Waiting for an Audience

On the Misuse of a Notion Associated with Listeners

by Steven Rice

Introduction: I have never heard the audience speak, but I have heard it clear its throat

An attempt to generalize about an audience involves a struggle to create a representation of an equivocal group, and thereby for the person that creates the representation to act as a spokesperson for that group. If the generalizations that are made about listeners are not verifiable, there is a potential for abuse in such representations, because these representations can be used as rhetorical devices to justify the expenditure of money on a project, whether this argument is made to a corporate board or to a jury that will award grants. If a particular group is seen as creating music that better satisfies the audience, they may be favored in such transactions. However, determining audience response to concrete situations is highly problematic. I will examine a few approaches to understanding audience expression and then examine the utility of the audience representation for the spokesperson and for individual listeners.

The audience represents a group of people that listen to music on a given occasion and sometimes even the people who might potentially be in attendance on a given occasion. What is somewhat awkward about this situation is that within the audience, there are no individuals in a strict sense. The audience is singular, which gives the impression of a crowd of people experiencing the same things, and a crowd that would articulate its experiences in the same way. Individuals are recognized as a kind of appendage to the audience; the audience member. Therefore, it would seem that what the audience expresses must be expressed in unity, en masse. The audience expresses the will, feelings, and intent of the totality of people that are in attendance. The public is a similar kind of representation, but one not connected to a single occasion. What is far less clear is how one might know what this totality expresses. Can the audience communicate (explicitly or implicitly) as a totality? Or is it necessary to conduct a poll, interview, or test of some or all individuals in attendance, from which one can infer knowledge of the totality?

It may seem that we can understand audience expressions based upon the semantic framework of applause, hissing, booing, or yelling in concert halls in response to music. Let us consider an account that Arnold Schoenberg gave of an audience:

In the front third of the hall, roughly, there was little applause and little hissing; most people sat unconcerned, many stood looking around in amazement or amusement towards the parts of the hall further back, where things were livelier. There the applauders were in the majority — there were fewer unconcerned, and a few hissers. But the most noise, both applause and hisses, always came from the standing space at the back and from the galleries… I never had the impression that the number of people hissing was particularly great. It never sounded full, like a chord of solid applause entering with precision, but more like an ad hoc group of ill assorted soloists… (1)

Notice that, in this example, Schoenberg had to break the audience into smaller constituent groups, which already semiotically destroys the expression of the totality. Even if we were to understand how groups of people were responding, how would we characterize the response of the totality in this instance? Were there multiple audiences or did the audience both like and hate the music? For that matter, imagine that someone left in the middle of that concert. Was that person no longer part of the audience, and would this change the constitution of the audience and its opinions (for example, if the person who left the concert hated the music and is no longer part of the audience, did the audience then like the music slightly more as a totality)? This situation calls into question the utility of a representation such as that “the audience liked the music.”

Even if we were to set this issue aside, we would have to confront another problem with informally assessing audience response, as Schoenberg did in the previous example, which is the lack of verifiability in our observations. One cannot call back an audience from a concert that one attended last night. Where did the audience from the concert last night go? Even assuming that we were able to correctly understand audience communications, we might be led astray in the interpretation of audience reactions a variety of ways. Our impressions of which audience members are participating in a particular action are skewed by our placement within the group of concert attenders, musicians, or stage crew. Also, the typical concert attender communications (if that is what they are) of applause, hissing, booing, etc., are rather semiotically ambiguous. Is she clapping because she enjoyed the music or because the people around her are clapping? At the least, we would have to say that interpreting these individual signs may prove challenging. As Artur Schnabel said, “the public is one of the vaguest notions existing. The public is mute, it cannot speak, can hardly utter a sound, as a whole… I know two kinds of audiences only — one coughing and one not coughing.” (2)

Represent by Number

Perhaps if we want to understand the audience expression as a totality, our only recourse is to conduct a poll or study of the audience’s opinions of music, and we could express the audience’s opinion on the basis of majorities and percentages. Presumably we could also replicate the study or record who responded, so that it would be verifiable. However, if we divide the audience into groups, we must decide what differences are significant enough to merit another grouping category and how to detect differences. The danger, if we are interested in knowing what people actually are thinking, or what the audience is thinking, is that if we divide listeners arbitrarily, the data that we obtain may be more a reflection the implicit ideology of our divisions than of what people actually feel. For example, a survey that asks “Did you like this music or not?” necessarily divides the audience into two groups, which may be a suspect division. If our study is a verbal survey, the questions that we ask to obtain the data, when we ask those questions, the ability of the survey subject to segregate the experience of the survey or test from the experience of the concert, the ability of the respondent to remember the experience of a concert, the ability of listeners to be articulate about their experience, and the possible inability of listeners to segregate different parts of the total experience of their evening in the concert hall all may play a part in distorting the data that we might obtain about listeners’ opinions of a given concert.

Let us examine how data was collected in a recent study of symphony orchestra attenders and potential attenders, conducted by the John S. and James L. Knight Foundation, as well as Audience Insight, LLC. Subjects were called at random on the telephone and asked questions from a script. One question reads: “Would you describe yourself as a critical listener, a casual listener, or an uninterested listener of classical music?” (3) Even if we assume that the subjects are honest and earnestly desire to answer the question to the best of their ability, this question seems difficult to answer, given the lack of explanation for the terminology (further explanation is prohibited by the script) and the apparently arbitrary categories. For example, one could be both casual and uninterested in listening habits or one could be focused without being critical. However, the interviewer is only allowed by the script to record one answer from the options provided. The question also seems to have an inherent bias. The term “casual listener” is an idiom, while “critical listener” and “uninterested listener” are not idiomatic, as well as possibly seeming to be undesirable labels to apply to oneself. Furthermore, given the wealth of possible responses to a question like “How do you listen to classical music,” the three-option multiple choice seems to be an awkward solution. However, the survey analysis summary (called the “Final Report”, which is published separately from the “Final Report Appendix” that contains the methodology and data), which is more likely to be read by businesspeople and administrators than the “Appendix,” finds the responses to this question to have great relevance:

Overall, just 10 percent of potential classical consumers think of themselves as “critical listeners” (self-defined), while 78 percent consider themselves “casual listeners” and 11 percent say they are “uninterested listeners.” Thus, the vast majority of potential customers for orchestras are casually involved with the art form. “Casual listeners” also dominate the audience base. Across the 15 orchestras, 42 percent of subscribers think of themselves as “critical listeners,” while 57 percent say that they are “casual listeners.” For single-ticket buyers, the figures are 28 percent for “critical listeners” and 68 percent for “casual listeners.” (4)

Without any further consideration of the data, this gloss serves as a foundation to suggest long-term projects and plans for symphony orchestras within the “Final Report,” frequently referring back to the importance of “casual listeners,” who, as the dominant group within this representation of the audience, have come to represent the whole. Perhaps research has been integrated into this summary in hidden ways, as there are mentions of focus groups that were not included in the data from this study that was released, but in either instance, the power of the representation of the idea of the audience as a “casual listener” is what makes the argument to a reader here.

Another question on the John S. and James L. Knight Foundation telephone survey reads: “If a friend or family member had FREE TICKETS to a classical music concert by a symphony orchestra and invited you to join them, would you like to go?” The gamut of answers has only four possibilities that are read to the respondent, which include ‘yes’, ‘maybe/it depends’, ‘no', and ‘[don’t know]/refused [to answer]’. (5) The last answer is not read, but represents a hidden scale degree on the gamut of possible subject responses. The scale of answers is more manageable here, but the hypothetical situation is lacking in detail. What is my relationship to the family member? What is on the program? How does it fit into my schedule? Or is the question whether I would go simply because the concert is free? Again, the “Final Report” uses the data gathered from this awkward question as a rhetorical device in the “Final Report”:

Two-thirds of Americans would accept a free ticket to a classical concert by a symphony orchestra, if offered by a friend or family member. So, why are some orchestras having difficulty filling their halls? What’s keeping orchestras from attracting the next 2 percent or 3 percent of adults in their communities? (6)

Not only was the Audience Insight representation painted with far too broad a brush, but its numbers and outlines invite a monochrome color fill of regions that should be shaded. Furthermore, the multiple choice questions leave the respondents with very little agency. They can respond to the Audience Insight, LLC survey by refusing to communicate their intention at all, or by conforming to a range of responses that may have nothing to do with their intention. As Theodor Adorno said, “[a] given alternative is already a piece of heteronomy. The legitimacy of [these] alternative demands has yet to be judged by the very consciousness that is asked to make the decision beforehand… [T]he realm of the essence is falsified by a resume of essentials.” (7) The survey conductors hold so much power to manipulate both the respondents and the data without fear of being gainsaid by the audience that they will represent.

On the other hand, it seems possible to assess certain details about the totality of concert attender response. I would like to hold up a contrasting example of reasonable methodological rigor in testing relating to the perception of musical form that was conducted with the model of a live performance of a Roger Reynolds composition entitled The Angel of Death. The authors involved in The Angel of Death Project had many goals, among them to determine if listeners were able to identify the formal organization that the composer intended when writing the work, as well as what kind of recall of repeated material they demonstrated. The methodology for this study involved two concerts, in which The Angel of Death was performed, where subjects were given sliders, either to indicate their level of emotional involvement or to gauge their familiarity with the material currently being performed. The slider data was recorded in relation to time as the piece progressed. (8) For example, subjects that were to monitor familiarity were given sliders with labels above that read “Resemblance to what has been heard from the beginning of the piece,” with a label on the left end of the slider reading “Completely different,” and on the right end was a label reading “Identical.” (9) On the simplest level, The Angel of Death study circumvented a variety of problems of bias by leaving the subject alone to react to the stimulus without much verbal prompting. The information that the subject is asked to provide is presented in a coherent manner, with a range of possible responses. Furthermore, the study was conducted for two performances to determine if variations in the presentation of the task to the subjects, variations in the performance, or cultural differences, influenced responses. McAdams and the other study authors characterized the data that they gathered by writing that “[t]here were global similarities between the mean response profiles… across concerts, despite differences in interpretation, concert hall, cultural milieu and language of the audience, and instruction/training protocol.” (10)

The situation created by the Angel of Death project was one in which those that conducted the study left the doors open for any kind of experimental response from the listeners. As Reynolds said of the experiment, “[s]everal composer colleagues (particularly Phillipe Manoury) were concerned about the possible implications of the process that I underwent in the Angel Project. ‘What,’ he asked, ‘would you have done if the data showed that your assumptions [about your compositional structures] were greatly at variance with the experimental results [of generalized listener perception]?’” (11) In other words, the listeners were in a position to respond in a way that might have issued a challenge either to the composer or the psychologists, including the possibility of invalidating their hypothesis. Furthermore, the language of McAdams and the other authors clings as closely as possible to the shape of the data. Noting the similarity of mean response profiles to the music, rather than saying that the audience understood the formal structure of the composition, reveals immediately that there were divergences in how the listeners responded, as well as how they responded to the music over time. Studies of concert attenders with rigor matching the Angel of Death studies seem to be a rarity, but if one bent their more scientific model to the task of sifting audience opinion, one could have more confidence in the results. At least for the moment, most surveys of audience opinion are comparatively haphazard and often seem to paint the audience with some kind of foregone conclusion.

A Production for the Mass

Despite the conceptual pitfalls in audience representation and the apparent lack of rigorous studies of audience opinion, certain musicians have explicitly stated that the audience influences their actions. For example, Michael Daugherty said that “[as] I write, I think of how the audience… might react [to my music].” (12) Because Daugherty thinks of how the audience might react, he also implies that he has some knowledge of the nature of the audience’s opinions, but as we have seen, it is difficult to know how anyone could understand audience expression with any certainty. I asked Daugherty at a lecture how he conceived of the audience as he composed, and he eventually clarified that thinking about how the audience might react to his music “is about putting yourself in the role of listener” (13), which collapses his rhetorical grasp at the totality of concert attenders and clearly articulates the more mundane fact that Daugherty occasionally stops to listen to his piece as he composes it, and decides whether it is something that he would like to hear, or perhaps whether it is something that he thinks someone else would like to hear.

At times the audience takes on the character of a wrathful god that might appear to punish his unfaithful servant. John Adams said that “[u]nfortunately, audiences have come to the point nowadays where they expect trouble whenever a new work is on the program… In my own work, I try to find a way to… have a larger cultural impact, be comprehensible to a level of educated people who are not musical specialists.” (14) Anecdotally, I have had conversations with people such as those that Adams alludes to, who are baffled by certain contemporary pieces and wish that they would not be played. I have also had conversations with concert goers that came specifically to hear the “new work on the program.” Again, because of the lack of rigorous studies on audience opinion, I have to assume that Adams is working from interviews of people conducted in just such an ad hoc fashion. There is not necessarily anything wrong with Adams being very concerned about the people that he knows have been alienated by the “new work on the program,” or even trying to write specifically to please them. However, because of the way that Adams frames this comment, instead of certain individuals expecting trouble, it is almost everyone who has come to the concert hall to listen that expects trouble. Rhetorically, this makes it seem that certain other composers (Adams names Milton Babbitt earlier in this interview [15]) compose in a way that will be incomprehensible to everyone that will attend the concert, whereas Adams composes in a way that attempts to be comprehensible to everyone that comes (or at least “educated people”).

The audience also makes appearances to justify or invalidate particular ways of making money as a musician. This often takes the form of the assertion that musicians can relate directly to the audience as an employer. For example, Cornelius Cardew states that “[a composer] can take employment in education, in the service of the state, teaching what he has learnt to other composers… Or he can take employment in industry, writing film or background music, or commercial music of other kinds. Or he can attempt to win the support of an audience.” (16) While I know of a few venues that pay performers and composers directly from ticket sales, such as The Red Room (17) of Baltimore, it is far more common in my experience for venues to pay performers a flat fee for their performance, while the composer is paid through royalties collected from licenses sold by an organization such as the American Society of Composers, Authors, and Publishers (ASCAP), Broadcast Music, Inc. (BMI), or the Society of Composers, Authors and Music Publishers of Canada (SOCAN). Commissions generally come from ensembles or organizations that have received a grant for that purpose or from institutions such as a university or an orchestra. Even “rock stars”, which we might think are supported by the audience, are generally paid by the corporation that distributes their music, and from the royalties for their recordings. In all of these instances, one does not interact with a listener totality.

This raises the question of the salient utility of representations of audience response. If one cannot relate directly to an audience as a totality, why speak as if one could? Of course, this anonymous, unverifiable, and mute entity will never contradict the wizard who conjures it up, so it is an easy trick to turn, and the term has a certain populist air about it. A bogeyman who composes for “musical specialists” is frequently evoked in opposition to the audience. Standing apart from such indifferent bogeymen, super-musicians are out “striving” to save the embattled audience. For example, Dawn Upshaw’s biography extols her “fierce commitment to the… communicative power of music” (18), the ensemble eighth blackbird is “widely lauded for its… efforts to make new music accessible to wide audiences,” (19) and “Bang on a Can strives to expose… music as broadly and accessibly as possible to new audiences…” (20)

More to the point, does the idea of composing and playing for the audience serve a listener well? A personal reaction seems almost excluded from within the mass. Perhaps an audience member twitches a little in sympathy, like a nervous eyebrow-tick, but the official reaction will come later from the audience mouthpiece. Therefore, the audience can actually silence the opinions of individual listeners. You may have liked that Schoenberg Orchesterstücke, but audiences hate it. Conceptual categories such as these make themselves manifest through the actions that people take while being informed by them. To return to demographic research vis-à-vis concert attenders, we should remember that to create constituencies, we must divide the audience with a conception of difference, and observe that deciding what differences are significant is a matter of aesthetics and worldview, which are expressed through terminology. Focusing on one element and making decisions about presentation and content on that basis is necessarily reductive, and a simplification such as that often relies on ideology to determine what elements are the most important. For example, marketing research, as applied to concerts, often only is interested in attendance or ticket purchases. While content might be considered in relation to whether someone will attend again, listeners have been reduced in this paradigm to consuming automata, which, however complex, are shuffled and prodded to see what will increase or reduce purchases. The audience is the music consuming mass that complements the music producing corporation. Therefore, the audience weakens the expression of the individual and transfers its voice to those that presume to speak for the audience. If we labor to serve the audience, who do we actually serve?

An Ensemble of Models

As the aforementioned examples illustrate, we are often tempted to generalize about an audience on the basis of far too little information. If we generalize about the totality of people in attendance, we should do so on the basis of verifiable information that we have obtained from individuals with a rigorous and circumspect methodology, and characterize the data we obtain as the expression of those interviewed and nothing else. Musical activity aspiring to populism that extends more than a tip of the hat in the direction of an audience abstraction would require that one represent the audience in a way that gives agency to the individual listeners. Also, there is often no cogent reason to attempt to express the intent of the totality of concert attenders other than the temptation to exaggerate for rhetorical effect. If, unrestrained by reference to concrete items, to models, or specific instances of an item, we jump to conceptions of the opinions of a totality, the representation we create will do little to dissuade us of our delusion. Concepts are more open to manipulation than models are. “A model covers the specific, and more than the specific, without letting it evaporate in its more general super-concept… [I suggest thinking in] an ensemble of analyses of these models.” (21) Too often, references to the audience are an effort to circumvent the complicated and messy world of individual reactions to music. When one deals with an ensemble of models, it is hardly even necessary to decide what to call the ensemble, since real people already have names.

Notes

1. Arnold Schoenberg, Style and Idea: Selected Writings of Arnold Schoenberg (New York: Philosophical Library, 1950), p. 98.
2. Artur Schnabel, My Life and Music (New York: Dover Publications, 1988), p. 204.
3. Audience Insight, LLC, and The John S. and James L. Knight Foundation. Classical Music Consumer Segmentation Study: How Americans Relate to Classical Music and Their Orchestras: Final Report Appendix. Miami: The John S. and James L. Knight Foundation , 2002. [Cited 19 September 2006], available from http://www.knightfoundation.org/research_publications/detail.dot?id=178183. Internet. P. 21.
4. Idem. Classical Music Consumer Segmentation Study: How Americans Relate to Classical Music and Their Orchestras: Final Report, p. 8.
5. Idem. Final Report Appendix, p. 17.
6. Idem. Final Report, p. 12.
7. Theodor W. Adorno, Negative Dialectics, translated by E.B. Ashton (New York: Continuum, 1983), p. 32.
8. Stephen McAdams et al, “Influences of Large-Scale Form on Continuous Ratings in Response to a Contemporary Piece in a Live Concert Setting,” Music Perception 22/2 (Winter 2004), pp. 307–311.
9. Ibid, p. 309.
10. Ibid, p. 323.
11. Roger Reynolds, “Epilog: Reflections on Psychological Testing with The Angel of Death,” Music Perception 22/2 (Winter 2004), p. 354.
12. Michael Daugherty, within McCutchan, The Muse that Sings, p. 174.
13. Michael Daugherty, Composition Symposium at the Eastman School of Music, 1 November 2007.
14. John Adams, within McCutchan, The Muse that Sings, p. 72.
15. Ibid. However, Adams perception of the dominance of composers that he describes as having a “bunker mentality” is grounded in the popular myth of serial composer dominance in America during the Nineteen Fifties, Sixties, and Seventies, which has at least in part been called into question by Joseph Straus in “The Myth of Serial ‘Tyranny’ in the 1950s and 1960s” (see pp. 301–343).
16. Cornelius Cardew, Stockhausen Serves Imperialism, p. 60.
17. The Red Room Collective, The Red Room at Normals Books and Records (Baltimore MD, 2008) [Cited 5 May 2008]. http://www.redroom.org
18. IMG Artists, Dawn Upshaw, Soprano: Full Biography [Cited 5 May 2008]. http://www.imgartists.com/?page=artist&id=95&c=2
19. eighth blackbird, About webpage [Cited 5 May 2008]. http://www.eighthblackbird.com/about
20. Bang on a Can, About Us webpage [Cited 5 May 2008]. http://www.bangonacan.org/about_us
21. Adorno, p. 29.

Bibliography

Adorno, Theodor. Negative Dialectics. Translated by E.B. Ashton. New York: Continuum, 1983.

Audience Insight, LLC, and The John S. and James L. Knight Foundation. Classical Music Consumer Segmentation Study: How Americans Relate to Classical Music and Their Orchestras: Final Report. Miami: The John S. and James L. Knight Foundation , 2002. [Cited 19 September 2006], available from; http://www.knightfoundation.org/research_publications/ ; Internet.

_____. Classical Music Consumer Segmentation Study: How Americans Relate to Classical Music and Their Orchestras: Final Report Appendix. Miami: The John S. and James L. Knight Foundation, 2002. [Cited 19 September 2006], available from; http://www.knightfoundation.org/research_publications/ ; Internet.

Cardew, Cornelius. Stockhausen Serves Imperialism. Ubu Classics, 2004. [Cited 21 January 2004.] http://www.ubu.com/historical/cardew

McAdams, Stephen, et al. “Influences of Large-Scale Form on Continuous Ratings in Response to a Contemporary Piece in a Live Concert Setting.” Music Perception 22 no. 2 (Winter 2004), pp. 297–350.

McCutchan, Ann. The Muse that Sings: Composers Speak About the Creative Process. Oxford: Oxford University Press, 1999.

Reynolds, Roger. “Epilog: Reflections on Psychological Testing with The Angel of Death.” Music Perception 22/2 (Winter 2004), pp. 351–356.

Schnabel, Artur. My Life and Music. New York: Dover Publications, 1988.

Schoenberg, Arnold. Style and Idea: Selected Writings of Arnold Schoenberg. New York: Philosophical Library, 1950.

Straus, Joseph. “The Myth of Serial ‘Tyranny’ in the 1950s and 1960s.” The Musical Quarterly 83/3 (Autumn 1999), pp. 301–343.