Chapter 4: Parameters for the Study of
Twenty Archival Sound Recordings

Methods of Approach to This Survey

As has been noted, the Grande Messe des morts neatly fits the suggested requirements for a survey of performance practice. This survey will now concentrate on matters in which the performances, and through them the putative performance practices, can be examined. Obviously, the earliest performances may be examined through contemporary reviews, correspondence or other commentary on the preparations and performances. The usefulness of this form of study is patent but necessarily limited, as witnesses and participants are unlikely to dwell on details in the performance, and in any event performances of the GMdm were (and are) rare enough that comparisons between different performing approaches are difficult if not impossible.

However, the advent of the recording medium now presents the scholar with a unique opportunity. Performances of the GMdm need no longer be ephemeral, since the taste of the music-buying public has been cause for numerous commercial and other ventures to preserve mountings of this difficult work. Today's musician need not content himself with reading the work in score, and may familiarize himself with its sound at leisure from a selection of phonograph recordings, which is precisely the selling point that has made the phonograph a commercial success.

For the sake of this study, it may be presumed that any recording made of the GMdm represents an organization of effort, usually tied to a live performance or series of performances, that will ensure that the conception of the performance/recording represents a considerable outlay of thought and decision-making by the performers. A recording of the work, therefore, may be considered as representative of such a particular approach and style at the time of the recording. (In some cases, as will be seen, the issued recording is in fact a document of a specific performance. These will be indicated as they occur, in order that their relevance to the study may be noted.) When the scholar has assembled a group of recordings of this work, it may be presumed that they are representative of a set of actual performing styles.

With the growing number of off-the-air tapes of radio broadcast performances, and the growing number of public radio broadcasts of notable performances by orchestras and other ensembles, a startling variety of performances of any repertoire work may be assembled. However, it was felt that an investigation of every broadcast performance of the GMdm might be futile, as such might take years for the scholar to assemble, and little guarantee would be available of recording quality from tape to tape.

Hence it has been arbitrarily determined that this study of GMdm recordings be limited to only those performances, commercial or otherwise, which have been issued for sale (or other consideration) in the traditional long-playing disc form. (There is no difficulty with early recordings here, since the only performance ever issued on 78 RPM discs was later transferred to LP, and as such is of course included in this survey.) Such a definition limits the field as well to one in which other scholars might more easily be able to apply their own investigations, which might not be as easily possible were the scope thrown open to performances preserved by other means.

An exhaustive survey of available resources, including record catalogues, journal references, academic institutions, domestic and overseas record merchants and on-line information searches yielded information on twenty-two such issued recordings, of which only two were unavailable and apparently unobtainable by the concluding stages of the research; and one single recording of an individual movement from the GMdm. The parameters for inclusion in this survey were simple: the performance must be a complete or essentially complete performance of the GMdm and must have been issued for sale or other distribution on a set of long-playing discs. Thus the tape recordings have been excluded, except in such cases as the Beecham and Mitropoulos recordings, derived from such tapes; rumored or unavailable recordings have been excluded, such as the two unobtainable recordings mentioned above; and the lone recording of a single movement (the Sanctus, sung by Luciano Pavarotti, conducted by Kurt Herbert Adler) has likewise been excluded, despite its interest from a solo vocal standpoint, as it is first of all not a complete performance, and secondly may not have been connected with any projected or actual performance of the work as a whole, and is therefore arbitrarily taken to be outside of the scope of this survey.

This leaves a field of twenty complete (or essentially complete) performances to be reckoned with. One copy of each of these recordings was obtained in preparation for this research. All were thereupon auditioned repeatedly on the identical reproducing equipment under controlled conditions. The results of these auditions, and the comparisons among them, are the basis for the remainder of this study.

For the sake of concision, these twenty recordings will here be enumerated and identified by the names of their conductors. For further clarity, each performance has been assigned a two-letter code based on the conductor's name. In the case of the one conductor who has recorded the work twice, arbitrary second letters have been assigned in order to differentiate the performances.

Of these twenty recordings, the following conductors have recorded the GMdm commercially and for public sale: Maurice Abravanel (AB); Daniel Barenboim (BA); Leonard Bernstein (BS); Matthias Büchel (BU); Sir Colin Davis (DA); Jean Fournet (FO); Louis Frémaux (FR); Theodore Hollenbach (HO); Lorin Maazel (MA); Fritz Mahler (MH); Charles Munch with the Boston Symphony Orchestra (MX); Charles Munch with the Bavarian Radio Symphony Orchestra (MZ); Eugene Ormandy (OR); André Previn (PR); and Hermann Scherchen (SC).

The following conductors never made commercial recordings of the GMdm, but were honored posthumously with the private release of broadcast performances: Sir Thomas Beecham, Bart. (BE), and Dimitri Mitropoulos (MI).

The following conductors have recorded the GMdm in the course of live performances given by academic institutions, recordings issued privately as mementos by those institutions: Jacob Avshalomov (AV) and Donn Weiss (WE).

The one remaining conductor has performed the GMdm on an occasion which was recorded and later issued by the state-run recording agency of the Soviet Union: Hovannes (sometimes transliterated as "Ogannes") Chekijian (CH). Biographical notes on each of these conductors will be found in Appendix C.

Types of Investigation

The comparison of this population of twenty issued recordings of the GMdm has been made in the following three fashions:

  1. Statistical comparison of movement timings. This is the one manner in which an empirical approach may be taken to compare the recordings in a quantitative and mathematically rigorous manner, that is, by comparison of the timings of individual movements. These timings will serve to indicate general trends and approaches to the tempos chosen by the conductors of the recordings, and as such may serve to delineate the general attitudes toward the nature of the passages in question. These timings may also be compared against the mean timings of the population, and against a presumed ideal timing based upon Berlioz' tempo indications.
  2. Observation of a few specific performance options. Here some varying performance choices and criteria for performance will be examined, along with some speculations offered in partial explanation of these options. As an example of some specific choices, the tenor solo recordings in the Sanctus will be examined critically.
  3. Subjective aesthetic comparison. This will take into account the performances as approaches to the score, and will needfully depend on the personal opinions and aesthetic reactions of the present author, along with some judgements on the observed approaches and options. General conclusions on the twenty-performance comparison will be made as well.

Copyright © 1983, 1995 by Matthew B. Tepper

Table of Contents

Chapter 3

Chapter 5

Back to The Hector Berlioz Page

Back to Matthew B. Tepper's Home Page