Introduction to the method of recreating recordings


Reflections on imitation


I remember that during my classical studies, my background of playing by ear was regarded as an unusual, but positive asset, but never as a natural part of learning classical music alongside studying scores. It is an interesting paradox that although classical performers of the day strive to connect with the true intentions of the composer, imitating recordings are in relation to this often considered to be a dangerous endeavor. Why is this? Is it perhaps because these recordings challenge our view on authenticity? Or are we afraid of plagiarism? Does the act of imitating a performance contradict the notion that a performance should come from the artist’s inner voice? In jazz music, meticulously accurate imitation of recordings is considered a natural part of learning - even when the ultimate objective is to find one’s own inner voice. Some art students spend countless hours at galleries copying old masterpieces – not because they are trying to become copyists, but because they are trying to learn the techniques and craftsmanship necessary to develop their own artistry. It is in my view naive to think that classical musicians in the modern world of anytime-anywhere access to music can somehow shut themselves off from recorded influences and thus remain “faithful” to the score. This research project aims to demonstrate how extreme imitation of historical recordings could result in artistic creativity and development. 


The method of recreating recordings in the context of my research project should not be confused with the term “learning by ear,” however, because this term is in classical music is often seen in juxtaposition to learning music by reading a score. It is interesting to note that classical music in my view seems to be the only direction in music where this distinction between learning by ear or from the score is emphasized. In other fields of music, such as folk music, jazz, rock, pop etc., it would seem that the two practices have a more natural relationship. The Suzuki Method is a good example of this distinction between the two practices, where children are encouraged from a very early age to learn by ear/imitation before learning how to read music.[1] Some might argue that learning to play by ear is more essential in improvised music such as jazz, but let us not forget that improvisation was an essential part of classical music pedagogy up until the end of the romantic era. In my use of the recreating method, I rely on first learning to play the music by studying the scores. In fact, I avoid listening too much to the recordings at this early stage because I want to be able to make a clear distinction between how I read and interpret the scores and what I experience through the recordings. In light of this, one could argue that my approach of recreating recordings is carried out by playing both from the score and by ear. Having some years of teaching experience, with beginner to advanced level piano students, I am aware of the danger of poorly imitating recordings that lack a necessary analytic understanding of the piece. This is an important point, because I believe that when such imitations become a substitute for score studies and analysis, it leads to performances that are neither true to the score nor true to a recording - and this I believe, is the reason that classical teachers often tell their students to avoid listening to recordings. I think it is important that anyone who engages in the recreating process has a clear idea of why they are doing this and what they want to achieve. The method does not need to be restricted to the study and embodiment of historical performance traditions,[2] but if that is the case, which it is in this project, it does require the performer to have some insight in these traditions if they are going to be able to critically reflect on what they are doing. 



Relatable research projects


Although the method of recreating recordings is becoming more widely used in artistic research, it is important to make a clear distinction between what violist-researcher Emlyn Stam describes as the “pick and choose-approach”[3] and the “all-in”-approach, or “all-or-nothing”-approach as described by pianist-researcher Anna Scott[4]. Emlyn Stam illustrates the problem of the “pick and choose”-approach in his dissertation:


Attempts to explore early-recorded performance style have been few and far between and are frequently limited by performers’ and researchers’ need to demonstrate their professionalism and skill in the context of current mainstream practices. Many have, as a result, taken what I call a 'pick and choose' approach, stopping well short of fully embracing the parameters evidenced by historical recordings.[5]


In relation to this research project, the “pick and choose”-approach is also problematic, because the researchers and performers who engage in it allow themselves to be influenced by historical recordings only to a certain degree, or attempt to imitate only selected parts of a recorded performance. As a result of this, these researchers/performers might willingly or unwillingly (unconsciously) influence the recordings with their own artistic practices. Thus, the “pick-and-choose” researchers/performers are able to embody certain aspects of a historical recording that they like, but also leave out those aspects that they might disapprove of. This conscious or unconscious “censorship” of historical performance practices, where the most problematic and currently disapproved-of tendencies are left out of the recreated performance, is in my view a problematic approach if the aim is to thoroughly investigate historical performance traditions. Although it is not possible to conduct a “true copy” in the most literal sense of the word, the “all-in”-approach is however a very different process, because here the researcher faithfully commits into imitating all aspects of a historical recording – also the aspects the researcher might disapprove of. There is to my knowledge only three completed doctoral level projects where the method of “all-in”-recreating historical recordings were at the core of the research. The researchers behind these projects are Norwegian pianist Sigurd Slåttebrekk[6], Canadian pianist Anna Scott[7] and Canadian-Dutch violist Emlyn Stam[8]. My project is closely related to these and it is therefore necessary to compare the slightly different recreating approaches here in order to clarify my own method/aims.


In addition to committing to the “all-in”-approach, the researchers Slåttebrekk, Scott and Stam have all undertaken research focused around the romantic performance tradition. They also have in common that they, in addition to this, conducted experimental recordings where they applied the gained knowledge to recording repertoires that had not been historically recorded by the source.[9] Despite these similarities, these researchers have slightly different aims and approaches.


In Sigurd Slåttebrekk’s research, the aim was first and foremost to embody the performance style of Edvard Grieg by recreating the recordings where Grieg played his own pieces in Paris 1903. The artistic results became recreated recordings made by using modern recording equipment on Grieg’s own piano at Troldhaugen, Norway[10]. The research was conducted through a close collaboration between Slåttebrekk and collaborative researcher and producer Toni Harrison. The original recordings have significant amounts of surface noise as well as unstable pitch, and an important agenda of Slåttebrekk/Harrison was therefore to make up-to-date “clean” recordings where the playing style of Grieg could be available to modern-day listeners. This was realized through a series of digital restorations before the artistic recreating process could begin.[11] Following this, Sigurd Slåttebrekk firstly undertook periods of extensive listening, trying to internalize the Grieg recordings before recording them himself. During the recreating process, he would then work together with Tony Harrison who had a recording setup station by the piano. They would start by listening to short fragments of each original recording, then Slåttebrekk would attempt to imitate what he heard on the piano (recording the same fragment). This process would be repeated a vast amount of times before going on to the next fragment (Tony Harrison recalled something around the area of a 50:1 ratio between the recorded material and the fragments that ended up in the final edit[12]). During this work, they made an interesting discovery: they found that although the final recreated recordings were technically quite accurate with regards to the timing of the original recording, they lacked some of the musical “sweeping” character of Grieg. Slåttebrekk and Harrison found that they, rather than working in small fragments, should focus on longer takes to solve this. In addition to recreating the original recordings of Grieg from 1903, Slåttebrekk also made two extra recordings. The first of these was the Sonata in E minor by Grieg which acted as a sort of “meeting point” in the research.[13] Grieg only recorded the complete third movement as well as most of the Finale (the development section and reprise is missing in the original recording). Slåttebrekk recorded the third movement as a recreated copy, but then for the fourth movement, he first recreated the parts that are in the original recording and then “filled in” the missing parts. He also recorded the first and second movement (which were not recorded by Grieg). The final artistic experiment was his recording of the Ballade in G minor by Grieg. Here he had to rely solely on the reference from the other recording since no fragments whatsoever exist of Grieg playing this himself. Having finished the recreated recordings, Slåttebrekk found himself contemplating the profound effect that the research had had on him:


In retrospect, I certainly find myself somewhat surprised by the way my own performance has become so fundamentally affected by this work. Not to say that the influence from Grieg came as a surprise to me, being the sole purpose of this project. I did however expect it to be more of a purely intellectual journey into the style of these historic performances, taking on various strategies and principles, still retaining a certain distance, allowing myself to move convincingly between styles for demonstration and lecture purposes. The reality, however, is that the painstaking work on these recreations has affected my musical instinct, and rewinding now seems very difficult. [14]


Anna Scott applied a similar approach in her research with regards to the recreating process. However, her agenda was not first and foremost to replicate historical recordings for the sake of restoration. By recreating the recordings made by pianists Adelina de Lara and Ilona Eibenschütz, who were both from Johannes Brahms’s inner circle (former pupils of Clara Schumann who had occasionally also been taught by Brahms himself), Scott developed a “Brahmsian” performance style. In addition to recreating recordings, Scott like Slåttebrekk also used her experience to make what she called “experimental extrapolation” recordings of Brahms pieces – challenging the modern-day notion of “Brahmsian identity” and how his music might sound when played differently. In contrast to Sigurd Slåttebrekk, Anna Scott committed to recording full performances (“one-takes”). Although the ground work leading up to these recordings was no less significant than Slåttebrekk’s, the decision to make unedited one-takes raises some important questions that are relevant to what I chose to do in this my research: on one hand, it could be argued that Slåttebrekk’s approach resulted in recordings that are more “spot-on” accurate compared to the originals[15]; on the other, if the agenda is to thoroughly understand the general tendencies and gestures of the historical performance, it might be beneficial to record them in the same manner that they were originally played – uncut and in one-takes. The benefits of working with longer fragments and even full-takes was also something Slåttebrekk and Harrison eventually came to terms with.[16]

Where Slåttebrekk came to the conclusion that he had to record on Grieg’s personal historic Steinway[17], Scott focused more on how the recreating process made her mentally and physically approach the piano on a more fundamental level. She emphasized that her results were not restricted to the use of period instruments: 

Indeed, our access to all kinds of historical tools still has not produced Brahms performances that capture the beguiling spirit captured on the recordings of those pianists who knew him, despite pianists' continued belief in both their historical awareness and creative agency.[18]


Violist Emlyn Stam chose a “full-take” approach similar to Anna Scott’s, although balancing the scale between full performances and “sparsely” edited recordings (aiming for 4-5 minute stretches in longer pieces).[19] Stam discussed how the “full-take” way of approaching the recreated pieces did come at the expense of some details that were lost: 

Generally, I have had to make trade-offs between capturing the overall sweep of the originals in a live performance or complete take and adhering to accuracy in the copying of details.[20]


Stam also investigated aspects of historical recording technology, recording with and comparing low-fidelity and high-fidelity microphones in collaboration with producer Geoffery Miles.[21] The intent of these “mock-up” historical recordings was not to produce authentic duplicates in a technological perspective, but rather it was a creative tool that helped Stam to delve deeper in the process of decoding the romantic performance style of performers such as Lionel Tertis and Oskar Nedbal. 

Unlike Sigurd Slåttebrekk and Anna Scott, Emlyn Stam also included chamber music works as a fundamental part of his research. While this certainly adds a new dimension to the research, it does come with its challenges, as Stam puts it: 

While pianist-researchers such as Scott and Slåttebrekk had only themselves to focus on during the copying process, I needed to focus on my own copying as well as that of my colleagues, all while relating their copying to the musical material I was playing.[22]


Playing chamber music allowed Stam to work on aspects such as “dislocation” (non-togetherness) between the musical parts of each musician, something which involved having the musicians “perform in ways that they sometimes felt to be counterintuitive or aesthetically displeasing.”[23]

Annotated scores were made and used during the recreating process in all three aforementioned research projects. It should be said that using such annotated scores as a way of documenting the differences between modern day and historical performance practice is problematic because then one must remain very clear about which tendencies/practices relate to a general performance practice and which are very personal statements or characteristics. However, using the annotated scores to guide the listener through the deviations between the recordings and the original scores can be very helpful. The annotated scores serve to this purpose in the research of Slåttebrekk, Scott and Stam (and indeed in my own).

Another factor these three research projects share in common is how they approached the recreating method first and foremost by alternating between first listening to the recordings, then imitating them, and finally to repeating the process as many times as they deemed necessary. Although playing alongside the original recording (while having the playback in-ear) was occasionally done in the process, it was not systematically incorporated in the recreating method. Common for Slåttebrekk, Scott and Stam is also that their research results were confined to the romantic performance tradition style. They did not further investigate if the embodied knowledge could be used in other kind of music.

In addition to this, researchers Anna Scott and Emlyn Stam both used the software Sonic Visualizer as a part of making detailed annotated scores when analyzing the historical recordings.


My agenda


While I acknowledge the potential that knowledge about Ricardo Viñes’s playing style could have in the field of historically informed performance (HIP), something which I to an extent investigated (particularly in the context of the recorded Debussy pieces) my main agenda for going into this research project was the artistic development work that could come out of it: what happens to me, the performer, when I embody a historical performance style? How does this work of recreating recordings affect my artistic practice? And in what ways are the research results applicable to new interpretations, not restricted to the specific historically recorded pieces?


As discussed earlier, the constant surface noise on the historical recordings of Ricardo Viñes are not, in my view, substantial enough to hinder the recreating process. The same goes for the recordings of the contemporaneous recordings I selected to recreate (made by Sergei Rachmaninov, Ignacy Friedman and Jesús María Sanromá). In fact, when listening to the recordings on a deep level during the process of recreating, the noise didn’t really bother me at all – it was one of the first layers that I managed to subconsciously peel away. In light of this, it was never my intent that the recreating process should be undertaken for the same conservational purposes for which the recreated Grieg recordings were partly made. If, however, the historical performances of Viñes should become more accessible to a broader audience because of my recreated recordings, this would be a positive contribution to Viñes’s legacy, and something that would please me very much.

With regards to the documentation of my recreated recordings, I decided from the beginning that this would take form as a collection of videos. These videos would be the ground work leading up to the artistic results. However, I have during the research period come to terms with the reality of this being a bit more complicated. As I shall later explain, as part of this video documentation I made “cross-cut” videos where my synchronized performances are superimposed onto the originals and where the videos alternate between my playing and the originals as a sort of musical dialogue. These performances are in my view neither merely documentation, nor are they limited to being copies. Rather they could be considered as artistic works in themselves – musical dialogues between the past and the present. 


Conversations with the dead


My approach to the method of recreating recordings is similar to the aforementioned researchers Slåttebrekk, Scott and Stam in several ways, but there are some differences. The most important is my systematic play-along” approach, which I consider to be a key factor in the embodiment phase. This approach is carried out by performing the imitated recording while having the original playback in-ear, striving for synchronization between the recreated performance and the recording. Being able to do this required substantial preparation work and I had to know the pieces very confidently beforehand to avoid being thrown off” during the “play-along” performances. This, I believe, is relatable to chamber music, where one needs to have a certain degree of automatization in one’s own musical execution to be able to listen and react to the other musicians on a deep level. For me, the experiences from playing-along led to a paradoxical discovery, related to my previous experiences with regards to chamber music works.[24] One might think that playing together makes careful listening more difficult, because one has to focus on hearing both the original recording and one’s own playing. My experience, though, is that when I practiced my play-along” approach over a period of time, I started hearing and experiencing the historical recordings on a deeper level, and as a result of this my recreated performances were also progressing more rapidly. 

It should be mentioned that how I used the relatively uncomplicated and sparse technical equipment was an important factor in the recreating process. At the beginning of the play-along period I would use conventional ear plugs or headphones. This partly muted my own playing, meaning that the original recording was superior in the mix”. This allowed me to firstly focus on the more general gestures of the historical recordings, such as tempo, rubato etc. Following this, I started focusing on more intricate details, such as “dislocation”, arpeggiation, articulation, the use of pedals etc. This phase required a fair amount of technical experimentation, and here I would mostly listen carefully to the recording first, and then imitate what I heard without the in-ear playback. Experimenting with different techniques, pedaling effects, articulation etc., striving for a sound and expression that would as closely as possible imitate what I heard in the recordings, enabled me to learn much about the Viñes performance style, including issues related to his technique and pedaling, as I shall later explain. As I gradually progressed and became able to recreate the more subtle details of the performance, I shifted over to using my “Aftershokz Bone conduction headset.”[25] This headset basically works like a conventional one, except that the sound is projected into the cranium right next to the ear, rather than directly into it. The sensation for the listener is in my view practically the same as using conventional ear-plugs, except that your ears (and thus your hearing) are not impaired by the in-ear music. Because the sound is not projected directly into the ear, I could hear my own playing clearly, with the appropriate level of volume, and it was possible to adjust the mix between the original recording and my live piano playing. I used the Aftershokz headset in all my recreated videos. I also experimented with the original recording being played back on conventional stereo speakers in the room, and although I believe that one could get similar results from this approach, I preferred using headphones/headset. The Aftershokz headset is affordable, and the only other technology I used was my smartphone to play and stop the recordings. Considering my rather sparse use of technical equipment, I believe that my approach on the recreating method could be accessible to any musician without the aid of a technician or complicated technical equipment.

The more I played together with Ricardo Viñes, the deeper I delved into his recordings. I experienced that I was now reacting to what I heard on a deeper level – it felt as if I had peeled off layers and I became aware of very subtle details previously lost to me. I came to the conclusion that this has to be related to my chamber music instincts. 

Although we tend to strive for absolute silence when we want to carefully concentrate on analytical listening, I don’t think we should underestimate the perception happening on a deep level when we as performers interact with other musicians. As I continued to play alongside Ricardo Viñes, strange things started to occur. It was as if the previously frozen-in-time “dead” recording started to come to life. In the span of the several weeks I worked on each piece, it was as if the original recording started organically changing its interpretation. On a few occasions, I actually had to check if something had happened to the playback, because I felt that the tempo had somehow changed. This was obviously my own mind playing tricks on me – and it was surely my playing and my experience of the recordings that were evolving – not the other way around. Nevertheless, it felt as some sort of a conversation. I found this intellectual versus artistic struggle that was going on in my mind to be very intriguing. The process allowed me at times to feel that even if I didn’t become the very being of Viñes, then I was certainly in his room playing alongside him. 

When the ear is exposed to two tracks of music being played simultaneously, but slightly desynchronized, it is in my view incredibly perceptive. I experienced this not just in the recreating process, but also when I was editing my videos - trying to superimpose the recreated soundtrack over the original recording-track to achieve perfect synchronization.[26] Even when zooming in to the degree of 1/30 of a second and looking at the visualization of the soundtrack layers in the editing software[27], the ear is far superior to the eye when adjusting them in order to achieve the synchronization. Once the tracks were in synch, they sort of “melted” together, and it became difficult to distinguish one from the other. Because of the slight variations in pitch between the historical recording and my recreations, the now double-layered recording sounded a bit “honky-tonk”, but the timing of the two recordings did indeed sound like oneThis experience of hearing the recordings “melting” together, was a sensation I strived for when I played my imitated performances with the original recordings in-ear. This “melting together” therefore became the benchmark of the recreating process. There would be moments of “free-flow” during my work where I started to unconsciously question who was leading this “musical dialogue” – was it me or Viñes? This probably sounds a bit absurd, much like this feeling of the original recording becoming “alive”. These are sensations that do not make much sense in an intellectual context, but they are in my view of fundamental value in the artistic process of embodying a performance. I believe that these sensations started happening because the imitation had become so advanced, and the physical movements so automated, that my mind at brief moments of time became slightly confused about whether it was reacting to the original recording - or the recreated one. Although these paradoxical experiences happened for only very brief moments of time, it was when they were starting to occur, far into the recreating process, that I knew I was ready to record my recreated performances. One could regard this as my “inner voice producer”, instructing me on when the time was right to document my results. 

It should be mentioned that I did not use software like the Sonic Visualizer as part of my recreating process. I am not disregarding the benefits that could come from using such software, and I do acknowledge that it is a great tool for visual analyses between historical and modern day performances. I am sure that advanced software is capable of deciphering these old recordings on a detailed level that far exceeds human capabilities, but such programs could only in my view, at least with the technology currently available, display the gestures and mannerisms objectively. The ability to understand the intent of a performance is something that I argue must be experienced to fully grasp. 


The recording process


When I felt ready to record my imitated copies, I would at this point be able to play through the entire piece alongside the original recording without major “hiccups”. I did, however, experience that these attempts of run-throughs came at the expense of some details that were lost in the process. With regards to the different approaches in the “full-takes” vs. edited “short-takes” in the research of Slåttebrekk, Scott and Stam, and the pros/cons of them, I decided to do both. The more fragmented approach of recording in smaller takes and aiming for synchronization was used on all my 11 recreated videos. Following this, I made a film recording at Ringve Museum in Trondheim, produced and filmed by Valkyrien Productions, approximating a “historical recording environment”, and here I chose to record “one-takes” of complete performances in the same manner as Viñes did in his day. By comparing both approaches I would ultimately be able to reflect on my own view of the differences and their values/weaknesses. The following text concerns my work on the 11 recreated videos, which were recorded in shorter takes.

Deciding on the length of each part of a piece to be recorded, depended on the complexity of the work, but generally I recorded parts that had an approximate duration between twenty seconds and up to two minutes.

In contrast to the research of Slåttebrekk, Scott and Stam, I integrated my “play-along”-approach also in the recording process. My bone conduction headset allowed me to have a “mix” of the original recording in my ears which was not picked up by the microphones recording my recreated performance.[28] I would record in sections and make several takes before moving on to the next one. Once more, it was the sensation of “melting together” with the original recording that was the benchmark I was striving towards, and I was excruciatingly thorough. When I finally had a “take” I was pleased with, I would stop playing and listen to the playback of my recording from a separate set of headphones plugged into the camera that was recording the sound.[29] Being quite pleased with my efforts before listening to this playback, I now made a disappointing, but fundamental discovery: much like Slåttebrekk had experienced, I felt that my recreated recording bits were missing something. I knew they were more or less perfectly in-sync (at least to the best of my abilities), but the recorded parts lacked the musical direction and “sweeping character” of the original recordings. The solution to this problem became the most demanding part of the entire recreating process of this project. I now had to record the same part once more, still striving to play in-synch, but at the same time attempting to evoke the musical direction of the original recording. This became counter-effective to the imitation process, because trying to follow the longer lines and direction in the music meant that I would have to exaggerate my musical gestures – something that slightly disturbed my attention to “perfect” synchronization. At this point I could have decided on recording in longer fragments without the in-ear playback, and the result would probably have been something that closely resembled the musical intention of the originals, allowing me to focus on the general musicality of the recording, rather than obsessing about the synchronization. However, this would come at the expense of some details lost. Having already spent so much time in the recreating process, focusing on these smaller details which I considered to have a fundamental value, I stubbornly proceeded and eventually succeeded in recording shorter fragments that I felt contained both the synchronization as well as the more general musical direction. I must acknowledge, though, that there will always be some level of compromise, but then usually in favor of the general musical intent. The results of this were, if not perfect copies”, then at least similar enough to superimpose them into the originals and intersect (“cross-cut”) between the two, as demonstrated in the videos of all my recreated recordings. 



The videos


I must admit that the way I chose to make my “cross-cut” videos was inspired from the beginning of the Recreating Grieg – an introduction video by Slåttebrekk and Harrison.[30] In their research, however, it was used as primarily an effective audiovisual demonstration only in the Weddingday at Troldhaugen performance, as a sort of promotional video that effectively demonstrated what they had been up to. I developed this idea into what became both a creative tool serving as a standard of my recreations, as well as artistic results in-and-of themselves. 

In making my videos, the first step was to synchronize all the recorded takes that would make up the complete performance together with the corresponding parts from the original recordings. Adobe Premiere Pro has a powerful auto-synchronization tool that is greatly time effecient when for example synchronizing the sound from the external microphones with the different cameras, which I needed to do with my own recorded sound.[31] However, this tool was not powerful enough to auto-synchronize my recordings with the original Viñes performances. The reason for this is probably a combination of the surface noise on the originals disturbing the analyzation process, as well as the fact that although the recreated recordings appear to be in synchronization, they are obviously not mathematically “correct” in that sense. The solution was to do this manually (the “old way”), by superimposing my recreated recording sound files over the historical recording track and moving them “frame by frame” (1/30 of a second) until the sound “melted” together in the same way I had experienced when I recorded them. Again, the ear is extremely perceptive and will easily hear if there is but the slightest de-synchronization between the two recordings. Because I had been so meticulously thorough about this when I recorded, it was, although time consuming, simply a matter of repeating the principle in the editing process. Having done this, the parts which make up the entire performance were edited together with the original recording (now synchronized) placed below in the recording timeline. It is important to state that I did not edit the original historical recording in any way other than adjusting the volume when the original “fades” into the recreated performance. In principle, editing the historical recording is quite possible and this could have been used to correct mistakes made in the recreated performances. This would obviously manipulate the historical performance and with regards to my research project, it would have been a highly un-ethical approach. For the sake of transparency, I have included three versions of each video in this reflection: a) the video where I am seen playing, but with the original recording being heard in the background (in black-and-white); b) the video of my full recreated performance (in color); c) the “cross-cut”-version which alternates between video “a” and “b”. Although I made recreations to the best of my abilities at the time and was pleased with the results, the human imperfections are there if one truly wishes to find them, with warts and all. The videos have 2-3 alternating camera views. Because I am not a professional video editor, the choices of camera angels, animated zoom-ins and outs etc., were simply based on my own taste and artistic choices. 

Because I regard my “cross-cut”-videos in this research as artistic results, a critique of them could be that making choices about when I played, and when the performer from the original recording played, is perhaps an example of the “pick-and-choose” approach, where I selected only the bits of my playing that I was happy with. Trying to be transparent in this matter, I have included my complete performance without the “cross-cut”-effects (video “b”). In addition to this, the full-performance “one-take”-recordings that make up the film from Ringve Museum are un-edited.


[1] The Suzuki method is an internationally known music curriculum and teaching philosophy dating from the mid-20th century, created by Japanese violinist and pedagogue Shinichi Suzuki (1898-1998).

[2] In the introduction to this research project I write about my different uses of the method in the past.

[3] Stam 2019, 11

[4] Scott 2014, 183

[5] Stam 2019, 11

[6] The notes must embrace the bars, not the bars the notes. (Slåttebrekk 2013)

[7] Romanticizing Brahms: Early Recordings and the Reconstruction of Brahmsian Identity. (Scott 2014) 

[8] In Search of a Lost Language: Performing in Early-Recorded Style in Viola and String Quartet Repertoires. (Stam 2019) 

[9] Anna Scott refers to this as her “experimental extrapolations” (Scott 2014, xvii).

[10] The villa of Edvard Grieg close to Bergen, Norway, now a museum.

[11]  The Grieg 1903 Recordings, accessed on October 30, 2021,

[12] Video: “Recreating Grieg – an introduction”, accessed on October 30, 2021,

[13] Ibid.

[14] Sigurd Slåttebrekk – A personal view, accessed on October 30, 2021,

[15] As demonstrated in the beginning of the video: Recreating Grieg – an introduction, accessed on  October 30, 2021,

[16] Prelude and trouble at Troldhaugen, accessed on October 31, 2021,

[17] Ibid.

[18] Scott 2014, xx

[19] Stam 2019, 177

[20] Ibid.

[21] Music producer and recording engineer Geoffery Miles also worked with Sigurd Slåttebrekk towards the end of his Grieg project due to the declining health of Tony Harrison.

[22] Stam 2019, 179

[23] Ibid.

[24] Discussed in the introduction of this research.

[25] Bone conduction headphones are widely used by individuals suffering from hearing loss, during scuba diving, and as an essential form of military communications. The model I used is mainly advertised for runners and is not affordable compared to conventional ear-plugs.   

[26] This absolute synchronization was necessary in order to achieve the “cross-cut” effect between the original recording and my recreated performance.

[27] I used the video editing software of Adobe Premiere Pro for both video and sound editing.

[28] I can be seen wearing this headset in all of my recreated “cross-cut” videos.

[29] The camera recording the sound was a Zoom Q8. I used the built-in stereo-microphones as well as two external Aston Starlight Stereo Pair plugged into the XRL-inputs of the camera.

[30] Recreating Grieg – an introduction, accessed on October 31, 2021,

[31] I recorded my videos with 2-3 static cameras (Sony cameras), a Zoom Q8 recorder and a pair of Aston Starlight microphones. The setup was standard, but some adjustments were made according to the acoustics and the light of the different rooms I recorded in.


Previous                                                                                                                                                                              Next