Audience reactions to the program notes of unfamiliar music

Familiarity and liking

The relationship between familiarity and liking has been explored by researchers for nearly half a century, with inconclusive results. Unlike Berlyne (1970, 1971), who proposed an inverted U-shaped relationship, North and Hargreaves (1995) found a positive association between familiarity and liking insofar as undergraduate students rated excerpts from recordings of pop songs that were familiar to them higher than those that were unfamiliar. In the context of live pop music, however, Brown and Knox (2016) have shown that audiences seek novelty rather than familiarity. By contrast, Thompson (2006) found no relationship between audience members’ enjoyment of a classical concert and their familiarity with the music played, although he proposed a preliminary model in which listeners’ connection with the music as well as the players may contribute to their enjoyment (Thompson, 2007).

Program notes, listeners and performers

Feld (1994) has suggested that music listeners, even when hearing music that is new to them, draw from a toolkit of “experiential anchors” and “interpretive moves” acquired through previous listening experiences. These anchors may be one of the factors influencing audience reactions in emerging research from the Guildhall School of Music and Drama (Halpern & Sloboda, 2015), where recent work has revealed that some listeners like music more upon a second hearing. Conversely, in practice-led research, Blom (2006) brought the concept of experiential anchors into question when she found herself without such anchors when learning a conceptually challenging work. An experienced pianist, Blom found it necessary to gain an understanding of the compositional process and context before practice could commence.

Building on this work, Viney and Blom (2015) observed that experiential anchors can be weak or non-existent when hearing new, stylistically challenging music, and that these situations require performers and listeners to build a new “interpretive platform” using additional sources such as text and metaphor.

Also reporting from the performer perspective, Bennett and Blom (2014, p. 177) began to question the assertion that additional sources are always useful and positive. In their study, information in the form of compositional notes was introduced part-way through the learning process of a newly composed work. The information overrode the initial interpretive thinking of the violist Bennett and was found to capture “unnecessary textual details”, having a negative impact on the preparation of the work. Recognising that the same notes had been given to audiences, Bennett and Blom (2014, p. 178) called for research that questions “how much compositional knowledge is useful in a program note and how much becomes counter-productive”.

In one of the few other studies that concerns program notes, US researcher Elizabeth Margulis (2010) concluded that, contrary to her hypothesis, text descriptions provided in advance of listening reduced listeners’ enjoyment of music. Margulis’ study employed excerpts of Beethoven string quartets and she distinguished (p. 289) between two types of program note text: dramatic notes that described, for example, a “hymn filtered through the ears of someone passionately connected to it”; and structural notes such as “a series of slow, sustained chords that grow louder and achieve resolution”. Surprisingly, the dramatic program notes reduced listeners’ enjoyment more than those that were structural.

Whilst the study reported here study sought to build on the work undertaken by Margulis, we incorporated several contrasting and novel features. For example, Margulis played listeners excerpts from the canon and asked them to rate their enjoyment after program note descriptions were provided. We requested open-ended responses both before and after the provision of descriptions, and we did this by performing each of the works twice. Bearing in mind that the Beethoven excerpts may have been familiar to Margulis’s participants and were likely to have had a familiar tonality, we sought to understand whether listeners might receive program notes more favourably when hearing works that were new to them, works in unfamiliar languages, and works with a challenging tonality. For this reason, the works we selected were new to the audiences and to us, and had a modern, somewhat dissonant tonality. The other novel aspect of our study is that we involved the voice, with works sung in Russian.

Lyrics and the listener

For singers to perform the words of songs and arias in such a way as to convey meaning to the audience, the words must be meaningful to the singer. As such, understanding lyrics is generally believed to be of vital importance for singers. Accordingly, most singers’ training includes the learning of languages such as Italian, French and German; they are assessed on their ability to communicate in these languages (Davidson & Coimbra, 2001) and until the advent of websites for lieder and libretti, every concert singer’s bookshelf would have included Prawer’s (1964) The Penguin book of lieder and Bernac and Radford’s (1970) The interpretation of French song, and every opera singer would have held key volumes of the Castels’ (n. d.) Operatic libretti series.

Furthermore, audiences in English-speaking countries are invariably given translations of the texts of oratorios, arias and songs included in recitals, whether these are integrated within or as adjuncts to programs, and whether spoken introductions to the repertoire are provided by commentators, as in the worldwide broadcasts from the Metropolitan Opera in New York, or by the performers themselves. This suggests that managers, promoters and performers all agree that audiences should be told what singers are singing about.

The extent to which listeners attend to the meaning of lyrics set to music, however, is likely to be widely variable. While researchers such as Collister and Huron (2008, p. 109) have asserted that “concertgoers and music listeners frequently complain of the difficulty in comprehending the lyrics of the music”, there is also evidence to suggest that the desire to understand the texts of vocal music is not universal. This is seen in the results of a study conducted by Fine and Ginsborg (2014), whose questionnaire study sought to identify the factors perceived by 143 listeners – professional and amateur musicians including singers, singing teachers and regular listeners to vocal music – to underlie the intelligibility of sung text. Respondents were explicitly asked if it was “not”, “quite” or “very” important to them to be able to understand sung text in both familiar and unfamiliar languages. Only 17% of respondents rated the ability to understand words in an unfamiliar language as very important, although 51% rated it as quite important. Content analyses were made of respondents’ open-ended comments. These included:

I remember when studying nineteenth century German lieder and French chansons in my twenties, I was bowled over by the music and the gist of what it was portraying, but not understanding the individual meaning of words. It was all very romantic and I projected my own meanings onto the sung words (F, 56, music teacher and amateur singer). (Fine & Ginsborg, 2014, n. p.)

By comparison, 61% of Fine and Ginsborg’s respondents rated the ability to understand words in a familiar language as very important and 32% as quite important. This varied not only, as in the previous example, with personal preference, but also with genre and listening context:

When I listen to jazz songs in the background, the text is not important and I am just sometimes happy to grasp a phrase. When I listen during driving a car on a motorway, I want to understand the text I think (M, 28, semi-professional singer). (Fine & Ginsborg, 2014, n. p)

In the present study, then, the authors chose to explore audiences’ perceptions of two songs for voice and viola in a language familiar to the singer but not to the violist or to any of the audience members, and presented with and without program notes and translations.

Theoretical framework

The study adopted Ferrara’s (1984) analytical framework that details “five steps in listening”. We employed Ferrara’s framework to analyse the syntactical, semantic and ontological meanings expressed by listeners, who also stated whether these personal meanings were positively or negatively impacted when the actual meaning of the words was communicated to them. The features, explanation and original listening foci are shown in Table 1.

OPEN IN VIEWER

Table 1.

Theoretical framework derived from Ferrara (1984).

Features	Explanation	Focus
Syntactical meaning	Sounds are considered in relation to their individual and connecting phonemic qualities. Semantic and ontological meanings are ignored.	Sound
Possible references	Interpretation of sounds and sound relationship at a fundamental level: relationships to known origins and complex understandings of the sounds within a work.	Semantic meaning
Relation to life-world	Interpretation of a whole work, which may include historical and socio-political contexts.	Ontological meaning

In Ferrara’s framework, listeners begin with a syntactical orientation in which they listen openly to “everything”. They then begin to separate sounds for their individual and phonemic qualities before considering the meanings behind, or relationships between the sounds: the semantic meaning of the music. Next, listeners consider ontological meanings: the work in its entirety. This might include, for example, the historical life-world of the composer and the circumstances in which the work was written. Ferrara, who also considered the role of feeling, movement, gesture, space, temporal relations and tactile qualities, was careful to point out that the “lived time” of a composition is unique and the outlooks and values that informed this lived time impermanent.

To understand the semantic and ontological thinking of listeners, we needed to learn whether listeners emphasized dramatic and/or structural elements of the music. For this aspect of the study, the program note research conducted by Margulis (2010) provided a framework with which to analyse listeners’ narratives for affective, imaginative language (dramatic features) and for objective, structural language (structural features).

The research also drew on principles of artistic or practice-led research: in this case, research led by performance. Practice-led research enables researchers to consider the knowledge inherent in their practice (Blom, 2006; Hannan, 2006), in the creative outcome (Bolt, 2006; Crossman, 2006), and in both (Odam, 2001). Of vital importance is the ability to be within the practice under scrutiny rather than “abstracted from the loom that produced it” (Carter, 2004, p. 1). As such we undertook the research from inside the practice, as performer-researchers who both studied and performed the works and managed the research itself. This answered Sloboda’s (2013, p. 13) well-founded criticism that “in the vast bulk of existing music perception research, the musicians involved in making the music don’t even know that the research on their music is taking place”.

Participants

Potential audience members were recruited through word-of-mouth musical and social networks with invitations issued via noticeboard, flyer or phone call. The 29 audience members were aged between 16 and 74 years; 15 were female. Seen in Table 2, 14 participants reported engagement in amateur music-making, three had undertaken post-secondary training in music and 19 had learned to play an instrument or to sing, mostly as children. Sixteen participants had attended live classical music concerts during the previous 12 months and 20 had attended other live music events. Participants listened to a range of music including rock, classical, jazz, big band and popular. No listeners had heard the works before and none of them understood the Russian lyrics. Preliminary discussions with music society contacts led to the inclusion of the works’ titles. These were given to lessen listeners’ concerns about having to “form an opinion that makes sense.”

OPEN IN VIEWER

Table 2.

Process and audience survey questions.

Performance 1: listeners have only the titles of each work Written survey 1: questions posed after performance 1.

• Have you heard either of these works previously?

• Did you know anything about the works prior to the performance?

• Did you understand the lyrics?

• What do you think the songs are about?

Program notes shared orally with audience Performance 2: listeners have listened to the program notes Questions posed after performance 2 (written survey)

• Did you listen to either or both of the songs in a different way once the background information was given? If yes, how did your listening differ?

Music-learning

• Have you taken formal music instrumental and/or vocal lessons? If so, on what instruments and for how long?

• Have you engaged in informal musical learning such as self-learning or learning with friends?

• Have you undertaken classes in music theory, appreciation, history etc?

• Have you undertaken any formal music training at a post-secondary level? If so, please describe this training.

Music-making

• Are you involved in any music ensembles or groups? If so, please tell us how often you are involved and what you do.

• Are you involved in any other types of music-making?

• Music-listening

• In the past year, how many concerts have you attended as a listener? Please include shows, popular music, jazz, musicals, operas etc.

• How often do you listen to or watch recorded music; what do you choose?

• Do you have any other feedback for the performers?

Procedure

Once informed consent was obtained from participants and ethical approval granted, we performed the music to two small audiences, one in the United Kingdom and one in Australia. As noted, and in contrast with Margulis’s (2010) use of Beethoven excerpts, to ensure that earlier listenings would not influence the findings we adopted music that was unfamiliar to listeners.

Two experienced musicians, a soprano and a viola player, gave the performances. The music was also new to the performers and comprised two rarely performed settings by Boris Tchaikovsky (1994) of poems by Rudyard Kipling loosely translated into Russian. The two paired works were short and contrasting, giving listeners the opportunity to comment on two pieces in a short amount of time. Once the data collection was complete, the performers presented a number of other short works that included collaborations with local musicians.

Prior to the first performance, listeners were given only the song titles: On far away Amazon and Homer. Prior to the second performance, brief program notes were given orally by the singer for two reasons: the researcher-performers wished the audience to attend to both performances in ways as similar as possible, rather than reading about the songs and following the text and translations as they listened, and because (for the purposes of the wider research project, reported in Ginsborg & Bennett, 2015) they needed to ask the audience afterwards if they had noticed that the viola player performed the first song from memory while the singer performed the second song from memory.

The oral program note explained that the performers had chosen the work because it is for the rare combination of mezzo-soprano and viola, and was not known to either performer before rehearsals had commenced. The singer then read the text of the poems by Kipling that had been translated into Russian (see Appendix), followed by their back-translation into English, and pointed out the liberties that had been taken by the original Russian translator: for example, On far away Amazon was originally titled The beginning of the armadillos, Kipling’s boat sailed weekly from Southampton rather than each Thursday from Liverpool, and the refrain “Rolling down to Rio” was replaced by “To Brazil!”.

After each performance, listeners completed a written questionnaire survey that posed closed and open-ended questions about their listening experiences and understanding (Table 2). The first three questions in survey 1 were closed. These established that no one had heard either of the works previously, no one knew anything about the works prior to the performance, and no one understood the lyrics. Listeners also recorded demographic details including music-learning, music-making and music-listening experiences. All 29 listeners returned completed questionnaires after the concert and 27 listeners provided responses to both works. Responses ranged from single-word answers to several sentences in response to the descriptive questions.

The design of the audience response questions was informed by the extant research, the need to ascertain whether listeners were familiar with the works or the Russian language, and the need to ensure sufficient background data for subsequent phases of the study. This article addresses the three key audience response questions:

Analysis

This was a qualitative study. To reduce error and bias in coding the transcripts (Mays & Pope, 2000), analysis began with independent, naturalistic coding. This involved multiple readings of listeners’ responses without applying pre-determined labels (Kreiner, Hollensbe, & Sheep, 2009). Following Glaser and Strauss (1967), we then used a constant comparative analytical scheme to unitise and categorise the text. Analysis moved gradually from a close association with individual cases towards a concern with broad analytic themes. This led to agreement on common coding categories, interpretations and conclusions (Lincoln & Guba, 1985).

We then coded deductively, using Ferrara’s features of listening (see Table 1) and Margulis’s program note types (dramatic and structural), and following the same steps outlined above. Finally, data were displayed in a way that is conceptually pure, making distinctions that are meaningful. This included further discussion of multiple themes including inter- and intra-theme connections.

The works performed

On far away Amazon is marked Moderato (at a moderate pace) and features two stanzas with bridge passages and a final coda. The viola accompaniment is almost entirely in staccato quavers, giving the piece what one listener described as a “driving feel”. The vocal line, too, is dominated by a quaver rhythm with a syllable for each note. Written in E♭minor, the dynamics range from pp (very soft) to ff (very loud). The lyrics are included here, and we include the original text of the poems in the Appendix.

На далекой Амазонке: On far away Amazon

1. На далекой Амазонке

On far away Amazon

Не бывал я никогда.

I never visited.

Только “Дон” и “Магдалина” -

Only Don and Magdalene -

Быстроходные суда

Fast-going ships

Только “Дон” и “Магдалина”

Only Don and “Magdalene”

Ходят по морю туда.

Go on the sea there.

Из Ливерпульской гавани

From Liverpool harbor

Всегда по четвергам

Always on Thursdays

Суда уходят в плаванье

The ships leave to swim (on a voyage)

К далеким берегам.

To far away shores.

Плывут они в Бразилию,

They swim to Brazil,

Бразилию,

Brazil,

Бразилию.

Brazil.

И я хочу в Бразилию

And I want to go to Brazil

К далеким берегам!

To far away shores!

2. Никогда вы не найдете

You’ll never find

В наших северных лесах

In our northern forests

Длиннохвостых ягуаров,

Long-tailed jaguars,

Броненосных черепах.

armored turtles.

Но в солнечной Бразилии,

But in sunny Brazil,

Бразилии моей,

Brazil mine,

Такое изобилие

(there is) Such abundance

Невиданных зверей!

[of] Never-seen beasts!

Увижу ли Бразилию,

Will I ever see Brazil,

Бразилию,

Brazil,

Бразилию,

Brazil,

Увижу ли Бразилию

Will I ever see Brazil

До старости моей?

Before my old age?

The second work, Homer, features a dissonant viola accompaniment with frequent, held harmonics and less of a sense of pulse than the previous work. Written in the brighter key of C major and again marked Moderato, the vocal phrases are separated by bridge sections and are not repeated. The lyrics are given here.

Гомер: Homer

Гомер все на свете легенды знал,

Homer all in the world legends knew,

И все подходящее из старья

And all suitable from old trash

Он, не церемонясь, перенимал,

He without hesitation adopted,

Но с блеском - и так же делаю я.

But with polish – and likewise do I.

А девки с базара да люд простой

And maids from the market as simple folk

И все знатоки из морской братвы

And all experts from the sea brotherhood

Смекали: новинки-то с бородой,

Understood: the new thing has a beard (the “new” story is old)

Но слушали тихо - так же, как вы.

But listened quietly - the same as you.

Гомер был уверен: не попрекнут

Homer was certain: [they] wouldn’t accuse [ridicule]

За это при встрече возле корчмы,

For this, while meeting near a tavern,

А разве что дружески подмигнут,

But may be with friendliness wink,

И он подмигнет - ну так же, как мы.

And he would wink - well, same as us (as we would).

As explained earlier, listeners heard the pieces twice. For the first hearing, only the titles of each work were given. The performers made no introductions until after the first performance.

The first performance: Listeners’ focus when hearing an unfamiliar work

As seen in Table 3, six common references were identified from the listeners’ responses to Amazon. These concerned negative emotions such as sadness and hardship, references to Brazil and a journey, positive emotions such as happiness and reconciliation, love, and an authoritative feel.

OPEN IN VIEWER

Table 3.

Listener responses to hearing On far away Amazon for the first time.

Common references (29, from 27 respondents)	Count (n = 29)
Sadness and hardship	13
A journey, leaving or returning home, references to a “homeland”	6
Reference to Brazil (forest or exotic sounds)	3
Love	3
Happiness or reconciliation	2
Authoritative	2

Of the six common terms shown in Table 3, the nine Brazilian and journey references related explicitly to the song title, which was the only information listeners had been given: for example,

A journey: “Exploration of something a long way away”

A journey and reference to Brazil: “Somebody walking to the Amazon”

Further readings of the complete narratives revealed that the title also influenced listeners’ references to sadness, love and happiness: for example,

Possibly missing someone who was far away from home.

Sad, repetitive, possibly about leaving home and remembering homeland.

In total, the title of the work was found to have influenced 18 of the 29 listener narratives.

Shown in Table 4, six common references were identified in the responses to Homer. Despite its major tonality, the most common response related to sadness or fear. However, five listeners noted a feeling of calm or reflection and four listeners suggested that the work might include a love theme:

The mood was sad and the piece was delicate and beautiful.

Reflective, sombre, and little sad.

OPEN IN VIEWER

Table 4.

Listener responses to hearing Homer for the first time.

Common references (34, mentioned by 27 respondents)	Count (n = 34)
Negative emotions: sadness and fear	14
Delicate, calm and reflective	5
References to Homer	5
Love	4
Comedic elements	4
A journey, making reference to the previous work	2

The pairing of this work with Amazon prompted two listeners to continue the theme of a journey, suggesting that they assumed a link between the two works even though none had been given. A deeper reading found that this assumption was implied in multiple responses: for example,

The futility of daily life in a poor country…

I think that the second piece was about loss again, but it was also about love.

Again, multiple responses were informed by the title of the work. Five listeners mentioned the poet Homer. One listener noted Homer’s blindness (which is not universally accepted – see Beecroft, 2011) in relation to the “thin sound of the harmonics”. Another listener, shown below, quoted from Shelley’s 1817 sonnet Ozymandias, whose ruined statue is all that is left of a civilisation destroyed “by the impersonal, indiscriminate, destructive power of history” (Sparknotes, 2015). In the same vein, a third listener referred to the act of passing on wisdom and knowledge, as detailed in Homer’s poem The Odyssey in the form of the Greek god Mentor.

An older person telling a story to a younger person (or peers) and passing on wisdom and love. Is this of the Homeric poem? Maybe.

Like hearing a well known piece of poetry – it felt like, “look upon my works ye mighty, and despair” – the Ozymandias poem.

The four mentions of comedic elements were a direct response to the facial expressions of the performers: a visual cue unlikely to have been noticed by all listeners and not a feature comparable with Margulis, who worked from recordings. One audience member suggested that listeners might link the “sombre” soundscape with this seemingly disparate visual cue:

Marriage/divorce – life’s ups and downs. Drinking/comic/tongue-in-cheek (??), taken from the facial expressions.

The first performance: Listeners’ consideration of ontological, semantic and syntactical features

Our second analysis from the first performance concerned Ferrara’s listening framework, and 27 comments from the 29 respondents were included. In particular, we looked for mentions of syntactical, semantic and ontological meaning as described in Table 1. A sample of responses is included in Table 5 together with the count. Of the listeners, 70% focused on the ontological meaning of the first song, and 63% of the second: for example,

A tale of hardship and toil, difficulty and triumph over adversity.

OPEN IN VIEWER

Table 5.

Features identified by listeners in response to the two works.

Features	Work 1	Work 2	Indicative comments
Ontological	19	17	Longing, reflective, thinking, of the “what only”, “what if”, etc.
Syntactical	2	3	I was concentrating on the overall musical sound rather than meaning
Ontological and syntactical	2	3	Slightly mournful, dirge-like. Authoritative tone, quite animated. Not particularly melodic
Ontological and semantic	2	1	Sounded scary at one point, as though character at risk. Accompaniment sounded light, as though accompanying a fairy tale
Ontological, semantic and syntactical	2	3	Themes/ideas from the poet, time & distance painted by the harmonics (fading sound from several centuries/millennia earlier) – wistfulness in the music, thin sound of the harmonics is the voice of the poet coming across the centuries, and maybe representing his original blindness

The second performance: Changes in listening

Once they had heard the pieces a second time, listeners received a second survey (see Table 2). They were asked, “Did you listen to either or both of the songs in a different way once the background information was given? If yes, please tell us how your listening differed”. Of the 29 listeners, 28 responded to the question, and all 28 reported that they listened differently having been given the program notes. Of these 28 listeners, 11 (39%) reported that the program notes had had a positive impact on their listening experience. As in the example below, positive comments largely related to being able to relate the text to the music, or to confirm aspects of their initial interpretations.

Knowing the subject matter it was easy to recognise a rollicking, maritime aspect to the tempo. Recognising words e.g. Brazil, Margherita [sic], made the narrative more accessible.

My ideas of Amazon were to do with exploring new places and seeing more new sights, and this seemed to be confirmed by the text.

Better understanding of the context; able to better relax and enjoy the experience.

This finding does not align with Margulis’s 2010 study; nor does it align with Fine and Ginsborg’s study (2014), which found that only 17% of listeners rated as very important the ability to understand words in an unfamiliar language. However, as previous research has suggested, listeners’ responses vary according to listeners’ personal preference and genre (Fine & Ginsborg, 2014), engagement and commitment (Pitts, 2005), social enjoyment (Pitts & Spencer, 2008), age (Sloboda & Ford, 2012), understanding (Dobson, 2010) and audience experience. It is likely that a further point of difference concerns works that have dissonant harmonies, such as the works in this study and the works that Viney and Blom (2015) found to be beyond their experiential anchors. Further, it is possible that the oral (rather than written) presentation of program notes made a difference to their reception.

As mentioned, some listeners had earlier noted “comedic” facial expressions that were in contrast with the “sombre” sound of Homer. Some of the positive responses to the program note related to a resolution of this conflict:

[I now] understand the tone and facial expressions relating to the words. Made much more sense. Listened to the words and music much more thoroughly, could relate much more.

The positive impact of the program note in alleviating such confusion or disparity could perhaps be assumed, and yet 17 of the 28 listeners were either neutral or negative about the inclusion of program notes. More in line with the studies mentioned above, neutral and negative receptions related to disparities between the text and the music soundscape and between the text and listeners’ own interpretations, and to the need to “listen” to more than the music itself:

For both pieces, my mind was trying to make the music fit the information given about the pieces, rather than just listen to the music and interpret it on its own merits.

I tried to interpret the music as a humorous story. It still sounds like a lament.

Our earlier interaction with listeners (Ginsborg & Bennett, 2015) suggested that those with less musical experience and knowledge might focus on the ontological meaning of new works, whereas the ontological meanings of more experienced listeners might be informed by musical elements (semantic meaning). Once they had the background information, we also found that less experienced listeners made semantic (and sometimes syntactical) relationships and adjusted their initial thoughts to align with the known meaning, whilst more experienced listeners often preferred and retained their original thinking about the works.

Similarly, within this larger sample, some listeners replaced their own thinking with that given in the program note, accepting the latter as authoritative:

Instead of thinking the song was about love, I knew it was about the Amazon.

Other listeners reacted by finding a way for their initial interpretations to align with the program note. For example, the following listener had originally thought that Amazon was about a memory. After the second performance, she wrote memory into her thinking even though this was not the case:

I can see why this one was a memory because they were wishing to go to Brazil and they had this memory of it happening.

For other listeners, often those with more experience, a contrasting interpretation presented by the program note was unwelcome and in some cases had a negative impact on the listening experience. This was particularly prevalent in reflections on Homer, with its light-hearted text and sombre soundscape:

I had hoped to have a deeper relation to the music, but I preferred my first interpretation. The words had a rather trite relation to the music, which I think I ignored.

Listeners unable to align their initial thinking about the work and the later program notes expressed similarly negative reactions:

I was completely off track and remain so. I failed miserably to find a connection.

I couldn’t tell that music was trying to tell us about a man’s life, even though I knew.

I was looking for the place names, e.g. Southampton, Brazil, and the ships. The Russian, guttural sound does not help the interpretation. I prefer my previous/original interpretation of a sad story.

Listeners’ alignment with Margulis’ descriptions

Margulis’s (2010) two types of program note descriptions – dramatic and structural – provided our third point of analysis and involved identification of “dramatic” notes, which related to listeners’ use of affective, imaginative language, and “structural” notes identified from listeners’ use of objective, analytical language. The 53 valid responses (26 and 27 for each work) were coded as dramatic, structural or both. This revealed that 47 (87%) of the reflections were entirely dramatic and six (11%) were both dramatic and structural. No responses were entirely structural.

At this early stage of the study, the results suggest that listeners variously welcome and reject others’ interpretations and/or others’ information in the form of program notes. Listeners appear to focus their listening on dramatic elements of the music, and it would seem that less experienced listeners are less likely to note structural features in favour of purely ontological or semantic listening. It follows that Margulis’s (2010) finding that listeners were particularly negative about dramatic program notes may relate as much to listeners’ emphasis on and, possibly, ownership of, their thinking about dramatic elements of the music, as to their interest in the works themselves.