Reflection on transcription – using a vocal, poetical & instrumental approach


Having made at least thirty transcriptions of Colloque sentimental for cor anglais, I have been reflecting continuously on every choice I made. First of all, I naturally made the decision to transcribe the poem for English horn and not for oboe or oboe d'amore. Because of its original key of A minor and quite low tessitura, the mélodie is suitable for a lower voice like baritone or in my case (mezzo-)alto and thus also for cor anglais – the alto version of the oboe, relatively close to the human voice and transposing a fifth lower than the oboe (the English horn part notated in my transcription a perfect fifth higher than it sounds), which makes the melody quite comfortable in terms of tessitura (not too high and not too low) and key: playing with one sharp being actually more easy on oboe(-related instruments) due to its fingerings. Crucial in this decision to play the art song on cor anglais, after experimenting playing the mélodie on oboe as well as on English horn for a while, was when I discovered that singing a particular phrase of the melody felt exactly the same as playing it on cor anglais (contrary to oboe), namely “Ah! Les beaux jours de bonheur indicible” (Experimentation and reflection: audio fragments 21a & 21b) – “indicible” meaning inexpressible or unspeakable. The (mezzo-)alto voice as well as English horn usually perform in a lower tessitura and have to work harder in order to obtain the same legato, connection, freedom and fullness of sound in higher written passages. Hence, due to their relatively high setting, particular phrases like “Ah! Les beaux jours de bonheur indicible” (audio 21) demand more energy from the singer or player, so the sensation is very similar for me to either sing it with my voice or with the cor anglais.

Another aspect that played an important role in this decision for cor anglais is the 'timbre', being close to an alto voice. Its soft and melancholic tone seems to be ideal for evoking the atmosphere of a frosty, deserted park. In addition, its possibilities in articulation and colouring – thanks to the (bigger) size of the reed and instrument – make the English horn different from the oboe, including onset (tonguing and articulating) and (breath) flow. What is so intriguing about the translation from voice to cor anglais, is indeed for instance this concept of attack and articulation. Whereas a singer is able to attack a tone completely silent, besides a breath impulse (avoiding any sound in the throat), an oboist always needs to tackle the resistance of the instrument by air(speed) and articulation in order to make the reed vibrate, which often leads to an audible attack that partly defines the instrument. An audio example that clearly demonstrates this inevitable difference between voice and cor anglais is the second half of fragment 23 (from Experimentation and reflection): “Et la nuit seule entendit leurs paroles” in which phrase the words 'Et' and 'entendit' both begin with a vowel sound (audio 23a) – [e] and [ã] – and should start without onset, very difficult to imitate on the English horn unfortunately (audio 23b, second half). On the other hand, while the beginning of a phrase or note can consequently be very defined and present with the oboe or English horn, the timing will be different with the voice, because of the placement of consonants slightly before or after the beat – in the case of final consonants (like the ending [r] in 'l'espoir', audio fragment 22a and 22b: the ending of the sung and played tone and syllable [pwar] slightly varying in decay)  to enable legato(,) singing on the vowels (the vowel being vital).

Therefore, it is especially interesting, in case of consonants and legato lines, to imitate the voice on oboe(-related instruments) since it will demand a flexible and varied articulation, many times sustained or fully legato, but also sometimes asking for little cesuras and commas for expressivity, esthetics and diction. To achieve a similar result in playing as in singing, one could for example consider writing a comma or breathing mark but also a short (sixteenth) rest in the transcribed English horn part when there would be a breath or small expressive pause in the sung part, still suggesting by slurs the continuation of the phrase(s) (as written and shown in the Transcription, for example in bars 29, 34 & 44). A perfect example of playing a musical (non-textual) comma is the phrase Pourquoi voulez-vous donc qu'il m'en souvienne?”, in which the written comma (bar 25 of the Transcription in PDF) is not only expressive but also indicating the time that it takes to pronounce two similar consonants: two times the [k] which is the ending sound of 'donc' and the starting sound of 'qu'il', therefore taking more time in the singing (audio 19a)  a timing I tried to match when playing the phrase on cor anglais (audio 19b). Sometimes, a different solution is indeed needed to create the same idea in the sung and played phrase, mainly if it concerns articulation. Even when the voice starts with a vowel, like this extended “Ah!” in the phrase mentioned before, the same note can actually demand an accent from the English horn player to provide a similar effect (instead of starting the tone only on air(speed)), because of the difference in energy from the beginning of a sung or played phrase – in which case it would actually help to effectively make the vocal sound/vowel 'a' [a] with your mouth whilst playing, in order to colour correctly (as can be heard in fragment 21b). 

When it comes to colouring, it would be interesting to pay particular attention to the sung vowels – at least as important as the consonants and articulation we discussed before. In the French language there are no less than eleven vowel sounds; even when excluding the four nasals [ã], [ɛ̃ ], [ɔ̃] and [œ̃], making fifteen (!) vowels in total. These eleven vowel sounds are all indicated by their phonetic symbols as seen in the diagram I made below, with the mixed vowels in the middle: literally between tongue and lips, mixing both in vowel sounds 9, 10 and 11 ([œ] and [ə] being virtually the same sound)). The sound depends upon the shape of the mouth cavity, which may be altered by changing the position of the tongue and the lips (Pierre Bernac suggested to try the following: [i]-[y]-[u], [e]-[ø]-[o] and [ɛ]-[a]-[ə]/[œ]-[ɔ]). When the tongue is near to the palate, the vowel is called closed: [i]. When the tongue is far from the palate, the vowel is called open: [a].



Obviously, I took these vowel sounds into account while experimenting with playing the mélodie on English horn, trying to match the vocal and instrumental colours as demonstrated in the audio files from the chapter Experimentation and reflection where I sing and play alternately (audio fragments 15 to 23, the first one ('a') being sung and the second ('b') played). In this way, the vowels highly influenced my playing on English horn, since I was constantly adapting my embouchure (the combination of mouth cavity, tongue and lips) in order to play the vowel sounds as faithfully as possible; playing nasal sounds nasally while (like in singing) aiming for open nasality, playing closed vowels with less space and a more focused tone and open vowels with as much room (in the mouth and body), looseness (in the lips) and fullness as possible – always looking for a rich palette of different (vowel) colours, depending each time on the meaning and character of the poetry and melody.

For example, considering the first phrase of the poem “Dans le vieux parc solitaire et glacé, deux formes ont tout à l'heure passé”, I am playing all the round vowels (such as the first nasal [ã], the schwa [ə], [ø], [u] and [œ]) with a very round embouchure (lips and oral cavity) as if I am speaking them and thus colouring them darker, fuller and rounder, in contrast with the [a] in 'parc' that I give a much brighter and more open colour as demonstrated in audio 15b and of course clearly audible in the sung part 15a. Furthermore, when displaying different characters like these of the narrator, the first and/or the second ghost, I use three opposite colours to portray them. This becomes very apparent in the dialogue between the two spectres, for instance when we compare the question from the first speaker – in a light, mild, sweet colour with a sense of desperation:“Te souvient-il de notre extase ancienne?” (audio files 18a, b & c: fragment 18c being closer than 18b to speech and to the vocal version of the phrase) with the response (also a question) from the second – in a dark, round, full colour with an undertone of anger: “Pourquoi voulez-vous donc qu'il m'en souvienne?” (audio 19a & 19b). In a sense, I was translating the meaning of the poem into sound, ranging from darker to brighter tones, from delicate and soft colours to big and full sounds. In this process I was literally (imaging) speaking or singing the vowels or actually the poem itself, experimenting in detail with the (musical translation of the) poetry and art song. Transferring the latter into song transcription was a real challenge for me, that resulted in many versions before the final one (that also still could change over time of course).

Moreover, the question I asked myself most of the time during transcription was about the repeating notes: “How to tackle these repetitive phrases full of triplets? Should I recompose them or keep them all the same?” – which led me to some sort of impasse. If I play them on English horn the same way as it would be sung, it might sound just like tones due to missing words and therefore meaning. After all, having tried out several versions, I chose to stay true to the words and the rhythm from Debussy, only connecting repeated notes (more) if they form one word, for example ending on a soft, weak 'schwa' [ə]. I set myself the challenge to play the repeated notes convincingly, by making the difference in articulation (diction) evident, nuanced and subtle and especially very varied, just like in spoken French. This part is the choice of the interpreter, since I mostly indicated the articulation and prosody by tenuti and legato slurs. Concerning the typical triplets and tenuti from Debussy, that were also remarkable in Pelléas et Mélisande – its English horn part appearing to be very similar in style to the transcription I have written – in the end I went for a logical musical notation. An interesting example in this case would be the one of “Tels ils marchaient dans les avoines folles” and then especially the word and triplet 'avoines' which has the emphasis on the [vwa], the second syllable (and second eighth note of the triplet) which is not on the beat, what made it difficult to logically notate and play on the English horn in a meaningful way (audio 23a and b), so I decided to keep the difference between the notes of the triplet subtle and organic. 


In my opinion, it really depends on the desired pitch and fingerings what would come with ease (in playing this mélodie) on the English horn; either a very prepared subtle tonguing to start a (often lower) tone or a tone that could actually start from nowhere, just beginning with airspeed. Sometimes, a phrase just starts on a more difficult tone on the English horn, like the written (or fingering) C-sharp (sounding a fifth lower: F-sharp) or G-sharp, which is especially a challenge in fragile phrases that need a warm, round, deep and soft tone and not an overly bright (basically sharp) tone – a huge difference with the singing voice, so it is a task to provide a similar tone quality and roundness on these notes on English horn in comparison to the voice. Nonetheless, I consider it to be essential and relevant for any artistic process, performance and approach to this mélodie to pay particular attention to musical aspects as sense of line, direction, phrasing, legato, breathing, colours, atmosphere and intention meanwhile taking into account poetical aspects as prosody, inflection, diction of the French language and the melody from the poem as well as the one from Debussy, following his music with the poem in mind and finding a suitable interpretation that resonates with the person and their instrument.


Finally, I would like to say that an extremely important challenge from my point of view, also regarding notation – being aware that a vocal part lacks any musical directions as dynamics, slurs, groupings (all notes having single flags and sticks) and articulations, often getting the information from the piano score – was to provide the information and explanation required to play the English horn transcription well, without necessarily knowing the poem nor the French language. A musical aspect I met all the time while transcribing, that challenged me primarily, is the idea of ties (connecting notes of the same pitch and in my transcription what should be one French word, like 'bonheur' and 'espoir') and slurs – of which at least two exist, namely legato slurs and phrasing slurs. These slurs connect longer lines, in the case of Colloque sentimental not only poetical (verse) but also musical and literary phrases (whole sentences), which I all tried to specify in the transcription, not every musical sentence corresponding with a textual sentence.

Just as essential as the slurs are the punctuation and breath marks that I took from the poem, the indications by Bernac such as the ticks (breathings), ties ('liaison') and breaks ('coupes': no 'liaison') that I adopted in the score (see the images with poem annotations in Analysis and annotation). For example, I have notated a tick, that Bernac proposed, in bar 44  in which bar I also added a sixteenth rest in the English horn part for timing purposes, because of the [l] in 'ciel'   and one in bar 47. Those small indications, such as commas and ticks, need to be respected as well as the smaller ties, indicating syllables that belong to each other in terms of words, meaning and 'liaison' (enchaining a final consonant and an initial vowel). In bars 24-26, small ties connect 'pourquoi', 'voulez-vous' and 'souvienne', and in bars 51-52 the 'liaison' with a [z] links the article 'les' with 'avoines', notated through ties in the English horn part, in order to avoid incorrect combinations of words (accidental 'liaison' with words that should definitely be separated in terms of meaning), unintentional accents or emphasis on an less important word or syllable. As a result, I think that the process of understanding the literary and vocal lines helps enormously in playing and phrasing more horizontally and legato with English horn. For me personally, I feel that during this process of singing and playing mélodies I have made enormous progress on both instruments voice and English horn – in terms of sound, phrasing, (breath)flow, confidence and especially in the (practise and performance) methods that I developed thanks to my research. Besides, I suppose that having the poetry, French language and vocal ideal in mind creates at least a richer, more nuanced and interesting articulation and a more colourful, characteristic and communicative interpretation of playing art song instrumentally – bringing forward subtleties and details that otherwise could be easily overlooked.


Transcription and reflection

The Alto voice/vocal part is only added for explanatory purposes (not included in the final transcription) and the English horn has been notated a perfect fifth higher than it sounds.

Bars 29, 34 & 44 of the transcription of Colloque sentimental for cor anglais.

15. “Dans le vieux parc solitaire et glacé

deux formes ont tout à l'heure passé.”

*15a: voice; 15b: English horn.

23. “Tels ils marchaient dans les avoines folles,

et la nuit seule entendit leurs paroles.”

*23a: voice; 23b: English horn.

18. “Te souvient-il de notre extase ancienne?”

*18a: voice; 18b & 18c: English horn.

19. “Pourquoi voulez-vous donc qu'il m'en souvienne?”

*19a: voice; 19b: English horn.

Bars 51 & 52 show the typical triplets, following the rhythm and prosody (arsis and thesis) of the spoken French. In 'avoines' the word-prosody first goes up (the voice raises) with the [vwa] (arsis) and then down (the voice is lowered) with the [nə] (thesis), which I tried to put into the transcription (and play) by 'hairpins' and accents – also in the case of 'folles' (the first syllable accented and the second diminished).

21. “—Ah! Les beaux jours de bonheur indicible où nous joignions nos bouches! —C'est possible.

*21a: voice; 21b: English horn.

Bars 35 & 36 are connected by the noun 'bonheur' that has been slurred because it is one word. 

Click media  to open

Bar 47 on the left and bars 22-26 below: