The Question of Authentic Performance

By Robin Phillips September 14, 2003

In recent years there has been a large movement for ‘authenticity’ in music, meaning music played on period instruments and in the style that it would have been played when originally performed. In Britain alone there are over a hundred professional music ensembles specializing in early music as ‘authentic’ performance has become the preference among the musical intelligentsia. Although I will argue that the early music movement is ultimately a good thing, I will also be suggesting that caution is required. As the ear of the twenty-first century music lover becomes re-trained to appreciate Bach’s music played on the harpsichord, or to appreciate Beethoven played with smaller orchestras that contain a greater balance between strings and winds (not to mention that the A note must be tuned to 435 cycles per second rather than our standard 440), it may be easy to assume that any performance deviating from period authenticity is musically inferior at least or invalid at worst. We may begin to cringe when we hear Bach played on the piano or when we hear a Vivaldi concerto played with violin vibrato, unconsciously feeling that a musical crime is being committed by violating the intentions of the composer. We must ask ourselves, therefore, what reasons exist for the performance of early music and whether these reasons are strong enough to give any inherent superiority to this kind of performance.

One very compelling reason in favour of early music performance is that it does justice to the intent of the composer. One would be scandalized at a performer who took it into his head to add a different chord to a Rachmaninoff concerto, or to substitute his own ending to a Chopin waltz. However, such alternations would represent less of a musical difference to the original as the alterations which naturally occur as a result of the gradual changes in instrument craftsmanship and style of playing. Therefore, one could argue, as much weight should be attached to achieving the period sound as we normally attach to making sure we have the right musical manuscript before us and not a corruption.

The above argument needs to be broken down into its constituent parts in order to be properly evaluated. We start with the need to be faithful to the intents of the composer. From where is this need derived? Is it derived from ethical considerations because we are under some kind of moral obligation to remain faithful to the composer’s wishes? Though it is philosophically problematic to suppose the existence of any such moral obligation, there is an intuitive appeal to such an argument. Any serious artist feels instrinctively that there is something disrespectful – perhaps even wrong – when he hears the first three bars of Mozart’s fortieth symphony beeping out as the ringer of a mobile phone, or when he sees a magnificent book like Le Misrables turned into a stupid cartoon, or when he sees Constable’s The Haywain outlined on black velvet. But I would argue that if such things are wrong, they are crimes against art and not against the dead composer, for we do not normally feel that a dead person’s wishes bind us morally unless there is some added factor that causes it to be so.

Let us assume, however, that the wishes of the composer do carry some special precedent; it cannot be assumed that being faithful to the wishes of the composer is equivalent to a period authentic performance. It is easy for us to impute to past composers thoughts and preferences of our own times and assume, therefore, that their wishes would be for us to perform their music as it would have been performed in their own day. But this is not necessarily the case. We know that in the Baroque period music was much more fluid, with many trios and other compositions being intentionally imprecise about which instruments they were written for. Many of Bach’s pieces were expressly written “for any keyboard instrument”, so one could argue that it is just as faithful to the composer’s intent to perform the works on the keyboard accordion as on the harpsichord. The closer we come to recent times, the more precise composers have striven to be, and we forget that aspects of Baroque pieces would often have been improvised and that the individual performers were given more latitude than they are today.

In addition to this, we cannot assume what a composer would have preferred had he lived long enough to witness the changes brought about in instrument manufacture, to use one example. For all we know, if Bach had lived long enough he may have preferred his pieces played on the piano to the harpsichord. It may be objected by way of Reduction ad Absurdum that we could hypothesize endlessly about what a composer would have preferred: perhaps he would have preferred this particular Eb chord to have been a C minor if he had of thought it. Such an argument is absurd, however, for there is clearly a difference between playing the same piece on an updated instrument and actually tampering with the score itself. It is a feature of musical masterworks that when they are adapted to new musical contexts they offer something new and exciting like fairy tales that can be retold in a multiplicity of ways.

Furthermore, as already mentioned, many ancient composers were not as particular as we assume they were with regard to the instruments on which their music was to be performed. We are conditioned to think that there is only one right way to perform a piece, or at least one right instrument or set of instruments on which an individual piece ought to be performed. As I have pointed out, such assumptions did not always exist within Western music.

Where it is possible to pin point the intentions of the composer, or to achieve a range of possible performances that would fulfil that criteria, is there any reason for assuming that these intentions carry a precedent? Take the example of Frescobaldi who left detailed instructions on how his compositions were to be performed. Many musicians disregard these instructions because they consider them to result in less successful performances. Or take the example of Chopin whose music, at the time, would have been performed in a way that, to our ear, would sound sentimental and ‘overdone.’ This style of performing may or may not be preferable to our own methods, but there seems no reason for assuming it carries any inherent precedent. Just as instrument manufacture may improve, so may performing techniques, and it seems unreasonable that on ancient pieces we should be bound by any imperfections of their day.

This is a point made by James O. Young in his article “The Concept of Authentic Performance.” Young attempts to achieve a precise definition of what is meant by ‘authentic performance’, adducing numerous objections to any formulation. First, he suggests that an authentic performance could be “a performance which reproduces music as it was heard at the time of its composition.” According to Young, such a performance would be either impossible or undesirable. For one, how are we then to perform music that was never performed at the time of its composition but merely written down? And what about cases where the state of instrument repair was in bad condition, as in the case of Haydn who had to contend with rotting oboes, or other composers who had to have their music performed by musicians who were not very competent. To circumvent such difficulties, Young suggests that an authentic performance is one in which a composition sounds the way its composer intended it to sound. But then Young objects to such a performance on grounds similar to my own, namely that even if such a performance were possible, this is not always an ideal worth attaining since the composer’s intentions may be aesthetically unsatisfactory. Young does not think that such a performance is even possible since there is no way of determining a composer’s intentions. Even when we have a recording of a composer performing his own work, there is no guarantee that he realized his own intentions. As a third definition Young suggests that an authentic performance is a performance that makes a piece sound as it would have sounded at the period of its composition, had conditions been ideal. Such a performance, Young argues, is impossible, given the distinction between hearing and hearing-as. Although we might be able to hear the actual noises that a Medieval or Classical listener would have heard, we still do not hear it as they would have heard it. To take a basic example, consider the major third. This is the most basic element of Western art music, at least until comparatively recently. We ‘hear’ the major third as consonant, yet James Young tells us that

In the middle ages a third was a dissonant interval. Medieval listeners hearing (under ideal conditions) a composition which contained a third would hear the interval as dissonant.

Young also points out that, after having been subjected to twentieth century music, it is difficult for us to hear the dissonances of even eighteenth and nineteenth century music. We hear the same notes, but they do not sound dissonant to us. Therefore, he concludes, we simply cannot hear music as it sounded in the past. His conclusion is that authentic performance is not possible and we should think instead about successful performance. However, he suggests that “an early music performance is valuable not because it bears some relation to past performances but because present listeners find it artistically appealing.”

Young’s objection to the possibility of authentic performance in his third formulation might be easily circumvented by defining authentic performance not as a piece that is “heard” as it would have been heard back then, but a performance in which a piece is “played” as near to possible as to how it would have been played at the time of its composition had conditions been ideal. Because the emphasis would then be placed on the performers rather than the hearers, the psycho-auditory perceptual difficulties raised by Young would no longer apply.

R. A. Sharpe raises a similar point in his own objections to Young’s article. Sharpe suggests that Young fails to distinguish between authentic performance and authentic reception. Young slips from “how music sounded” to “how music would have sounded to”. These are not the same and the performer should aim at the former. Thus, Sharpe suggests that an authentic performance can be conceived as a performance “in which the music sounds as it would have done at the time of composition had it been played by competent musicians familiar with the style and performing conventions, playing on decent instruments of the sort which the composer assumed when he wrote the music.”

Having suitably circumvented the problems of definition, Sharpe attempts to briefly defend the primacy of authentic performance. He gives two reasons: an external reason and an internal reason. The external argument is that authentic performance places the music in its appropriate historical-cultural context. Our reception of such music is enriched by this. As Sharpe writes, “If we interpret such art on the basis of our own culture and its assumptions we lose something valuable, an encounter with a different way of doing things.”

The internal argument given by Sharpe is that a musical work commonly has internal relations (the balance of parts, the orchestral colours, etc.) that can best be grasp by hearing the works in the context of their original instrumentation. Granted that Sharpe has a point here, may it not also be the case that additional features of a work may be revealed when the work is performed with different instrumentation. The instrumentational balance of an orchestra of Beethoven’s day was different to the orchestras of our own day; therefore, a performance of the same piece with the former orchestra will have a different quality to a performance of that piece with the later kind of orchestra, and it is reasonable to assume that something will be both gained and lost in either case. To take another example, when recorder music is performed on the flute there is an aesthetic quality that is absent when that music is performed on the recorder; but equally one loses something that can only be captured by the recorder. Similarly, Sharpe acknowledges that one might prefer the aesthetic qualities that come across when Bach is played on a nineteenth century organ (in which the harmonic shifts are more striking) rather than a Baroque organ (in which the counterpoint stands out more predominantly). But, Sharpe adds, this is not the preference of the educated and, he writes, “My case would fail if musicians and music lovers, after listening carefully and intelligently to competent and more or less authentic performances, did not come to prefer the latter.” It seems here as if Sharpe, after everything is said and done, is falling back on aesthetic reasons to make his case (at least his ‘internal’ case) for authentic performance. If this is the case then there is no inherent precedent for authentic performance, only an accidental precedent, for it may always be that in one case competent music lovers may prefer the non-authentic performance. If, at the end of the day, we are simply discussing which kind of performance is aesthetically preferable, then we might dispense with any reference to authenticity at all.

Ironically, the final conclusions of both Young and Sharpe are not as different as they may at first appear. Young says that “an early music performance is valuable not because it bears some relation to past performances but because present listeners find it artistically appealing.” That is also what Sharpe implies when he says that “My case would fail if musicians and music lovers, after listening carefully and intelligently to competent and more or less authentic performances, did not come to prefer the latter.” For both men it is aesthetic considerations that have the final say at the end of the day. Young thinks that the aesthetic reasons furnish good reasons for performance of early music, and Sharpe prefers authenticity for aesthetic reasons, so the disagreement is (or so it seems to me) mainly semantic.

Sharpe’s external argument, namely, that our reception of past music is enriched by an encounter with a different way of doing things carries much appeal with me, yet I recognize that this appeal may, in some cases, have nothing to do with the music itself. I may always prefer the way Chopin is played today yet still be intrigued to hear him ‘over-done’ in the 19^th century style. In some cases, what began as merely historical interest may turn into musical interest, as music which at first sounded strange and alien to us begins to expand our field of enjoyment.

I usually prefer to hear music played on period instruments in period style, but my preference is aesthetically derived: I simply like how it sounds better. In my opinion, this is the only rational ground on which authentic music can be defended; yet it always leaves room for the possibility that, in some cases, the authentic performance may be the aesthetic inferior. Furthermore, just as it is beneficial to arrange the music of the past masters for new combinations, so it is surely beneficial to adapt their music to new instruments. A potential problem with the authentic music movement is that we begin to think of their being only one right way for music to be performed, forgetting that it is a feature of musical masterworks that when they are adapted to new musical contexts they offer something new and exciting.

Return to Robin Phillips HOMEPAGE

You are invited to join my mailing list!

As a member of my mailing list, you will receive automatic notification about additional material and features on this site, as well as occasional newsletters. To join, send a blank email to

largerhope @ tiscali.co.uk

with “Join” in the subject heading. To unjoin, send a blank email with “Unjoin” in the subject heading.

(Note: for anti-spam purposes, the above email address has had spaced inserted before and after the @ sign. The address will only work after deleting these spaces.)