 1936) were a time of transition for Toscanini, both from a career and an interpretive perspective. During this period, he began to move from being primarily based in the opera house to the symphony hall, culminating in his resignation from La Scala in 1929. From an interpretive standpoint, Toscanini began moving away from the rhetorical emphases and broader tempos we hear in the earlier recordings to the more streamlined approach of his final Philharmonic discs.

The Haydn and Mozart symphonies which begin our program are examples of his earlier style. Although he made a prescient nod toward Historically Informed Performance by using reduced forces (the recording sheets list 55 players for the Haydn and 58 for the Mozart), he was still using a corrupt edition of the Haydn, which “corrected” the dissonance at the beginning of the Trio in the third movement. The approach is more expansive than what he was to do later with the NBC Symphony, lingering over passages which would later be taken more in tempo. In addition, the playing of the Philharmonic, honed by Mengelberg over several years, favors string portamenti to a greater degree than Toscanini would later allow.

A word is in order at this point about the alternate takes. In the early days of electrical recording, it was common for the Victor label to make at least three takes of each side. One would be chosen as the master, with the others marked to be held or destroyed. During the early 1940s, Victor replaced the originally-issued takes on many of its best-selling recordings due to wear on the metal mothers which were used to grow new stampers. Several Toscanini recordings experienced substitutions of this nature, some (as we shall see in Volume 2) having more than one alternate take published. While the alternates are usually fairly similar to the originally-published versions, some noticeable differences occasionally appear.

After frustrating experiences in 1929 with the frequent stops and starts of recording on limited-duration 78s, Toscanini vowed never again to make records. Victor was keen on preserving his interpretation of the Beethoven Fifth Symphony, however, and devised two different methods to record it live in 1931 and 1933. The first attempt (to be featured on Volume 2 of our series) was made direct-to-disc using a pair of cutting tables in relay. Most of the sides ultimately had to be dubbed, which compromised the final sound. The second try, heard here, was recorded initially as the soundtrack on an optical film, which allowed for longer takes and a wider frequency range. The film was then dubbed to disc matrices with greater success than the previous version; but neither recording was approved by Toscanini. In reply to a 1933 EMI request to make additional recordings with the conductor, RCA Victor A&R director Charles O’Connell wrote:

Our last two experiences with Toscanini have been such as to discourage any further attempts to record him; furthermore, having spent in the neighborhood of $10,000 in these attempts [nearly $200,000 in 2019], we are fairly well cured of our ambitions in this direction.

Nevertheless, toward the end of his final season as music director of the Philharmonic in 1936, RCA was able to reach a compromise with the Maestro. Recordings would be made during studio sessions, which would allow for retakes; but they would be taken down without fully stopping after each side. Instead, at pre-arranged points in the score, the conductor would pause momentarily while the next side was begun on a second cutter.

The first movement of the Beethoven Seventh made at these sessions was issued in two very different versions. Take 1 was initially chosen for the first side, and appended to the second takes used for Sides 2 and 3. When the original metal part for Side 1 was damaged during processing in 1942, the second take was substituted. The first take has a broader introduction, and runs more than 20 seconds longer than its replacement.

While there are no titles still unissued from Toscanini’s Philharmonic sessions, we are pleased to present a previously-unpublished studio recording with the NBC Symphony as a bonus to this volume. In March of 1941, Toscanini recorded the Prelude to Act 1 of Wagner’s Tristan und Isolde, but he did not set down the Liebestod at that time. A year later, he returned to record the Prelude together with the Liebestod; but only the latter was issued on 78 rpm. In 1998, I brought out the 1942 Prelude and Liebestod on a Pearl CD set. For the present volume, the 1941 Prelude is being released for the first time, coupled with the Liebestod from the following year. The rubato used by Toscanini in this version of the Prelude (particularly leading up to the climax) and the greater use of string portamenti is reminiscent of the Maestro’s earlier style, which may be one of the reasons he did not approve it.

Mark Obert-Thorn