Early twentieth-century recordings in higher music education: a preliminary analysis of the students’ views

Abstract

This article examines how classical music students understand early twentieth-century recordings in higher music education. A qualitative research method was chosen to investigate the beliefs and self-reported practices of 16 students enrolled in a European conservatoire, whose attitudes were considered through the administration of a semi-structured questionnaire and an interview. Their responses identified seven main themes: (1) beliefs, (2) sources, (3) self-reported practices, (4) repertoires/performers, (5) educational implications, (6) limits and (7) benefits.

Results show how much students value early twentieth-century recordings – especially when it comes to analysing the performance practices of the past and developing new interpretations – and how articulate their responses can be with regard to specific stylistic and technical issues. However, possibly due to informal learning strategies and the lack of curricular teaching activities focusing on listening to and analysing recorded interpretations, some responses highlighted a misrepresentation of our recent musical past and the need for a more structured curricular activity. This last should benefit from a vast body of scholarly literature whose relevance is still underestimated among music practitioners.

Listening to early twentieth-century recordings: still a controversial issue

Listening to recordings can be encountered in many informal music learning settings, and advanced musicians often admit to listening to others’ recordings (Elverdam & Brock-Nannestad, 2008; Hallam, 1995; Smart & Green, 2017). Recent investigations confirm the role and importance of listening to others’ recorded performances as a strategy often used to learn a new piece, choose and become familiar with new repertoires, and gain a reference point for comparison (Antonini Philippe et al., 2020; Volioti & Williamon, 2017, 2021). Studies on practising strategies also touch on the use of recordings in different stages of the preparation process and for different purposes, e.g. developing an interpretation, analysing the music and evaluating learning strategies (Chaffin & Imreh, 2001, 2002). Moreover, today’s musicians have countless opportunities to listen to a variety of performers’ interpretations, thanks to the vast availability of past and contemporary recordings in different formats. This becomes all the more relevant if one considers that the training of musicians often involves the notion of imitation, and modelling is a learning strategy common among music practitioners (Lindström et al., 2003). This is the case even when the plea for originality makes it musically suspect to take reference from others’ interpretations (Lisboa et al., 2005).

While CDs are sometimes perceived as relics of an older age, YouTube, podcasts and streamed playlists assume a pivotal role in musicians’ listening habits, thus representing an invaluable learning tool for both music practitioners and scholars (Monkhouse & Forbes, 2015; Volioti & Williamon, 2017). In particular, the use of YouTube in music education settings may help students to develop better critical awareness and judgement and to understand how knowledge of history may inform their performance practice (Monkhouse & Forbes, 2015). YouTube is often used as a video repository to assist teaching and learning in the performing arts (DeWitt et al., 2013), even though the educational value of beginning instrumental music tutorials available via YouTube may depend on many factors, such as technical adequacy, musical quality and the credibility of those who produce the videos (Hanson, 2018). In addition, online tools such as blogs and forums could have complementary functions in scaffolding students’ self-regulated strategies and learning, thus supporting the development of a virtual community of practice (Biasutti, 2017).

However, if the ubiquity of YouTube has confirmed its potential as a teaching and learning tool in different educational settings attracting increasing scholarly attention (Hanson, 2018), the same cannot be said of the manner in which recordings available via YouTube, and early recordings, in particular, may represent a valuable teaching and learning resource for training professional musicians. In the existing literature, there is a dearth of approaches considering the pedagogical implications of using early recordings in educational contexts. This becomes even more surprising if one considers that since Robert Philip and Timothy Day’s ground-breaking volumes made their first appearance (Day, 2000; Philip, 1992), much has been written on early recordings (i.e., recordings realized in the first decades of the twentieth century) and their importance in documenting the connection between a silent past and a multifaceted musical present (Leech-Wilkinson, 2009, 2010a). In line with Robert Philip’s formulation, we take early recordings to mean recordings of the Western Art music repertoire that were made in the early twentieth century and stand, to paraphrase Philip, at the end of the era in which knowledge of practices was preserved in written sources, and at the beginning of that in which this knowledge became accessible through wax rolls and discs (Philip, 1992, p. 2). This concept covers the first three decades of the century and sees the technological leap of 1925, with the advent of electric recording, as the turning point into a new, more advanced phase. Of no less importance is that the artists who recorded at the beginning of the last century had been trained in the previous one, and are to be considered not only the most authoritative witnesses of the late nineteenth-century’s musical aesthetics but also the unsurpassed masters of an entire epoch. The explosion of digital reissues of early twentieth-century recordings and the role played by platforms such as Spotify and YouTube in making them available worldwide have provided undeniable evidence of significant changes in the interpretations of canonical repertoires while raising numerous questions about the legitimacy of many interpretative approaches that we continue to consider appropriate today.

Among others, early recordings have shifted the listening experience from the social to the solitary (Clarke, 2007), have extended the object of musicological studies from score to performance (Cook, 2010) and, most importantly, have opened new perspectives on the complex and multifaceted dimensions involved in the music-making processes (Fabian, 2015). Despite some initial scepticism about quantitative investigation (Bowen, 1996), an entirely new branch of musicological research has emerged, which sees researchers use computer-assisted analysis to examine expressive timing and dynamics in recorded interpretations of the Western Art music repertoire. The piano soon came to hold a prominent position, with Bruno Repp’s seminal studies on Beethoven, Schumann, Debussy and Chopin forming the first invaluable corpus of scholarly literature (Repp, 1990, 1997, 1998, 1999a, 1999b). More work has been done in recent years, with studies often involving the detection and analysis of inter-onset intervals, for which a variety of analytical techniques have been tested. For example, Andrew Earis used a simple manual beat-tapping system and the continuous wavelet transform to correct the beat-tapped times in Johann Sebastian Bach’s keyboard music (Earis, 2007), while Georgia Volioti used principal components analysis as a dimension reduction method for exploring timing profiles and performance strategies in Edvard Grieg’s recordings of two Lyric Pieces Op. 43 (Volioti, 2017). Other studies focused on Chopin’s mazurkas (Cook, 2007, 2009; Dodson, 2009) and preludes (Dodson, 2011a, 2011b; Rink, 2001; Spiro et al., 2010), while descriptive statistics were adopted to analyse the singing style that can be found in early recordings of Schubert’s Lieder (Leech-Wilkinson, 2010b; Timmers, 2007). A large body of research was carried out by Dorottya Fabian on the interpretation of Bach’s music for violin solo by performers like Joseph Joachim, Pablo Serasate, Eugène Ysaÿe, Jascha Heifetz and Nathan Milstein (Fabian, 2003, 2006, 2017; Fabian et al., 2014; Fabian & Ornoy, 2009). The extent of expressive tempo modifications as a function of textual and musical content was investigated by Massimo Zicari in the recordings of operatic arias from Adelina Patti, Luisa Tetrazzini, Marcella Sembrich and Nellie Melba (Zicari, 2014, 2017, 2019).

What is most characteristic of this vast body of scholarly literature is the quantitative analysis of tempo-related interpretative devices, used to bring empirical performance analysis into dialogue with other branches of musicological studies and, where possible, to draw evidence-based conclusions on early twentieth-century performance style.

Although the quantitative approach seems to be predominant in the scholarly work addressed above, other authors have demonstrated how fruitfully early recordings can be investigated without the help of computer-assisted analysis. As early as 1983, Will Crutchfield’s pioneering work on ornamentation in Verdi’s operatic arias interrelated early recorded evidence with written documentation of Verdi’s own preferences (Crutchfield, 1983). Verdi’s singing style, voice characteristics, timbre and technique were also the focus of Roger Freitas’ investigation of 2002, in which he considered early recordings from relevant singers in relation to Verdi’s own preferences and opinions (Freitas, 2002). Marco Beghelli and Raffaele Talmelli used early recordings to complement traditional text-based source documents to investigate the development of the contralto voice during the nineteenth century (Beghelli & Talmelli, 2011). Similarly, Peres da Costa investigated early recordings of late romantic pianists to show how radically their performing practices differ from the present (Peres da Costa, 2012). Massimo Zicari’s extensive study on Luisa Tetrazzini’s recorded interpretation of the nineteenth-century Italian operatic repertoire demonstrates how much can be learned by transcribing and comparing multiple renditions of the same arias (Zicari, 2022). Whether conducted using a quantitative or a descriptive approach, the analysis of multiple recorded performances of the same piece has proven particularly effective when investigating inter- and intra-subjective differences among performers (Zhukov, 2015).

Interestingly, as soon as we attempt to draw larger conclusions and call for the idea of a stylistic kinship or a trajectory of historically patterned styles, any attempt to reconstruct how interpretative styles evolve over time within and across individuals becomes problematic (Volioti, 2010). Individual differences tend to prevail over more largely shared stylistic traits and, despite a strong consistency over time, individual performers follow their own specific interpretative trajectories regardless of the models shaped at an early stage of their career (Fabian & Ornoy, 2009). Moreover, the individual liberties observed in early twentieth-century recordings have affected musicians’ approach to canonical composers, thus challenging the normative thinking that often surrounds the interpretation of the Western Art Music repertoire (Fabian, 2015, p. 7). The notion itself of tradition as a homogeneous analytical, interpretative category to be adopted when studying the transmission of performance practices is not supported by the analysis of recorded interpretations; these last present a variety of interpretative approaches that defy the notion of tradition as a homogeneous corpus of shared conventions (Volioti, 2010).

All these studies notwithstanding, and despite that the value of early recordings as a source for the documentation of our musical past has been generally accepted (Peres da Costa, 2012), the study of early twentieth-century recordings (let alone the use of computational and descriptive analysis) for educational purposes still needs to be expounded. If, on the one hand, the ubiquity of YouTube confirms its enormous potential in the educational sphere, and if, on the other hand, a vast body of studies on early recordings fuels critical reflection on the interpretative canons of the past, one might wonder whether this discussion has entered the music conservatories. It is not clear to what extent early twentieth-century recordings are present in educational settings, how they are perceived by music students and what their impact is in the context of higher music education institutions. Do they receive sufficient critical attention among young music practitioners? How do music students learn to use them? If so, how do they come to value these recordings? Do they even value them at all? How does the ubiquity of YouTube and the vast availability of recorded interpretations affect teaching and learning strategies? These questions were addressed in an exploratory study involving 16 students from a European conservatoire, for which a qualitative method was chosen.

Method

Aim of the study

The aim of the current study is to examine how conservatoire students understand and conceptualize early twentieth-century recordings. As already suggested, by early twentieth-century recordings, we understand the recordings that were realized during the first three decades of the last century by interpreters trained in the late nineteenth century. A qualitative research method involving a semi-structured questionnaire and an interview was chosen to investigate the beliefs and self-reported practices of students while using early recordings. These were analysed as indicators of their engagement with early recordings. The overarching objective was to develop themes for reflection on the opportunities offered by these recordings in music conservatories, with the idea that they may enrich the curriculum and innovate the teaching-learning approach. The following research questions drove the investigation:

1. 1. How do instrumental students understand early twentieth-century recordings?

2. 2. How do students use early twentieth-century recordings to develop a musical interpretation?

3. 3. How do students use early twentieth-century recordings as part of their learning strategies?

Participants and context

The study recruited 16 students attending a European conservatoire. Participants were 6 females and 10 males, with a mean age of 23.9 (range 19–30, SD = 3.42). Fourteen participants were attending the Master of Arts in Music Performance, one the Master of Arts in Music Pedagogy and one was enrolled in the Bachelor of Arts in Music. Their main instruments were violin (n = 7), cello (n = 3), viola (n = 2), piano (n = 2) organ (n = 1) and harp (n = 1) (Table 1).

Table 1. Participants

Participants’ study programmes place great emphasis on individual tuition, chamber music and orchestral and ensemble productions, leaving little room for elective activities. In particular, in the MA in Music Performance, elective activities are given 14 credits in total out of 120 over two years. Early, modern and contemporary repertoires are part of their training, while no other curricular activities are offered in connection with recordings, whether early or new. With the exception of one participant, none had previously participated in a learning experience involving early recorded interpretations. Regarding the number of public concerts given in the last five years, participants reported 30 on average (SD = 15.1), suggesting a solid commitment to music performance. To the question ‘Do you have a historically informed approach to the repertoire you perform (yes/no)?’ all master participants answered yes, thus demonstrating a strong interest in the issue, while the Bachelor student answered no. Participation in the study was voluntary and respondents received no payment or financial reward.

The semi-structured questionnaire and interview

The semi-structured questionnaire aimed to elicit students’ beliefs about early twentieth-century recordings and to explore their self-reported practices. The initial section of the questionnaire aimed at collecting information about the participant’s background. The core questions focused on the situations in which students use early twentieth-century recordings in their musical activity, asking them to provide examples of specific situations. The sources through which students have access to early twentieth-century recordings were considered, and aspects such as how helpful they can be in understanding the music, developing an interpretation and improving style and technique were investigated. Additional questions regarded the repertoires and the performers to which historical recordings could be of particular relevance. Participants were then asked whether they had been already involved in educational activities concerning early twentieth-century recordings and whether they had ideas or suggestions for further educational uses of those recordings. Finally, the strengths and the limitations of early recordings were considered. In order to collect qualitatively richer data, three participants were asked to respond verbally to the same questions in the form of a semi-structured interview. This resulted in a more nuanced, richer palette of information. The full list of questions is reported in the appendix.

Ethical statement

The study was conducted in agreement with the standards of the Code of Ethics and Conduct of the British Psychological Society and respecting the Declaration of Helsinki. All participants were informed about the anonymisation of the data, and about each step of the research, including data collection and analysis, reporting, public presentation and publication.

Procedure

The first author recruited participants by circulating an announcement following a collective theory class at a conservatoire. The class focused on early twentieth-century recordings, provided an appropriate definition, described the historical context, discussed technical limitations and included guided listening to a number of selected recorded interpretations. Attendees had various backgrounds and competences, playing different musical instruments, thus ensuring adequate variability within the setting in which data were collected. The semi-structured questionnaire was administered in writing and sent per email; additionally, three participants (P. 14, 15 and 16) were involved in a real-time interview to ensure more richness of data. The interviews, conducted in person, lasted approximately 45 minutes and were transcribed verbatim for analysis. Respondents were encouraged to provide tangible examples to illustrate their experiences and elaborate on their beliefs and self-reported practices. The transcripts were sent to the participants with a request to review and verify the contents and correct any misunderstandings; this offered them the opportunity to ensure that their ideas had been accurately documented. The approved interviews and the written questionnaires were then anonymized by assigning each participant a number and merging all the texts into one single Word file.

Qualitative analysis

The collected data were examined using a qualitative three-phase approach based on the six-phase method that is typical of thematic analysis (Braun et al., 2019; Braun & Clarke, 2006; Nowell et al., 2017):

1. 1. Familiarizing yourself with your data;

2. 2. Generating initial codes;

3. 3. Searching for themes;

4. 4. Reviewing themes;

5. 5. Defining and naming themes;

6. 6. Producing the report

As the research was underpinned by a pre-existing conceptual framework and driven by the analytical interest of the authors, a mixed deductive-inductive approach was considered. The pre-existing conceptual framework led to the definition of the questions and guided the formulation of the analytical categories into which the emerging codes were placed. The analysis started with an immersion phase, which was conducted by means of repeated readings of the questionnaires and interviews, to acquire a high level of familiarity with the material. Each researcher carried out this task individually. The second phase focused on the coding process, which started with the identification of brief portions of the text to extract codes of content. The discernibly different answers were determined and distinct codes were assigned to the statements while redundant or ambiguous codes were deleted. While assigning the codes, a revision of the initial codes was undertaken by rereading carefully the responses to avoid repetitions due to the nature of the coding process. In this phase, the codes were checked several times and, before creating a new code, it was verified whether it was possible to set the statement in an existing code. Only if none of the codes fit, a new one was assigned. In the third phase, larger themes were defined and discussed with reference to the analytical categories derived from the initial conceptual framework; these were named in relation to the codes and a coding scheme was developed through discussion until agreement was reached among the researchers on the codes identified. The comparison of emerging themes and analytical categories helped to check for consistency between them and to highlight possible conceptual implications.

The coding scheme was used to examine all the textual data. Later, each researcher validated the coding of the interviews independently, applying the coding scheme and verifying the coding of the interviews. Any contrasting findings were discussed until full agreement was obtained among coders. The following seven themes emerged at the conclusion of the coding process:

1. 1. Beliefs

2. 2. Sources

3. 3. Self-reported practices

4. 4. Repertoires/performers

5. 5. Educational implications

6. 6. Limits

7. 7. Benefits

Results

The seven themes are described in the following paragraphs. The quotations reported below are marked with ‘P1’; ‘P2’ … (where ‘P’ indicates ‘participant’), to allow the reader to compare insights and comments belonging to the same or different participants.

Beliefs

This theme relates to what students understood from early twentieth-century recordings. It should be remembered that in Italian, the language of the study, the term ‘registrazioni storiche’ is commonly used to refer to early recordings; since the adjective ‘storico’ (historical) can be used to express both ‘early’ and ‘of historical importance’, this definition was explained and illustrated during the course in order to clarify the time coordinates and the context of the recordings in question. Nonetheless, participants’ responses refer to different criteria, including the type of the discs, the period in which they were made and their historical relevance and value as shown in the following statements: ‘I think of an historical recording as a recording made in the last century, possibly in the first half, I think of vinyl… or even of wax reels!’ (P15), ‘by historical recording I mean a recording as close as possible to the era to which it refers… historical recording, maybe one refers to someone who interpreted something in a convincing way for the time, maybe sixty years ago; yet it is simply a dated recording, not necessarily a reference recording’ (P2). Similarly, P 3 said: ‘I would say those recordings that are prior to the 1950s. Those for me are very distant and are already historical. I think that there are some recorded works that are part of our history in the sense that they are cornerstones of the discography and the interpretation of certain pieces. These are very important and in this sense, they are historical. If I take Karajan’s recording of [Dvořák] Symphony [n. 9] “From the new world,” that can be considered a historical recording … in the sense of an exemplary, very important one, which can be taken as a model’ (P3).

In some cases, participants referred to the manner in which early twentieth-century recordings are particularly helpful when early twentieth-century composers come into question, as was the case with Edward Elgar and Claude Debussy, whose music was recorded by themselves or by performers who worked in close contact with them. In general, participants considered early twentieth-century recordings helpful for broader cultural reasons: ‘They [twentieth-century recordings] are useful for feeding my music culture and expanding my knowledge of historical periods, composers and performers’ (P9). ‘A constant and careful glance on the past is needed to be more aware of what we do every day’ (P2).

Sources

When asked how they have access to early recordings, participants suggested that the Internet is by far the most frequently used source. Although private collections of CDs and even LPs were occasionally mentioned, it was clear that YouTube and other online music platforms were considered the easiest, cheapest and most accessible source of musical information: ‘Thanks to the Internet; YouTube and other audio/video file sharing sites’ (P1); ‘Mainly YouTube. In the past I often listened to performers from the fifties and sixties because at home I have so many vinyl records (33 rpm) that belonged to my dad’ (P2); ‘I think that a lot can be found on the internet today, also on YouTube. Then if you really can’t find anything, you can look into a sound archive, especially for vinyl recordings maybe, those are more difficult, I think … but it never happened to me’ (P16). These statements confirm how pervasively YouTube is present in the participants’ listening habits. Only one participant drew attention to the ‘embarassment of abundance’ and expressed her concern with regard to the ‘difficulty of distinguishing between good and bad recordings among the thousands that are now available via the internet’ (P14).

Self-reported practices

This category refers to the manner in which early recordings may shed new light on the interpretation of the relevant repertoire, focusing on specific interpretative practices. Participants mentioned issues regarding musical style, performance practice, interpretation and instrument-specific differences, thus suggesting the perceived value of early recordings. For instance, students reported: ‘They [early recordings] serve to broaden one’s awareness while trying to interpret a piece. (…) I think they are more useful for the style than for the technique’ (P8). ‘They [early recordings] are useful to understand how a music piece was performed, even if not in the period of composition and by the composer himself; at least they give us a general overview of what was done and how it was done. You can then make comparisons with recent recordings, while paying attention to the editions’ (P12).

In some cases, participants referred to technique and expressivity as two distinct analytical categories and showed different attitudes with regard to the manner in which early recordings can inform a modern interpretative choice: ‘Taking reference from a historical recording, we can decide which interpretative level to take into account, whether expressive or technical, such as phrasing, vibrato and portamento’ (P5). Tempo was considered crucial by one participant: ‘I found myself wondering about the metronome cues in some of Debussy’s preludes and resorted to early [contemporary] recordings to look for answers. Timing can also be a topic to be addressed using early records’ (P12).

Another example regards the violin duets Béla Bartók transcribed for viola, which derived from traditional folk songs. One participant (P16) reported that he found the early recordings of these traditional folk songs a source of inspiration for shaping his own interpretation of Bartók’s duets and getting an idea of the atmosphere from which they originated. In particular, the traditional folk songs provided cues as to the correct tempo and character, as was suggested by the lyrics.

Early recordings were considered an invaluable source of information not only with regard to the interpretation of a given piece but also to the manner in which it came to assume its final form. As was reported by one participant (P16), such is the case with the Concerto for Viola and Orchestra by Sir William Walton, which was recorded in 1937 by Frederick Riddle, then the first Viola of the London Symphony Orchestra. It has been reported that Walton, who was not a violist, had to rely on the soloist, who suggested some modifications in the articulation and in the bowing. The music featured a very difficult passage of semiquavers, which Riddle advised Walton to write legato in order to have a smoother movement of the bow, precisely as he plays it in the recording. Walton gladly accepted this advice and wanted the score to be published with the modifications recommended by the soloist. In 1961, however, the composer decided to write a second orchestration, for which the publisher mistakenly used the solo part written before Riddle’s changes. This last remained the reference score for decades (Dunham, 2006). ‘I didn’t know about this episode’, the student reported, ‘and listening to the earliest recording offered me the opportunity to learn about it’ (P16). The wrong solo part set an interpretive standard that only recently, and thanks to Riddle’s recording, could be reconsidered to better reflect on the compositional process.

Furthermore, participants elaborated on the moment when an early recording can best fit the preparation of a piece. Some participants prefer to get an idea from listening to recordings before learning a new piece, while others tend to do the opposite and listen to an early recording only once the piece has been learned. This last approach reveals particular critical acumen, especially since it highlights the risks involved in aural modelling, especially when potentially conflicting interpretative standards come into question: ‘once the piece has been resolved from a technical point of view, I look for an interpretation that is not influenced by today’s clichés’ (P13).

Repertoires/performers

This theme refers to the musical repertoires and significant performers of early twentieth-century recordings. As far as repertoires are concerned, these recordings were considered particularly useful for the romantic and late romantic repertoire until the beginning of the twentieth century: ‘Certainly with the romantic and late nineteenth-century repertoire and above all with the early twentieth-century repertoire’ (P5). Composers such as Béla Bartók, Claude Debussy and Maurice Ravel were mentioned together with interpreters like Jascha Heifetz, Joseph Joachim, Friedrich-Max Kreisler, Georg Kulenkampff, Yehudi Menuhin, David Fëdorovič Ojstrach, William Primrose, Sergej Rachmaninov and Eugène Ysaÿe. Participants mentioned individual pieces such as Rachmaninov’s Piano Concerto No. 2 played by the composer himself, Brahms’s Hungarian Dances played by Joseph Joachim, Mendelssohn’s violin concerto played by Eugène Ysaÿe and Bach’s cello suites played by Pablo Casals. Astonishingly enough, some participants suggested that also earlier repertoires, such as the Baroque and even the Renaissance could be relevant with regard to early discs, although the reasons could not be sufficiently elucidated. In this regard, it can be argued that some participants confused the notion of historically informed performance with that of early disc. This was also the case with some reference interpreters, as suggested by the fact that two Dutch cellists, Anner Bijlmsa (who died in 2019) and Pieter Wispelwey (still living) were mentioned alongside many other late nineteenth-century instrumentalists whose names and careers are strongly connected with either the contemporary composers, or the repertoire in which they were trained, the so-called late romantic. These examples suggest that participants were often looking at specific interpretations and that their searches, although goal-oriented, were not informed by sufficiently robust knowledge of the historical positioning of the interpreters they had in mind. Again, some participants confused the notion of early music with that of early recordings.

Educational implications

In general, students considered early recordings to be a useful learning tool. When asked how these could be better used for teaching purposes, some even suggested that a class on the analysis of early recordings should be permanently inserted in the curriculum of the conservatoire: ‘It would be helpful to introduce a course on early recordings in the curricular activities’ says P11, or ‘it should be mandatory’ recommends P15.

Regarding further educational activities involving early recordings, one participant reported that these should encourage students to reflect on unconventional interpretative aspects, avoid the ‘mummifications of classical music’ (P15) and ‘debunk false myths on interpretation’ (P9). The value of comparative listening was often highlighted, suggesting that listening exercises could involve different recordings of the same piece to encourage discussion and enhance critical awareness: ‘We could better understand how different ways of bowing and fingering, such as those belonging to the French-Flemish and the Russian violin school, may lead to different musical results’ (P14).

Regarding the educational aspects that could be stimulated by early recording, participants reported elements focusing on transferable skills such as critical thinking, awareness development and critical listening.

Limits

Participants were asked about the constraints that might affect or hinder the use of early recordings. Possible limitations or disadvantages might relate to the interpretations they reveal, the poor technologies used in pre-electric recordings and the ways in which these technologies might have affected the recorded interpretation: playing uncomfortably in front of the horn, taking faster tempos to fit the length of the disc.

With regard to the interpretations, one participant recognized a danger in the overwhelming personality of the great masters from the past: ‘Listening to the great performers of the past can be intimidating as they show their strong personality, which the listener may copy and paste’ (P1).

Other issues regarded variables related to early recording technologies: ‘The conditions in which performers recorded, the lack of time and space, the need to take whatever came out of the recording process, the fact that musicians accepted to take the first performance as good and were not always in their prime represent serious limitations’ (S 14).

In some cases, early discs reveal an interpretative approach to the repertoire different from the current one, suggesting that one century ago, the accepted aesthetic did not necessarily involve flawless renditions. One participant reported that ‘for the piano, the performers often convey greater expressive freedom at the expense of the technical perfection typical of today’s recordings. You can obviously argue that today the recordings can be repeated and modified later, in the past you had one chance’ (P12). This statement highlights the manner in which the work done in a recording studio is understood as a key moment in the process leading to a recorded interpretation and suggests how prominent a role technical proficiency has come to assume in today’s standard.

The poor quality of early twentieth-century audio recordings was one of the major and most widely shared concerns, especially because of its musical implications: ‘the poor quality of the audio does not allow us to grasp many details concerning the sound and the expressive nuances used by the musicians’ (P5).

One issue was that early recordings offer no visual information on how these interpretations were realized. This could be of particular relevance with those instruments where bow strokes and fingerings could make a difference in terms of instrumental technique and interpretative style: ‘Videos give you something more. Visual aspects give you gestures that can be interesting on a technical level, both for me, but even if I had to tell a student’ (P16). This quote highlights the differences between audio and video recordings and the pedagogical potential of using images and videos in educational contexts.

Benefits

Finally, participants were asked to describe and elaborate on the strengths and the advantages of early twentieth-century recordings. They recognized their importance as a valuable source when trying to understand the performance practice of the past or, more generally, when looking for inspiration, as is reported here: ‘They are certainly a very important source of artistic inspiration’ (P1); ‘They help people understand how a certain repertoire was really played’ (P4); ‘They can open new avenues and perspectives for interpretation’ (P8); ‘Early recordings offer interesting ideas for the development of an original interpretation, different from today’s standard’ (P9). ‘Early recordings provide the possibility of studying how the pieces were performed at the time, what was the technique, the use of vibrato, the portamento …’ (P10). ‘Early recordings help you to understand the performance practices that were in use at the time of recording. They also allow us to become aware of how the musical taste, the aesthetics, changed over the years’ (P13).

As already mentioned, special emphasis was placed on those recordings that were realized when the composer was still alive and can be considered as testament to a specific compositional intention, interpretative style or performance practice: ‘I consider working on early discs crucial as it allows me to get as close as possible to how the composer intended the work. In this way, you almost have the illusion of being able to work with them. (…) it could be an enormous source of artistic inspiration’ (P1). ‘The historical closeness with the composer with whom we come into direct contact is one of the strengths of these recordings. They help us get closer to the composer’s ideas’ (P10). These statements suggest how listening to early twentieth-century recordings may offer the opportunity to have a closer connection with past composers: many participants viewed some recordings as embodying composers’ intentions.

Discussion

The purpose of this study was to examine how music students understand and conceptualize early twentieth-century recordings in higher music education. In formulating the research question, the authors worked on the assumption that these recordings have come to represent an incredibly rich and easily accessible source material, and that an important body of scholarly studies is available to both music students and teachers, which address issues of style, performance practice and technique.

As to the first research question (How do students in higher music education institutions understand early recordings?), participants’ responses clearly suggest how much they value early recordings when it comes to analysing the performance practices of the past and developing new interpretations. These results are consistent with previous studies on the use of recordings as learning and practising resources among music students and, to a more limited extent, professionals (Monkhouse & Forbes, 2015; Volioti & Williamon, 2017), and are in line with the idea that these may play an important role in shaping one’s musical development (Volioti & Williamon, 2021) and enhancing musicians’ expressiveness (Lindström et al., 2003). Regarding where early recordings can be accessed, participants recognized the importance of the Internet and the pivotal role played by online platforms like YouTube. These findings are in line with similar studies by Monkhouse and Forbes (2015), who identify the significance of the use of YouTube in music educational environments for the development of skills such as critical awareness and aesthetical appreciation. However, some confusion emerged about the distinction between early nineteenth-century recordings, past recordings of historical relevance and recordings involving the performance of early music. This confusion, which was to some extent confirmed by the responses to the following questions, involves four dimensions: time coordinates, historical relevance, proximity to the composer and type of interpretative approach. Time coordinates were still unclear, for example, the difference between acoustic and electric recordings, despite that the participants had attended a course on the subject. The historical positioning of the discs in question was often overridden by the notion of historical relevance and exemplary value. The degree of closeness between composer and interpreter was sometimes suggested as a criterion for defining a particular recording as historically relevant. Recent recordings of historically informed interpretations of the past repertoires, especially the Baroque, were mentioned as examples of early recordings. Although the responses of our participants suggest a broader understanding of early recordings than the definition adopted by the researchers and presented in the course, the confusion discussed above could have critical implications, especially if the distance between composers, repertoires and interpreters is misjudged, and if these figures are misplaced along the timeline. As has already been suggested, it is also possible that the ambiguity of the Italian term, which is used to refer to both early and historical discs, may have influenced some of the responses. This confusion is strongly suggestive of the manner in which informal learning and the lack of curricular teaching activities focusing on listening to and analysing audio sources might lead to a misrepresentation of the musical past. This was the case with those students who referred to living interpreters or to recordings that have been realized in the last two decades.

Regarding the second research question (How do students refer to early recordings to develop a musical interpretation?), participants showed a strong sense of awareness of the usefulness of early recordings, providing evidence that a historically informed approach can be developed also through the analysis of early discs. Their responses show how they were able to address issues of musical style, performance practice, music interpretation and expressiveness, drawing attention to specific instrument-related characteristics and the opportunity to consider instrumental proficiency and interpretative style as two distinct analytical levels. Most crucially, students emphasized the importance of early recordings for documenting the compositional intentions and the interpretative conventions relevant to early twentieth-century music, as the emblematic anecdote on Béla Bartók and William Walton, reported by one of the participants, illustrates. Participants were also aware of the limitations and constraints typically associated with early discs, such as the poor technologies and the challenging working conditions in recording studios.

Regarding the third research question (How do students use early recordings for educational purposes?), participants’ responses reinforce the idea that early recordings represent an invaluable learning resource and that learning activities involving the critical listening of early and recent recorded interpretations should be more strongly incorporated into music curricula and educational contexts. Students were able to offer examples of the educational implications of early recordings and argued that these may offer insightful views, inspire actions and help develop new learning strategies. Several relevant skills were highlighted, including critical listening, critical thinking and the development of a new awareness of the potential of early recordings.

One issue does not emerge from the participants’ responses, regarding the relationship between early discs and the master-apprentice teaching-learning model (Carey et al., 2013; Gaunt, 2007). It still needs to be verified whether and to what extent sources like early discs may impinge on a learning model that relies strongly, and sometimes almost exclusively, on one single figure, the instrumental teacher. As has been already suggested, students tend to perceive that there is little integration between the work done in the one-to-one tuition and other curricular activities, which are often considered peripheral or even irrelevant. Sometimes, it is the student’s responsibility to integrate experiences from different areas of the curriculum and a discussion with the main teacher can even prove to be problematic (Gaunt, 2010). This seems to be consistent with the work of those instrumental teachers who conceptualize vocational skills in terms of passing on their own experiences and transmitting them to the next generation. Their teaching routine focuses on the student performing or playing through some repertoire, study or technical exercises, and the teacher providing feedback on detailed technical and musical aspects, with little room left for other reflective practices. In this regard, one-to-one teaching is not perceived as an environment in which to explore ‘outside the box’ (Gaunt, 2007). Interestingly, only one participant (P15) highlighted the dichotomy between the traditional one-to-one tuition model, where ‘prohibitions and dogmas’ in combination with a ‘passive attitude on the part of students’ tend to prevail, and the liberating effect that listening to past recordings had on her. This statement opens the way for new investigations and suggests the need to analyse how the teacher-student relationship leaves room for more differentiated learning models and tools.

Limitations and further developments

The current research presents the students’ beliefs and self-reported practices regarding early twentieth-century recordings and explores the manner in which they use them to develop a musical interpretation. Although the findings stimulated several reflections, the study presents limitations mainly due to the qualitative nature of the investigation and the small number of participants (16), all coming from one single institution. Not all musical instruments were represented, with their different characteristics, repertoires and reference interpreters. However, even though the results cannot be generalized, they have strong implications for future research into the use of early and recent recordings for learning and teaching purposes, suggesting that further research in this direction is needed. Aspects such as the redefinition of educational principles, curriculum design, goal settings, teaching strategies and evaluation criteria could be influenced by the use of early recordings and should be the focus of future investigation. Early recordings could provide stimuli for innovative teaching approaches based on student-centred methodologies and, at the same time, invite further reflection on the constraints and limitations typically associated with one-to-one tuition.

Another research strand could lead to a stronger connection between computational and descriptive analysis of early recordings and the music teaching, building on a vast body of scholarly literature whose relevance is still underestimated among music practitioners.