The individual volume of the various instruments is a major concern when many different instruments play together. The sound of a trumpet is more powerful than that of a flute, but exactly how much more powerful is it? And how does the volume of specific instruments vary between different registers? While a low flute is very soft (and has very limited penetration power due to few overtones), a flute in the highest register penetrates easily. And a high-pitched solo for piccolo will often be easily audible even in a full orchestral tutti. Conversely, a low oboe is far more penetrating than the same instrument in its highest register. Knowledge of the characteristics of volume and timbre in the different registers of individual instruments is a fundamental requirement for any instrumentation. In the case of some instruments, even adjacent notes may involve considerable differences.
How powerful is a group of violins, for example? Richard Strauss (and many others) generally required 16–18 1st violins and 16 2nd violins for a full-scale symphony orchestra. But will 16 violins be 16 times louder than one? This would not be an unreasonable assumption. However, would it not make life difficult indeed for a violin soloist, struggling alone to be heard above a crowd of 32?
The problem involves a difference between the measured sound pressure level (SPL) and the volume as perceived by the human ear. Two people may perceive a sound to be “twice as loud” as another, but may well disagree on the exact meaning of this. The SPL unit decibel (dB)became widely used in the 1920s. Another unit of measurement, phon, was introduced shortly after the war in an attempt to establish a standard for volume as perceived by the human ear. However, phon soon sank into oblivion (the technical and scientific aspects of this are beyond the scope of this presentation), but the difference between dB and phon may still illuminate why two violins are not perceived to be “twice as loud” as one. Measured in decibel, they are indeed twice as loud, i.e. their SPL is twice as high. But the human ear perceives the sound to be only slightly louder. When 16 (or 32) violins play together, the ear recognizes the sound to be somewhat louder, but certainly not many times louder. Instead (and notably), it will be perceived as a different type of sound – full, rich, more mellow and far more dense.
In his enormous four-volume work Traité de l’orchestration (1939–43, published 1954–59),the important but today unfairly neglected French composer Charles Koechlin does in fact suggest a thought-provoking distinction between the volume of a sound and the power of a sound, where “volume” refers to sound density, regardless of power. Koechlin’s use of the term associates a horn or a tuba with greater volume than a violin. Greater volume in this specific sense may be linked to instrumental sound spectra with few high overtones. To describe the sound of a large group of strings as having a greater volume than a solo violin would thus more aptly indicate their difference than to express this difference in terms of increased sound power. And as already mentioned earlier, a doubling at the perfect unison leads to an increase in volume (in the above sense) rather than to an increase in sound power.
Today, decibel is universally used as a standard unit of measurement, partly because perceived sound power has now been integrated into the measurement scale. The threshold value, the softest sound the human ear can perceive, is expressed as 0 dB, and the so-called “pain threshold” as 130 dB. In practice, however, this precise measurement of sound power proves to be practically useless when it comes to dynamics and balance in music. Measuring the power of individual instruments in an orchestral tutti in terms of dB is unlikely to be of any use to orchestrators, due to far too many constantly changing factors: different traditions in instrument manufacture, playing style, hall acoustics, spatial placement of instruments, individual musicians’ instinctual adjustment of balance, etc.
What is more, balance is not merely a matter of sound power and volume in the traditional sense, but to a great extent linked to the specific character and function of musical elements. A melodic line (or the highest part) always attracts attention – activity and movement capture more attention than, for example, sustained notes. Contrast in sonority or register can also have a major impact on how dynamic relationships are actually perceived.
A particular problem area concerns the chamber orchestra, such as the nowadays very popular “sinfonietta” ensemble in which solo string players are pitted against solo woodwind or brass players. While string instruments have remained practically unchanged since the end of the 1700s, brass instruments have changed substantially and are considerably more powerful today than they were during the classical era. This inevitably calls for meticulous attention to balance when solo strings interact with solo brass instruments. And it might be mentioned in passing that while doubling 1st and 2nd violins in a full orchestra merely results in greater density (or “volume”), two solo violins in perfect unison often sound poor, if not to say “impoverished” …
When it comes to handling internal orchestral balance and dynamics in practical terms we must – as in so many other areas of orchestration – avail ourselves of accumulated knowledge, craftsmanship and many centuries of surviving experience. Plainly, such knowledge cannot be substantiated by well-tested theoretical, let alone scientific arguments, but this is partially due to the absence of anything resembling a general theory of the orchestra in the theoretical literature on music.
Possibly the only, certainly the simplest knowledge handed down by history can be found in the form of simple numerical relationships in Rimsky-Korsakov’s textbook on orchestration, hereinafter referred to as “balance ratios.” Based solely on experience, this represents an attempt to define the basic sound power relationships between orchestral instruments. These balance ratios represent a purely “statistical” average – all the inevitable differences in instruments, musicians, acoustic issues, etc. have already been outlined above. Besides, a conductor can always adjust the balance at his or her discretion. Nevertheless, practical examples show that these ratios, in spite of their extraordinary simplicity, contribute to a greater insight into the effects and workings of orchestral balance. As a practical tool to understand and achieve proper and/or desired balance, this method is unique in the general literature on orchestration.
Rimsky-Korsakov attempted to take into account instrumental differences in piano and forte, but this differentiation is likely to overestimate the applicability of balance ratios. As a statistical foundation, forte seems to be most suitable, since instrumental power varies considerably in piano. A muted brass instrument, for example, or a string group, can play extremely softly, while very soft dynamics can be tricky for a low oboe, a very high clarinet or a high trumpet without mute. Because the number of strings can vary between different orchestras, Rimsky-Korsakov suggested a further differentiation, but this is not corroborated by modern SPL measurements, and composers rarely specify the desired number of strings in the score – today, standard classical works are performed with anywhere from about 20 to more than 60 string players.
The present text will use a simplified version of Rimsky-Korsakov’s balance ratios:
One group of strings (e.g. 1st violins) · 2
One woodwind instrument · 1
One horn (or a saxophone) · 2
One brass instrument · 4
It may seem surprising that four clarinets are needed to match one trumpet. Nevertheless, these numerical relationships are largely confirmed by practical experience, although they admittedly are beyond scientific proof. Their applicability is largely limited to tutti, chords, melody and accompaniment, etc. In smaller groups, each musician will adjust his or her dynamics to the rest of the ensemble.
With regard to the traditional repertoire, one must bear in mind the more limited power of brass instruments until the mid-1800s, and the fact that natural brass instruments possess only a limited repertoire of notes in the low register. Since all of these factors may influence the balance ratio (BR), its interpretation and use should be taken with a grain of salt.
· · ·
· · ·
 Regrettably, this monumental textbook has never been translated.