Methodology

1. Scope and purpose

Beethoven.is is an ongoing project devoted to documenting complete recorded cycles of Ludwig van Beethoven's nine symphonies issued on CD or SACD. Its principal aim is to provide an accessible reference for collectors, musicians, researchers and others interested in the history of Beethoven interpretation on record.

The catalogue is intended to be as comprehensive as reasonably possible, but it does not claim to include every complete cycle ever recorded or released. The sheer number of recordings, the existence of numerous historical reissues and the sometimes-limited availability of information make absolute completeness difficult to achieve.

A fundamental distinction is made between an individual recorded cycle and the various commercial editions in which that cycle may have appeared. The primary purpose of the catalogue is to document the recordings themselves rather than every subsequent edition, remastering or reissue.

2. Criteria for inclusion

To qualify for inclusion, a cycle must comprise all nine of Beethoven's symphonies and must have been commercially released on CD or SACD. Recordings issued exclusively on LP, DVD, Blu-ray or in digital-download formats are excluded, regardless of their historical or artistic importance.

The catalogue includes studio recordings, live concert performances, radio broadcasts and archival recordings, provided that they have been issued in the required physical formats.

Not all complete cycles were originally conceived or recorded as unified projects. Some were assembled retrospectively from performances recorded over several years, sometimes with different orchestras and under different recording conditions. Such compilations are included where they constitute a complete recorded cycle issued on CD or SACD.

Consequently, the term complete cycle, as used on Beethoven.is, does not necessarily imply that all nine symphonies were recorded as part of a single planned project, with the same orchestra or within a relatively short period.

A case in point is Carlo Maria Giulini's Beethoven cycle. Between 1991 and 1993, Giulini recorded the first eight symphonies with the Filarmonica della Scala but retired from conducting before completing the project. By then, however, he had already made two commercial recordings of the Ninth Symphony: with the London Symphony Orchestra in 1972 (EMI/Warner) and the Berliner Philharmoniker in 1989–1990 (Sony). The earlier London recording has been used to complete the cycle listed on Beethoven.is, illustrating how a complete cycle may be assembled from recordings originally made independently of one another.

One deliberate omission is the cycle conducted by Maximianno Cobra (Tempus Collection). Described by its publisher as adopting a “re-evolutionary tempi approach that fully liberates both the Apollonian and Dionysian content of these universal masterpieces”, this interpretation bears little relation to Beethoven’s original intentions.

3. Sources and verification

The catalogue has been compiled from a variety of sources, including my personal collection of recordings, classical music discussion forums, record company catalogues, online discographies, advertisements and other internet resources. Printed sources, including books, specialist magazines, journals, newsletters and historical record catalogues, have also been consulted.

Wherever possible, I have sought to establish that a particular recording or release exists or has existed, rather than relying solely on unverified references. Nevertheless, the availability and reliability of information vary considerably, particularly in the case of historical recordings, obscure releases and recordings that have been out of print for many years.

Information relating to recording dates, orchestras, record labels and track listings has been gathered from the available documentation. Despite efforts to ensure accuracy, discrepancies may remain between different sources, and some details may be incomplete or uncertain.

The catalogue should therefore be regarded as a continuously developing reference rather than a definitive or closed discography. Corrections, additional information and details of previously unidentified cycles are always welcome.

4. Recording dates and historical chronology

The dates given in the catalogue refer to when the performances were recorded, not when the recordings were first commercially released.

This distinction is important because a recording may have been issued several years, or even decades, after the performance took place. It is particularly relevant to historical recordings originally made for radio broadcast, recordings preserved in archives and performances released commercially only after the conductor's death.

Where the nine symphonies of a cycle were recorded over several years, the relevant recording period is indicated rather than treating the date of commercial release as the date of the cycle itself.

The use of recording dates also provides a more meaningful basis for examining historical developments in performance practice and interpretation. A recording issued in the twenty-first century may, for example, document an interpretative approach belonging to a much earlier period.

5. Record labels, editions and reissues

Beethoven.is is not intended to provide an exhaustive listing of every commercial edition or reissue of a recorded cycle.

Many historical recordings, particularly those no longer protected by copyright in the relevant jurisdictions, have appeared on numerous labels. Some have been reissued by the original record companies, while others have been made available by different companies, often with new remastering, packaging or catalogue numbers.

Where I am aware that a cycle has appeared on more than one label, I indicate this by adding et al. after the name of one of the labels, for example, "Andromeda et al." This notation indicates that other editions or releases are known to exist, without attempting to provide a complete list of them.

The presence of a particular label in the catalogue should therefore not necessarily be interpreted as identifying the original recording company or the first commercial release.

The emphasis throughout is on the recorded performances rather than the commercial history of every edition in which they have appeared.

Different commercial releases containing the same recorded performances are generally treated as a single cycle, even where their track divisions, indexing or published timings differ. In such cases, the timings presented on Beethoven.is are taken from the edition used as the reference for the catalogue entry.

6. Track listings and recorded durations

The catalogue provides movement-by-movement timings for the symphonies included in each cycle. These figures are taken directly from the track listings of the relevant CD or SACD releases and are reproduced without adjustment.

In a small number of cases, a listed track duration may include a brief period of audience applause, particularly in recordings of live performances. Pauses between movements and other supplementary material are not included in the timings.

The figures should therefore be understood as the published track durations of the recordings concerned rather than independently measured durations of the musical performances.

This distinction is relevant when comparing recordings, particularly where differences amount to only a few seconds. Small discrepancies may reflect differences in track indexing or the inclusion of brief non-musical material rather than differences in the performances themselves.

The timings nevertheless provide a useful basis for examining the considerable variation in the duration of Beethoven's symphonies across conductors, orchestras and historical periods.

7. Tempi, repeats and the interpretation of timings

Differences in recorded duration do not necessarily correspond directly to differences in tempi.

Although the conductor's choice of tempi is an important factor in determining the length of a performance, the observance or omission of repeats can have an equally significant effect. Two performances of the same movement may differ substantially in duration even when their basic tempi are relatively similar.

Beethoven's scores contain numerous explicitly notated repeats, particularly in sonata-form movements, where the exposition is frequently repeated. The treatment of these repeats has varied considerably throughout the history of recorded performance.

Further complications arise from historical performance conventions, including the treatment of repeats in scherzos and trios and their subsequent da capo returns. Depending on the work and the interpretative approach adopted, conductors may observe or omit repetitions that are associated with Classical-period performance practice.

Consequently, a longer performance is not necessarily a slower performance, nor is a shorter performance necessarily a faster one. A conductor who observes an exposition repeat may take considerably longer to perform a movement than another conductor who omits it, even if the former adopts a faster tempo.

Other factors, including tempo fluctuations within movements, differences in phrasing and the treatment of transitions, may also influence the overall duration.

Meaningful comparisons therefore require consideration of the musical structure of each movement and the repeat conventions observed in individual performances.

To assist readers, Beethoven.is provides an overview of the notated repeats in each of Beethoven's nine symphonies, based on the Bärenreiter Urtext edition, together with information relevant to historical performance practice.

The timing tables are intended as a starting point for comparative investigation, not as definitive measurements of the relative speed of different conductors.

8. Statistical charts and historical comparisons

Beethoven.is includes statistical charts designed to illustrate the relationship between recording dates and performance durations.

All timings used in these charts are taken directly from the catalogue's track listings, without alteration. Each point represents one recording and is positioned according to its recording year.

Where a complete cycle was recorded over several years, its position on the chronological axis is determined by the midpoint of the recording period. This provides a consistent means of representing cycles that were not recorded within a single calendar year.

The charts also include a running median, intended to illustrate broader historical trends while reducing the influence of unusually long or short performances.

For each year, the median is calculated from the recordings made within a window extending six years before and six years after that year. The median represents the middle value of the recordings within the relevant period, rather than their arithmetic average.

To avoid presenting trends based on an excessively small number of observations, the running median is displayed only where at least twelve recordings fall within the relevant window. The resulting line is then lightly smoothed to make longer-term developments easier to identify.

A shaded band represents the middle 50% of the recordings within each window, extending from the 25th to the 75th percentile. This provides an indication of the variation in recorded durations around the median.

Recordings for which reliable recording dates are unavailable are excluded from the chronological charts, although they remain included in the catalogue and timing tables.

These charts are intended to illustrate broad patterns rather than establish definitive conclusions about historical changes in tempo or interpretation. In particular, changes in the observance of repeats, the uneven distribution of surviving recordings and the varying number of recordings available from different periods must be taken into consideration when interpreting the results.

9. Limitations and continuing development

Despite the considerable amount of information assembled, Beethoven.is remains a work in progress.

The catalogue is necessarily dependent on the availability and reliability of published documentation. Some recordings may have escaped identification, while others may contain incomplete or conflicting information concerning recording dates, participating orchestras, record labels or track divisions.

The statistical comparisons are subject to similar limitations. Recorded durations provide valuable evidence of interpretative differences, but they cannot, on their own, explain the musical decisions underlying those differences.

The website does not claim to offer a definitive assessment of individual performances or a comprehensive history of Beethoven interpretation. Rather, it seeks to provide a reliable and accessible body of information that may serve as a starting point for further listening, comparison and research.

I welcome corrections, supplementary information and suggestions from collectors, musicians, researchers and other visitors, either via email or Facebook. Such contributions are invaluable in improving the accuracy and completeness of the catalogue and in supporting the continued development of Beethoven.is.