Rimsky-Korsakov had long been engrossed in his treatise on orchestration. We have in our possession a thick note book of some 200 pages in fine hand writing, dating from the years 1873-1874, containing a monograph on the question of acoustics, a classification of wind instruments and a detailed description of the construction and fingering of the different kinds of flute, the oboe, clarinet and horn.1
In his "Memoirs of my musical life" (1st edition, p. 120) the following passage occurs: "I had planned to devote all my energies to the compilation of a full treatise on orchestration. To this end I made several rough copies, jotting down explanatory notes detailing the technique of different instruments. What I intended to present to the world on this subject, was to include everything. The writing of this treatise, or, to be more exact, the sketch for it took up most of my time in the years 1873 and 1874. After reading the works of Tyndall and Helmholtz, I framed an introduction to my work, in which I endeavoured to expound the laws of acoustics as applied to the principles governing the construction of musical instruments. My manual was to begin with a detailed list of instruments, classified in groups and tabulated, including a description of the various systems in use at the present day. I had not yet thought of the second part of the book which was to be devoted to instruments in combination. But I soon realised that I had gone too far. With wind instruments in particular, the different systems were innumerable, and each manufacturer favoured his own pet theory. By the addition of a certain key the maker endowed his instrument with the possibility of a new trill, and made some difficult passages more playable than on an instrument of another kind.
"There was no end to such complications. In the brass, I found instruments with three, four, and five valves, the mechanism varying according to the make. Obviously, I could not hope to cover so large a field; besides, of what value would such a treatise be to the student? Such a mass of detailed description of the various systems, their advantages and drawbacks, could not but fail to confuse the reader only too eager to learn. Naturally he would wish to know what instrument to employ, the extent of its capabilities etc., and getting no satisfactory information he would throw my massive work aside. For these reasons my interest in the book gradually waned, and finally I gave up the task."
In 1891 Rimsky-Korsakov, now an artist of standing, the composer of Snegourotchka, Mlada, and Shéhérazade, a master of the orchestral technique he had been teaching for twenty years, returned to his handbook on instrumentation. He would seem to have made notes at different times from 1891 to 1893, during which period, after the first performance of Mlada, he gave up composition for a while. These notes, occasionally referred to in his Memoirs, are in three volumes of manuscript-paper. They contain the unfinished preface of 1891, a paragraph full of clear, thoughtful writing, and reprinted in this book.2
As the author tells us in his Memoirs (p. 297), the progress of his work was hampered by certain troublesome events which were happening at the time. Dissatisfied with his rough draft, he destroyed the greater part of it, and once more abandoned his task.
In 1894 he composed The Christmas Night; this was the beginning of his most fertile period. He became entirely engrossed in composition, making plans for a fresh opera as soon as the one in hand was completed. It was not until 1905 that his thoughts returned to the treatise on orchestration, his musical output remaining in abeyance through no fault of his own. Since 1891 the plan of the work had been entirely remodelled, as proved by the rough drafts still extant. The author had given up the idea of describing different instruments from their technical standpoint, and was more anxious to dwell upon the value of tone qualities and their various combinations.
Among the author's papers several forms of the book have been found, each widely differing in detail from the other. At last, in the summer of 1905 Rimsky-Korsakov brought his plans to a head, and outlined the six chapters which form the foundation of the present volume. But the work suffered a further interruption, and the sketches were once more laid aside. In his Memoirs, Rimsky-Korsakov explains the fact by lack of interest in the work and a general feeling of weariness: "The treatise remained in abeyance. To start with, the form of the book was not a success, and I awaited the production of Kitesh, in order to give some examples from that work" (p. 360).
Then came the autumn of 1906. The composer experienced another rush of creative energy; his opera, The Golden Cockerel made rapid strides, and kept him busy all that winter and the following summer. When it was finished, in the autumn of 1907, his thoughts reverted to the treatise on orchestration. But the work made little progress. The author had his doubts as to the adequacy of the plan he had adopted, and, in spite of the entreaties of his pupils and friends, he could not bring himself to broach the latter part of the book. Towards the end of 1907 Rimsky-Korsakov was constantly ailing in health, and this materially affected his energy. He spent the greater part of his time reading old notes and classifying examples. About the 20th of May he set out for his summer residence in Lioubensk, and having just recovered from a third severe attack of inflammation of the lungs, began to work on the first chapter of the treatise in its present, final form. This chapter was finished on June 7/20, about 4 o'clock in the afternoon; the same night, the composer was seized with a fourth attack which proved fatal.
The honour fell on me to prepare this last work of Rimsky-Korsakov for publication. Now that Principles of Orchestration has appeared in print I think it necessary to devote a few words to the essential features of the book, and to the labour imposed upon me in my capacity as editor.
On the first point I will say but little. The reader will observe from the Contents that the work differs from others, not merely by reason of its musical examples, but more especially in the systematic arrangement of material, not according to orchestral division in groups (the method adopted by Gevaert for instance), but according to each constituent of the musical whole, considered separately. The orchestration of melodic and harmonic elements (Chapters II and III) receives special attention, as does the question of orchestration in general (Chapter IV). The last two chapters are devoted to operatic music, and the sixth takes a supplementary form, having no direct bearing on the previous matter.
Rimsky-Korsakov altered the title of his book several times, and his final choice was never made. The title I have selected seems to me to be the one most suitable to the contents of the work, "principles" in the truest sense of the word. Some may expect to find the "secrets" of the great orchestrator disclosed; but, as he himself reminds us in his preface, "to orchestrate is to create, and this is something which cannot be taught."
Yet, as invention, in all art, is closely allied to technique, this book may reveal much to the student of instrumentation. Rimsky-Korsakov has often repeated the axiom that good orchestration means proper handling of parts. The simple use of tone-colours and their combinations may also be taught, but there the science of instruction ends. From these standpoints the present book will furnish the pupil with nearly everything he requires. The author's death prevented him from discussing a few questions, amongst which I would include full polyphonic orchestration and the scoring of melodic and harmonic designs. But these questions can be partly solved by the principles laid down in Chapters II and III, and I have no wish to overcrowd the first edition of this book with extra matter which can be added later, if it is found to be necessary. I had first of all to prepare and amplify the sketches made by Rimsky-Korsakov in 1905; these form a connected summary throughout the whole six chapters. Chapter I was completed by the author; it is published as it stands, save for a few unimportant alterations in style. As regards the other five chapters, I have tried to keep to the original drafts as far as possible, and have only made a few changes in the order, and one or two indispensable additions. The sketches made between 1891 and 1893 were too disconnected to be of much use, but, in point of fact, they corresponded very closely to the final form of the work.
The musical examples are of greater importance. According to the original scheme, as noted on the 1891 MS., they were to be drawn from the works of Glinka and Tschaikovsky; those of Borodin and Glazounov were to be added later. The idea of choosing examples solely from his own works only came to Rimsky-Korsakov by degrees. The reasons for this decision are partly explained in the unfinished preface of 1905, but other motives may be mentioned. If Rimsky-Korsakov had chosen his examples from the works of these four composers, he would have had to give some account of their individual, and often strongly marked peculiarities of style. This would have been a difficult undertaking, and then, how to justify the exclusion of West-European composers, Richard Wagner, for example, whose orchestration Rimsky-Korsakov so greatly admired? Besides, the latter could hardly fail to realise that his own compositions afforded sufficient material to illustrate every conceivable manner of scoring, examples emanating from one great general principle. This is not the place to criticise his method; Rimsky-Korsakov's "school" is here displayed, each may examine it for himself. The brilliant, highly-coloured orchestration of Russian composers, and the scoring of the younger French musicians are largely developments of the methods of Rimsky-Korsakov, who, in turn, looked upon Glinka as his spiritual father.
The table of examples found among the author's papers was far from complete; some portions were badly explained, others, not at all. The composer had not mentioned which musical quotations were to be printed in the second volume, and which examples were to indicate the study of the full score; further, no limit was fixed to the length of quotation. All this was therefore left to the editor's discretion. I selected the examples only after much doubt and hesitation, finding it difficult to keep to those stipulated by the composer, as every page of the master's works abounds in appropriate instances of this or that method of scoring.
I was guided by the following considerations which agreed with the opinions of the author himself: in the first place the examples should be as simple as possible, so as not to distract the student's attention from the point under discussion; secondly, it was necessary that one example should serve to illustrate several sections of the book, and lastly, the majority of quotations should be those mentioned by the author. These amount to 214, in the second volume; the remaining 98 were added by me. They are drawn, as far as possible, from Rimsky-Korsakov's dramatic music, since operatic full-scores are less accessible than those of symphonic works.3
At the end of Vol. II I have added three tables showing different ways of scoring full chords; all my additions to the text are marked with asterisks. I consider that the careful study of the examples contained in the second volume will be of the greatest use to the student without replacing the need for the study of other composers' scores. Broadly speaking, the present work should be studied together with the reading of full scores in general.
A few words remain to be said regarding Rimsky-Korsakov's intention to point out the faulty passages in his orchestral works, an intention expressed in his preface to the last edition. The composer often referred to the instructional value of such examinations. His purpose however was never achieved. It is not for me to select these examples, and I shall only mention two which were pointed out by the composer himself: 1. The Legend of Tsar Saltan 220, 7th bar—the theme in the brass is not sufficiently prominent the trombones being tacet (a mistake easily rectified); 2. The Golden Cockerel 233, bars 10-14, if the marks of expression are observed in the brass, the counter-melody on the violas and violoncellos doubled by the wood-wind will hardly be heard. Example 75 may also be mentioned, to which the note on page 63, in the text, refers. I will confine myself to these examples.
In conclusion I desire to express my deep gratitude to Madame Rimsky-Korsakov for having entrusted me with the task of editing this work, thereby providing me with the opportunity of performing a duty sacred to the memory of a master, held so deeply in reverence.
St. Petersburgh, December 1912.
MAXIMILIAN STEINBERG.
Our epoch, the post-Wagnerian age, is the age of brilliance and imaginative quality in orchestral tone colouring. Berlioz, Glinka, Liszt, Wagner, modern French composers—Delibes, Bizet and others; those of the new Russian school—Borodin, Balakirev, Glazounov and Tschaikovsky—have brought this side of musical art to its zenith; they have eclipsed, as colourists, their predecessors, Weber, Meyerbeer and Mendelssohn, to whose genius, nevertheless, they are indebted for their own progress. In writing this book my chief aim has been to provide the well-informed reader with the fundamental principles of modern orchestration from the standpoint of brilliance and imagination, and I have devoted considerable space to the study of tonal resonance and orchestral combination.
I have tried to show the student how to obtain a certain quality of tone, how to acquire uniformity of structure and requisite power. I have specified the character of certain melodic figures and designs peculiar to each instrument or orchestral group, and reduced these questions briefly and clearly to general principles; in short I have endeavoured to furnish the pupil with matter and material as carefully and minutely studied as possible. Nevertheless I do not claim to instruct him as to how such information should be put to artistic use, nor to establish my examples in their rightful place in the poetic language of music. For, just as a handbook of harmony, counterpoint, or form presents the student with harmonic or polyphonic matter, principles of construction, formal arrangement, and sound technical methods, but will never endow him with the talent for composition, so a treatise on orchestration can demonstrate how to produce a well-sounding chord of certain tone-quality, uniformly distributed, how to detach a melody from its harmonic setting, correct progression of parts, and solve all such problems, but will never be able to teach the art of poetic orchestration. To orchestrate is to create, and this is something which cannot be taught.
It is a great mistake to say: this composer scores well, or, that composition is well orchestrated, for orchestration is part of the very soul of the work. A work is thought out in terms of the orchestra, certain tone-colours being inseparable from it in the mind of its creator and native to it from the hour of its birth. Could the essence of Wagner's music be divorced from its orchestration? One might as well say that a picture is well drawn in colours.
More than one classical and modern composer has lacked the capacity to orchestrate with imagination and power; the secret of colour has remained outside the range of his creative faculty. Does it follow that these composers do not know how to orchestrate? Many among them have had greater knowledge of the subject than the mere colourist. Was Brahms ignorant of orchestration? And yet, nowhere in his works do we find evidence of brilliant tone or picturesque fancy. The truth is that his thoughts did not turn towards colour; his mind did not exact it.
The power of subtle orchestration is a secret impossible to transmit, and the composer who possesses this secret should value it highly, and never debase it to the level of a mere collection of formulæ learned by heart.
Here I may mention the case of works scored by others from the composer's rough directions. He who undertakes such work should enter as deeply as he may into the spirit of the composer, try to realise his intentions, and develop them in all their essential features.
Though one's own personality be subordinate to that of another, such orchestration is nevertheless creative work. But on the other hand, to score a composition never intended for the orchestra, is an undesirable practice. Many musicians have made this mistake and persist in it.4 In any case this is the lowest form of instrumentation, akin to colour photography, though of course the process may be well or badly done.
As regards orchestration it has been my good fortune to belong to a first-rate school, and I have acquired the most varied experience. In the first place I have had the opportunity of hearing all my works performed by the excellent orchestra of the St. Petersburgh Opera. Secondly, having experienced leanings towards different directions, I have scored for orchestras of different sizes, beginning with simple combinations (my opera The May Night is written for natural horns and trumpets), and ending with the most advanced. In the third place, I conducted the choir of the Military Marine for several years and was therefore able to study wind-instruments. Finally I formed an orchestra of very young pupils, and succeeded in teaching them to play, quite competently, the works of Beethoven, Mendelssohn, Glinka, etc. All this has enabled me to present this work to the public as the result of long experience.
As a starting-point I lay down the following fundamental axioms:
I. In the orchestra there is no such thing as ugly quality of tone.
II. Orchestral writing should be easy to play; a composer's work stands the best chance when the parts are well written.5
III. A work should be written for the size of orchestra that is to perform it, not for some imaginary body, as many composers persist in doing, introducing brass instruments in unusual keys upon which the music is impracticable because it is not played in the key the composer intends.
It is difficult to devise any method of learning orchestration without a master. As a general rule it is best to advance by degrees from the simplest scoring to the most complicated.
The student will probably pass through the following phases: 1. the phase during which he puts his entire faith in percussion instruments, believing that beauty of sound emanates entirely from this branch of the orchestra—this is the earliest stage; 2. the period when he acquires a passion for the harp, using it in every possible chord; 3. the stage during which he adores the wood-wind and horns, using stopped notes in conjunction with strings, muted or pizzicato; 4. the more advanced period, when he has come to recognise that the string group is the richest and most expressive of all. When the student works alone he must try to avoid the pitfalls of the first three phases. The best plan is to study full-scores, and listen to an orchestra, score in hand. But it is difficult to decide what music should be studied and heard. Music of all ages, certainly, but, principally, that which is fairly modern. Fairly modern music will teach the student how to score—classical music will prove of negative value to him. Weber, Mendelssohn, Meyerbeer (The Prophet), Berlioz, Glinka, Wagner, Liszt, and modern French and Russian composers—these will prove his best guides. It is useless for a Berlioz or a Gevaert to quote examples from the works of Gluck. The musical idiom is too old-fashioned and strange to modern ears; such examples are of no further use today. The same may be said of Mozart and of Haydn (the father of modern orchestration).
The gigantic figure of Beethoven stands apart. His music abounds in countless leonine leaps of orchestral imagination, but his technique, viewed in detail, remains much inferior to his titanic conception. His use of the trumpets, standing out above the rest of the orchestra, the difficult and unhappy intervals he gives to the horns, the distinctive features of the string parts and his often highly-coloured employment of the wood-wind,—these features will combine causing the student of Beethoven to stumble upon a thousand and one points in contradiction.
It is a mistake to think that the beginner will light upon no simple and instructive examples in modern music, in that of Wagner and others. On the contrary, clearer, and better examples are to be found amongst modern composers than in what is called the range of classical music.
My aim in undertaking this work is to reveal the principles of modern orchestration in a somewhat different light than that usually brought to bear upon the subject. I have followed these principles in orchestrating my own works, and, wishing to impart some of my ideas to young composers, I have quoted examples from my own compositions, or given references to them, endeavouring to show, in all sincerity, what is successful and what is not. No one can know except the author himself the purpose and motives which governed him during the composition of a certain work, and the practice of explaining the intentions of a composer, so prevalent amongst annotators, however reverent and discreet, appears to me far from satisfactory. They will attribute a too closely philosophic, or excessively poetic meaning to a plain and simple fact. Sometimes the respect which great composers' names command will cause inferior examples to be quoted as good; cases of carelessness or ignorance, easily explained by the imperfections of current technique, give rise to whole pages of laborious exposition, in defence, or even in admiration of a faulty passage.
This book is written for those who have already studied instrumentation from Gevaert's excellent treatise, or any other well-known manual, and who have some knowledge of a number of orchestral scores.
I shall therefore only just touch on such technical questions as fingering, range, emission of sound etc.6
The present work deals with the combination of instruments in separate groups and in the entire orchestral scheme; the different means of producing strength of tone and unity of structure; the sub-division of parts; variety of colour and expression in scoring,—the whole, principally from the standpoint of dramatic music.
The following is the formation of the string quartet and the number of players required in present day orchestras, either in the theatre or concert-room.
| Full orchestra | Medium orchestra | Small orchestra | |
| Violins I | 16 | 12 | 8 |
| " II | 14 | 10 | 6 |
| Violas | 12 | 8 | 4 |
| Violoncellos | 10 | 6 | 3 |
| Double basses | 8-10 | 4-6 | 2-3 |
In larger orchestras, the number of first violins may amount to 20 and even 24, the other strings being increased proportionately. But such a great quantity of strings overpowers the customary wood-wind section, and entails re-inforcing the latter. Sometimes orchestras contain less than 8 first violins; this is a mistake, as the balance between strings and wind is completely destroyed. In writing for the orchestra it is advisable to rely on a medium-sized body of strings. Played by a larger orchestra a work will be heard to greater advantage; played by a smaller one, the harm done will be minimised.
Whenever a group of strings is written for more than five parts—without taking double notes or chords into consideration—these parts may be increased by dividing each one into two, three and four sections, or even more (divisi). Generally, one or more of the principal parts is split up, the first or second violins, violas or violoncellos. The players are then divided by desks, numbers 1, 3, 5 etc. playing the upper part, and 2, 4, 6 etc., the lower; or else the musician on the right-hand of each desk plays the top line, the one on the left the bottom line. Dividing by threes is less easy, as the number of players in one group is not always divisible by three, and hence the difficulty of obtaining proper balance. Nevertheless there are cases where the composer should not hesitate to employ this method of dividing the strings, leaving it to the conductor to ensure equality of tone. It is always as well to mark how the passage is to be divided in the score; Vns I, 1, 2, 3 desks, 6 'Cellos div. à 3, and so on. Division into four and more parts is rare, but may be used in piano passages, as it greatly reduces volume of tone in the group of strings.
Note. In small orchestras passages sub-divided into many parts are very hard to realise, and the effect obtained is never the one required.
String parts may be divided thus:
| a | { | Vns I div. Vns II div. |
b | { | Vns II div. Violas div. |
c | { | Violas div. 'Cellos div. |
d | { | 'Cellos div. D. basses div. |
Possible combinations less frequently used are:
| e | { | Vns I div. Violas div. |
f | { | Vns II div. 'Cellos div. |
g | { | Violas div. D. basses div. etc. |
Note. It is evident that the tone quality in b and e will be similar. Still b is preferable since the number of Vns II (14-10-6) and Violas (12-8-4) is practically the same, the respective rôles of the two groups are more closely allied, and from the fact that second violins generally sit nearer to the violas than the first, thereby guaranteeing greater unity in power and execution.
The reader will find all manner of divisions in the musical examples given in Vol. II. Where necessary, some explanation as to the method of dividing strings will follow in due course. I dwell on the subject here in order to show how the usual composition of the string quartet may be altered.
Stringed instruments possess more ways of producing sound than any other orchestral group. They can pass, better than other instruments from one shade of expression to another, the varieties being of an infinite number. Species of bowing such as legato, detached, staccato, spiccato, portamento, martellato, light staccato, saltando, attack at the nut and at the point, 
and 
(down bow and up bow), in every degree of tone, fortissimo, pianissimo, crescendo, diminuendo, sforzando, morendo—all this belongs to the natural realm of the string quartet.
The fact that these instruments are capable of playing double notes and full chords across three and four strings—to say nothing of sub-division of parts—renders them not only melodic but also harmonic in character.7
From the point of view of activity and flexibility the violin takes pride of place among stringed instruments, then, in order, come the viola, 'cello and double bass. In practice the notes of extreme limit in the string quartet should be fixed as follows:
for violins: 
, for violas: 
,
for 'cellos: 
, for double basses: 
.
Higher notes given in Table A, should only be used with caution, that is to say when they are of long value, in tremolando, slow, flowing melodies, in not too rapid sequence of scales, and in passages of repeated notes. Skips should always be avoided.
Note. In quick passages for stringed instruments long chromatic figures are never suitable; they are difficult to play and sound indistinct and muddled. Such passages are better allotted to the wood-wind.
A limit should be set to the use of a high note on any one of the three lower strings on violins, violas and 'cellos. This note should be the one in the fourth position, either the octave note or the ninth of the open string.
Nobility, warmth, and equality of tone from one end of the scale to the other are qualities common to all stringed instruments, and render them essentially superior to instruments of other groups. Further, each string has a distinctive character of its own, difficult to define in words. The top string on the violin (E) is brilliant in character, that of the viola (A) is more biting in quality and slightly nasal; the highest string on the 'cello (A) is bright and possesses a "chest-voice" timbre. The A and D strings on the violin and the D string on the violas and 'cellos are somewhat sweeter and weaker in tone than the others. Covered strings (G), on the violin (G and C), on the viola and 'cello are rather harsh. Speaking generally, the double bass is equally resonant throughout, slightly duller on the two lower strings (E and A), and more penetrating on the upper ones (D and G).
Note. Except in the case of pedal notes, the double bass rarely plays an independent part, usually moving in octaves or in unison with the 'cellos, or else doubling the bassoons. The quality of the double bass tone is therefore seldom heard by itself and the character of its different strings is not so noticeable.
The rare ability to connect sounds, or a series of sounds, the vibration of stopped strings combined with their above-named qualities—warmth and nobility of tone—renders this group of instruments far and away the best orchestral medium of melodic expression. At the same time, that portion of their range situated beyond the limits of the human voice, e.g. notes on the violin higher than the extreme top note of the soprano voice, from
upwards, and notes on the double bass below the range of the bass voice, descending from
(written sound)
lose in expression and warmth of tone. Open strings are clearer and more powerful but less expressive than stopped strings.
Comparing the range of each stringed instrument with that of the human voice, we may assign: to the violin, the soprano and contralto voice plus a much higher range; to the viola, the contralto and tenor voice plus a much higher register; to the 'cello, the tenor and bass voices plus a higher register; to the double bass, the bass voice plus a lower range.
The use of harmonics, the mute, and some special devices in bowing produce great difference in the resonance and tone quality of all these instruments.
Harmonics, frequently used today, alter the timbre of a stringed instrument to a very appreciable extent. Cold and transparent in soft passages, cold and brilliant in loud ones, and offering but little chance for expression, they form no fundamental part of orchestral writing, and are used simply for ornament. Owing to their lack of resonant power they should be used sparingly, and, when employed, should never be overpowered by other instruments. As a rule harmonics are employed on sustained notes, tremolando, or here and there for brilliant effects; they are rarely used in extremely simple melodies. Owing to a certain tonal affinity with the flute they may be said to form a kind of link between string and wood-wind instruments.
Another radical change is effected by the use of mutes. When muted, the clear, singing tone of the strings becomes dull in soft passages, turns to a slight hiss or whistle in loud ones, and the volume of tone is always greatly reduced.
The position of the bow on the string will affect the resonance of an instrument. Playing with the bow close to the bridge (sul ponticello), chiefly used tremolando, produces a metallic sound; playing on the finger-board (sul tasto, flautando) creates a dull, veiled effect.
Note. Another absolutely different sound results from playing with the back or wood of the bow (col legno). This produces a sound like a xylophone or a hollow pizzicato. It is discussed under the heading of instruments of little sustaining power.
(These instruments give all chromatic intervals.)
Black lines on each string denote the general range in orchestral writing, the dotted lines give the registers, low, medium, high, very high.