A Callas Recording Update

Since I evaluated the U.S. pressings of EMI's 1997-8 "Callas Edition" in The Opera Quarterly (spring 2000), recorded Callas material has been continually reworked and reshuffled by EMI and other firms, with mixed results. U.S. pressings of the "Callas Edition" have been discontinued, so only the superior Dutch pressings are available. At my suggestion in 1999, EMI corrected a disfiguring editing flaw at Callas's entrance in the 1997 remastering of the 1953 Tosca (EMI 56304-2). Editing and pitch problems introduced into EMI's 1997 remasterings of Rigoletto and six recital discs, however, remain unaddressed despite EMI's assurances. Ironically, the corrected "Callas Edition" Tosca, though still in EMI's catalogue, has largely been replaced by a 2002 remastering flying under two catalogue numbers in the mid-priced "Great Recordings of the Century" (GROTC) series and a third in the budget "Historical Series." The 2002 remastering, again by Allan Ramsay, retains the 1999 correction but begins at a pitch level of a´ = 435 Hz, drifting to standard a´ = 440 Hz by the end, thereby adding well over a minute to a 108-minute opera. Pitch standards in opera houses during the 1950s were, if anything, higher, not lower, than a´ = 440 Hz, as the original EMI Scala LPs attest. The 2002 Tosca sounds vocally distant and artificially lush; previously corrected sudden dynamic shifts at some edit points are back. One should avoid this multimarketed muddle, and seek the now-corrected EU "Callas Edition" version, EMI 56304-2, if available.

The GROTC Tosca initiated a third EMI Callas go-round in that series. The LP sound of the 1954 Norma wasn't very vibrant, having been recorded in Milan's acoustically lifeless Cinema Metropole; but whereas earlier EMI CD versions added a touch of bloom to the sound, the 2003 GROTC version removes all acoustic space, producing a harsh, compressed sound devoid of context and accentuating the overload distortion of the original. Moreover, a cache of pressings made from a defective master contains an unplayable final track on the third CD. The 2004 GROTC issue of the 1953 Lucia is likewise harsh and airless. All of EMI's 1953 Lucias on CD begin low in pitch, drifting upward through the opera. Identical pitching, timings, and retention of misguided CD pauses at some original LP side breaks—for example, the five frustrating seconds of silence between beats before "Ma di', l'amato giovane" in Norma, act 1, scene 2— indicate that, like the "Callas Edition," the GROTC versions may be remastered from 1980s DATs rather than the original analog tapes, if these even survive. The GROTC Trovatore has none of the sweetness, articulate bass, focus, or clarity of the LPs, and compared with previous CD incarnations, it is distorted at some points: the LPs had no distortion. This Trovatore is probably an indifferently modified version of the 1997 "Callas Edition" mastering, which, particularly (if oddly) in its U.S. incarnation, was the best CD compromise. For EMI's latest reissues of the 1954 Puccini and 1958 Verdi recitals in the "Great Artists" and "Legend" series respectively (EMI has endless "series"), we must contend with EMI's confusing 1980s practice of filling out original LP contents with chronologically unrelated material.

The lapse of the fifty-year European Union copyright has legitimized "unofficial" LP transfers of studio recordings. Callas LP transfers on Naxos are thus appearing on the heels of each copyright expiration. So far, we have the 1953 Lucia, Puritani, and Tosca, the 1954 Norma, and the 1952 Cetra Gioconda and 1953 Cetra Traviata. Mark Obert-Thorn transfers the EMI sets, while Ward Marston handles the Cetra recordings. Both men have impeccable reputations for splendid work with 78s. LP transfers, however, present different challenges.

The ubiquitous use of the CEDAR declicking process, with its side effect of "hearing" upper-octave transients as noise and subtly erasing them, is no problem with 78s, which seldom have much high-frequency content. CEDAR's unpredictable characteristic, however, is a distinct problem with full-range LPs. Obert-Thorn's LP transfers are scrupulously—even...