Poetics 35 (2007) 277–300

www.elsevier.com/locate/poetic

Media modes of poetic reception

Reading lyrics versus listening to songs§
Sibylle Moser *
University of Vienna, Hyegasse 3/46, 1030 Wien, Austria

Available online 21 March 2007

Abstract
This paper introduces the comparative study of modalities of poetic language (print/song) and
corresponding modes of reception (reading/listening). Results of semi-standardized focused expert inter-
views are presented on the background of a constructivist model of media self-organization. The interviews
were conducted with 18 creative professionals in Austria and Canada and focus on Laurie Anderson’s song
Kokoku (1984). The aesthetic experience of the example and the systematic comparison of ‘‘reading lyrics’’
and ‘‘listening to songs’’ allow for the inductive differentiation of the categories ‘‘perception of media
modality’’ and ‘‘metaelaboration of media mode.’’ Detailed explication of these categories suggests that
media-specific perception and text processing occur independently of linguistic competence. Based on the
interview results, four dimensions of the media specificity of song reception are outlined: (1) nonverbal
dimensions of language; (2) text fragmentation versus text coherence; (3) genre-specific interplay of lyrics,
text performance and music; and (4) intermediality of listening and reading.
# 2007 Elsevier B.V. All rights reserved.

1. Medium print: the blind spot in empirical studies of literature

In his introduction to cognitive poetics, a rising field in the empirical study of literature, Peter
Stockwell (2002) frankly states on the first page: ‘‘Cognitive poetics is all about reading
literature.’’ Despite the emergence of digital media, which enable the convergence of audiovisual
and verbal signs, the majority of literary scholars continue to observe poetic communication
within the print-dominated paradigm of written texts and reading audiences. Accordingly, most

§
Parts of this paper have been presented at the IX Congress of the International Society for the Empirical Study of
Literature and Media (IGEL) in Edmonton, 2004, and at the Conference ‘‘The Art of Comparison’’ of the ESA-Network
in Rotterdam, 2004. The research was funded by a grant from the Austrian Academy of Sciences (APART – Austrian
Program for Advanced Technology) 2002–2005.
* Tel.: +43 1 966 17 11.
E-mail address: sibylle.moser@univie.ac.at.

0304-422X/$ – see front matter # 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.poetic.2007.01.002
278 S. Moser / Poetics 35 (2007) 277–300

empirical studies in literature refer to reading processes in (post-)modern societies, which are
realized in individualized situations, foster silent reception and follow the conventions of
polyvalent interpretation and fictional reference.1 In contrast to this ‘‘monomedia’’ perspective,
media historical studies in literary communication have demonstrated that the dominance of
print is a relatively recent development. The wider distribution and the reception of literature
emerged in the centuries that followed the rise of the printing press. Gutenberg’s
groundbreaking invention, the reproduction of script with movable type, enabled more
efficient publishing structures and finally contributed to the spreading of literacy in the 19th
century (McLuhan, 1962; Schön, 1987; Triebel, 2001). Accordingly, the aesthetic value
systems of a reading bourgeoisie inform literary studies to the very day. In focusing on a canon
of printed ‘‘high literature’’ and its reception by academic readers – most often scholars and
their students – literary researchers tend to overlook that poetry is right out there, on radio
channels, on the Internet, in music stores and in concert halls. Songs and their lyrics bear
witness to the pertinent presence of oral genres in contemporary systems of literary
communication. Moreover, multimedia performances of the historical avant-garde (dadaism,
expressionism, futurism), beat literature and the performances of the 1960s and today’s
spoken word artists brought orality back into the literary canon. While being treated as mere
folklore, romanticized as ‘‘natural’’ by the idealistic aesthetics of the 19th century and put into
sharp opposition to the ‘‘high art’’ of aesthetic distance (Finnegan, 1977:30–41; Storey, 2003),
spoken, recited and sung poetry has in fact been a vibrant part of artistic linguistic practice
throughout the 20th century.
From a historical perspective, poetic practices throughout the world originally occurred in the
form of song and musical, multimedia performance (Finnegan, 1977:13; Zumthor, 1990:142;
Danesi, 1993:7). Songs are a multisensorial mode of linguistic communication which has never
ceased to exist. Until today, the vast majority of oral poetry is communicated through the various
genres of popular music in the tradition of African-American music that represents a vibrant
counterpart to print-oriented ‘‘high culture’’: blues, jazz, rock ‘n roll and its diversification into
uncountable subgenres – hip hop, punk, new wave, soul, funk, reggae and grunge to name but a
few – spread the sounded word up to the present and reach wide and diversified audiences. It may
be true that in multimedia societies printed poetry does not reach wide audiences, since functions
of literary reading such as entertainment or identity formation are increasingly fulfilled by
audiovisual media offerings (cf. Schreier and Rupp, 2002:262–263); however, given the
prevalence of popular music in everyday life, it is likely that poetic texts are experienced,
interpreted and enjoyed as song lyrics by a significant number of people. The question remains
how exactly the oral reception of poetic texts takes place and in which respect it differs from
reading processes. So far, few empirical researchers have paid attention to the fact that lyrics
occur in different media modalities, namely oral (e.g., on CD or on the radio), printed (e.g., CD
booklets) and audiovisual (e.g., music videos). When studied by psychologists and literary

1
For a documentation of this focus on reading processes see for example the online proceedings of the more recent
congresses of the International Society for the Empirical Study of Literature and Media (IGEL): http://www.arts.
ualberta.ca/igel/igelconf.htm. Out of 28 empirical studies concerned with literary communication on the conference in
Pécs 2002, 71% focused on printed texts, one of them investigating the reading of song lyrics. From the 28 papers from
the congress in Edmonton 2004, 61% were concerned with the reading of written texts; topics and readings of the summer
institute preceding this congress exclusively discussed the reading of literature. Participants of a symposium entitled
‘‘literature in oral cultures,’’ which was held at IGEL 2006 in Chiemsee, Munich, did not investigate reception processes
but focused on the linguistic and rhetorical analyses of historical texts.
 S. Moser / Poetics 35 (2007) 277–300 279

scholars, lyrics are most often analyzed in cognitivist and/or hermeneutic frameworks, which are
based on representationalist, print-determined presumptions. Existing studies focus on the
‘‘decoding’’ of textual meaning, thereby implicitly suggesting the independence of semantics
from the pragmatics of media technologies and language practice (Stratton and Zalanowski,
1994; Sousou, 1997; Alkalay-Gut, 2000; Whitten and Graesser, 2002). Sousou (1997), for
example, in a study on the impact of songs on mood and memory, interprets the voice of the
singer as a confounder that spoils the purity of musical effects. In order to figure out the impact of
music and lyrics on the mood of her subjects, she has them read the lyrics with classical
background music, ‘‘to eliminate the possibility of singing becoming part of the melody’’
(ibid.:33). This divorce of text and music for the sake of an experimental design defies the
purpose of investigating the reception of lyrics, which by definition are sung texts and not read
texts. The notion of a song text deprived of melody and of a singing/intonating voice is a
contradiction in terms. Sousou therefore does not study the effect of listening to songs, but rather
investigates a particular kind of reading experience – namely, the reading of lyrics with
background music.
The study of popular music, on the other hand, is traditionally inflected by a tension between
the interpretation of sociocultural lifestyles and subcultures on the one hand and the structural
analysis of musical forms on the other. Recent trends in popular music studies therefore try to
establish a dialogue between cultural studies and musicology and formulate a semiotic bridge
between musical structures and cultural meaning (Middleton, 2000; Burns and Lafrance, 2001).
The following paper will build on this promising semiotic mediation and complement its
integrative approach to popular music with a media-theoretical perspective on the reception of
pop lyrics. My initial observation is that there is a vast difference between a reader who is lost
between the silent pages of a book and a listener who hears the voice of a performer through the
sounds of a live or recorded song. What is the function of the text in a song? Which role do media-
specific qualities of language such as sound and rhythm play in poetic reception? And last, but not
least: Why do so many people listen to popular songs? What is their vast appeal? The general
nature of these questions does not address the concrete diversity of musical and textual styles in
popular songs. Rather, it points to the specific mediality of texts in songs and the connection
between poetic language and the musical form of song.
In what follows, I will explore the media specificity of oral text reception with results from a
series of expert interviews that focused on the song Kokoku (1984) by multimedia performer
Laurie Anderson. Laurie Anderson represents an interesting example for the transgression of the
borders between popular culture and so-called high art. She started her career in the New York art
scene of the 1970s, became famous with multimedia performances in the 1980s and has reached a
wider audience with her songs ever since her piece ‘‘O Superman’’ made its way into the British
charts in 1981.2 Most of her songs are extractions from multimedia performances; they are
guided by short poetic narrations and various adoptions of electronic music.
My exemplary and tentative discussion of the reception of Anderson’s piece will be situated
on the background of a media theoretical framework, which focuses on the corporeality and
embodiment of poetic experience. Since oral poetry and song respectively have traditionally been

2
Despite its minimalist instrumentation ‘‘O Superman’’ sold a million copies worldwide. My attendance of some of
Anderson’s latest shows (Toronto/September 2001, Berkeley/March 2002, Innsbruck/September 2003, Montréal/Feb-
ruary 2004) proved, that she still has a wide and diverse audience: they were all sold out, in Montréal the organizers had to
schedule an extra show for a Sunday afternoon due to the high demand.
280 S. Moser / Poetics 35 (2007) 277–300

theorized in contrast to written texts, I will first introduce media theoretical approaches to the
distinction between orality and literacy and argue that theories which assume a ‘‘great divide’’
between oral and literate modes of language tend to reify the theoretical distinction at stake. As
an alternative, second, I will discuss the constructivist media theory of Siegfried J. Schmidt as a
promising perspective for the integration of psychological, social, cultural and media-specific
dimensions of poetic reception. Third, I will present data from interviews with aesthetic experts
that suggest a range of different cognitive and communicative dimensions for the distinction
between oral and literate modes of poetic reception. Finally, I will discuss the consequences of
these interview results for the setup of experimental research designs, which would aim at a
media sensitive study of both printed poetic texts as well as orally performed lyrics.

2. Orality versus literacy: the ‘‘Great Divide’’ theory

The Toronto School of Communication (e.g., Harold Innis, Eric Havelock, Walter J. Ong,
Marshall McLuhan) provides the most prominent media theoretical framework for the
observation of the difference between oral and literate modes of language. Theorists of this
posthumously (re-)constructed ‘‘school’’ of media theory assume the genuine existence of oral
and literate forms of cognition and social organization. That is, these thinkers claim that
communication technologies such as the mnemonic devices of oral tradition and the Greek
phonetic alphabet determine cognitive as well as social dimensions of language use. On this
perspective, media modalities of language – such as orality and literacy/print – determine
psychological and social processes of linguistic processing and textual interpretation.
Walter J. Ong (2002/1982:11), for example, characterizes the thought processes of primary
oral cultures, that is, of cultures ‘‘totally untouched by writing or script,’’ by qualities such as
‘‘additive,’’ ‘‘redundant,’’ ‘‘close to life world,’’ ‘‘empathetic and participatory’’ and
‘‘situational and homeostatic’’ (ibid.:37ff). In contrast, he portrays literate societies as
dominated by deductive reasoning, individualism and linear abstract thought. Accordingly,
Marshall McLuhan (1962) claimed in his famous book The Gutenberg Galaxy that Western print
culture has exchanged the space of aural immediacy for the Cartesian space of visual print
culture, summarizing this perceptual change with the catchy formula ‘‘an eye for an ear’’
(McLuhan and Fiore, 1996/1967:44). He furthermore claims that qualities of this lost acoustic
space, such as its immersive nature, its immediacy and its emotionality, reoccur in the
communication practices of the ‘‘electronic age.’’ The new ‘‘secondary orality’’ (Ong, 2000/
1982:134ff.) is ‘‘(re)composed based on writing’’ and nowadays most often implemented
through audiovisual communication technologies (Zumthor, 1990:25). The oral communica-
tion practices of radio, television and digital media therefore do not transcend the distinction
between orality and literacy. According to McLuhan, they reestablish the oral mindset under the
electronic conditions of the global village. In the same line, media theorist Derrick de
Kerckhove argues for a sharp contrast between oral and literate ‘‘modes of listening.’’ In
correspondence with Ong’s profile of oral and literate mindsets, he distinguishes between these
two modes of listening as follows:

The basic difference between the two modes is that oral listening tends to be global and
comprehensive, while literate listening is specialized and selective. One is attending to
concrete situations and to persons, whereas the other is interested in words and verbal
meanings. One is context-bound, the other is relatively context-free. The first is cosmo-
centric and spatial, the latter is linear, temporal and logocentric. (de Kerckhove, 1995:104)
 S. Moser / Poetics 35 (2007) 277–300 281

Since de Kerckhove formulates his distinction on background of an analysis of contemporary

media systems, we can assume that his distinction is meant to cover current listening practices.
Hence, listeners of contemporary popular songs are to perceive lyrics either in an oral or in a
literate mode and they can change between these two modes: ‘‘We switch on one mode or another
depending upon our circumstances and our need’’ (ibid.). However, de Kerckhove seems to
suggest that listeners cannot apply the two modes at the same time. They cannot, for example,
focus on single lines or words of lyrics and simultaneously experience a song as spatial and
immersive.
Distinctions such as oral and literate listening underpin ‘‘Great-Divide Theories’’ (Chandler,
1994). These theories are guided by the dichotomy of orality and literacy, which is often driven
by a problematic contrast of primitive versus civilized (magic versus rational, female versus
male, etc.) cultures (cf. Biakolo, 1999). While the insight that the specific ways through which we
communicate determine our knowledge of the world improves the understanding of semiosis and
cognition, the technological determinism of the Great Divide theories and the missing evidence
for the mutual exclusion of media modes of language remain problematic.

3. Orality and literacy: constructivist media theory

Despite similarities and intersections, the constructivist media theory of Siegfried J. Schmidt
(1996, 2000) differs in at least two aspects from the Toronto School of Communication. From the
systems theoretical perspective that informs constructivist reasoning, the Great Divide approach
suffers from two theoretical insufficiencies:

(1) It equates cognition with information processing in the framework of cognitivist approaches,
which elaborate on the computer model of mind. The comprehension of printed texts is
modeled as the ‘‘decoding’’ of meanings that are stored in the text and can be ‘‘retrieved’’ (cf.
Ong, 2002/1982:138). Likewise, language, knowledge and communication technologies such
as the alphabet are often theorized in analogy to disembodied computer software (de
Kerckhove, 1995:28). In contrast to this cognitivist position, constructivist models of
cognition emphasize the concrete embodiment of semiotic meaning (cf. Varela et al.,
1991:172–173), no matter if it emerges out of acoustic, visual, audiovisual or multimedia
reception. The notion of embodiment addresses a range of qualities which all aim at the
contextualization and dynamic sampling of cognitive representation (cf. Varela et al.,
1991:147–184; Ziemke, 2003). Meaning emerges from the history of interaction between a
cognitive system and its environment (Ziemke, 2003:1307). This ‘‘historical embodiment’’ is
realized through concrete movements based in the feedback between perception (sensors)
and motor activity (effectors) (ibid.). In humans, sensorimotor knowledge becomes most
obvious in the intrinsic interdependence of sound and gesture in linguistic processing (cf.
Danesi, 1993; Iverson and Thelen, 1999; Roth, 2002). Finally, the notion of embodiment
encompasses the idea that in contrast to man-made machines, humans act as living systems
whose ‘‘fundamental variable to be maintained constant is its own organization’’ (Ziemke,
2003:1308). This dimension of ‘‘organismic embodiment’’ constitutes the autonomy and
self-organization of human representation and emphasizes the functional relevance and
situatedness of cognitive operations (cf. Núñez and Freeman, 1999). Thus, recipients of
poetic texts actively construct and physically embody semiotic meanings on the basis of their
somatic, social, psychological and cultural conditions and according to their individual and
cultural needs (cf. Schmidt, 1996).
282 S. Moser / Poetics 35 (2007) 277–300

(2) Many Great Divide theorists interpret media technologies as deterministic causes. While
constructivist models of self-organization emphasize the complex interplay and the
interdependence of different variables, most proponents of a Great Divide assume a linear
causal correlation between literacy and modernity, thereby introducing writing as a necessary
and sufficient cause for modern civilization (see, for example, Havelock, 1991:24). This
assumption could not be validated by current empirical studies on the complex roles literacy
plays in developing countries (Olson and Torrance, 2001:11–13; Triebel, 2001). In order to
unfold its merits, literacy has to interact with a number of crucial parameters such as
educational policies and economic welfare. Since Great Divide theorists lack the idea of self-
organization, they cannot sufficiently account for the systemic interplay of body, mind,
society and culture.

In contrast, Siegfried J. Schmidt’s media model (Fig. 1) introduces ‘‘cognition’’ as the

enactment of relevant distinctions or ‘‘observations’’ in complex environments. Based on
writings of cognitive scientists such as Humberto R. Maturana, Francisco Varela and Ernst von
Glasersfeld, this understanding of cognition elaborates on the idea of embodiment and
consequently it is not confined to deductive problem solving or symbolic representation.
Cognition means to enact an environment successfully, an ability that encompasses knowledge
about perceptual, conceptual, emotional and practical operations alike. Perceptual pattern
formation, emotional evaluation and higher functions like the formation of categories are realized
through the same logic of self-constitution. ‘‘Communication’’ complements these individual
cognitive processes through social knowledge. It encompasses the coordination of actions and
processes of social attribution such as the formation of social identity. In fact, both sides are
intrinsically intertwined and permanently translate into each other. Cognition enables
communicative structures, which on the other hand inform cognitive processes. ‘‘Programs’’
of cultural distinction allow for the evaluation of semantic possibilities on both sides (Schmidt,
2000:39). Finally, media systems enable this complementary process by the structural coupling
of cognition and communication; whatever changes on one side has an effect on the other.
According to this model of media self-organization, orality and literacy are both modes of
cognitive embodiment, each of which represents a complex body of knowledge. Reception
modes such as listening to songs or the reading of lyrics represent two kinds of media-specific
knowledge that relate to each other. Hence, it is rather unlikely that printed and oral texts occur in

Fig. 1. Self-organization of media, culture, cognition and communication (based on Schmidt, 2000:98).
 S. Moser / Poetics 35 (2007) 277–300 283

mutually exclusive ways or that the reception of printed and sung lyrics requires mutually
exclusive cognitive and social skills. While representing a media-specific form of knowledge,
which can, for example, be observed on the level of language use, the respective oral and literate
modes of reception have to be understood in relation to each other. This interplay of different
media practices reflects the cognitive and communicative dynamics of media systems. As social
systems, they are organized dynamic units based on selective structures. Hence, the cognitive and
communicative functions of different media systems and the media modalities they deploy
provide actors with agency by means of limitation. Reception processes unfold as the interplay of
the possibilities and limitations represented by media modalities and their respective modes of
reception. By using media technologies in specific ways, communicating actors extend their
cognitive faculties and create media-specific environments, which represent specialized
psychological and social spheres of existence.
The systems theoretical definition of observation as a ‘‘selective distinction’’ which creates
sociocultural environments meets with McLuhan’s idea that each medium enhances the active
selection and integration of information and at the same time reverses into the reduction of
behavioral possibilities:
Examination of the origin and development of the individual extensions of man should be
preceded by a look at some general aspects of media, or extensions of man, beginning with
the never-explained numbness that each extension brings about in the individual and
society. (McLuhan, 1997/64:6)
Yet, I would not interpret these ‘‘dialectics of media’’ to be driven by an ‘‘equilibrium of the
senses,’’ an assumption McLuhan may have inferred from his fascination with the first-order
cybernetics (‘‘automation’’) of his time (ibid.:346–359). The idea that media demarcate a realm
of possible actions rather corresponds with the position of ‘‘the embodied mind’’ as outlined
above. It furthermore resonates with Schmidt’s assumption that media systems exist and function
in relation to each other (Schmidt, 2000:194). I therefore assume that popular songs represent an
‘‘intermedia practice,’’ which enacts and thus embodies the interplay and integration of oral,
literate and audiovisual modes of linguistic communication.

4. Talking about ‘‘Kokoku’’: ‘‘observing’’ experts

Given the lack of media sensitive studies on the reception of lyrics, the central question for the
empirical observation of sonic practice is how to operationalize listening processes that
encompass reflective as well as affective, bodily and social operations. What are the most salient
dimensions for the observation of media modalities of language and their corresponding modes
of reception? Which items would enhance the ecological validity of a questionnaire that
compares the effects of printed lyrics and songs in an experimental research design? The
following section will sketch first tentative answers derived from a series of semi-standardized
focused expert interviews.

4.1. Material

The interviews focus on the lyrics of Laurie Anderson’s song Kokoku (1984). These lyrics are
presented as a print text from a CD booklet, as a song recorded on CD and as an audiovisual
performance in a concert video. Besides the fact that the lyrics of this song naturally occur in
three different media modalities, the piece was chosen for its richness of possible verbal and
284 S. Moser / Poetics 35 (2007) 277–300

musical enactments of language. The lyrics of the song alternate a Japanese chorus sung by two
female background singers with English lines melodically recited by Laurie Anderson. Likewise,
the music mixes electronic instrumentation with traditional Asian sounds. The lyrics are
syntactically simple and resemble a range of linguistic features associated with oral speech such
as paratactic sentence structure, formulaic expressions and unspecified deictic expressions. At
the same time, they represent a relatively abstract and opaque narrative, a semantic quality
traditionally associated with printed poems. The choice of a song by Laurie Anderson limits the
generalization of some results in this study to the genre of avant-garde pop. However, this
limitation is outweighed by the fact that Anderson’s piece corresponds with the artistic life
worlds of the interviewees and challenged their expertise. The song and its lyrics are therefore
able to serve as a rich conversational impulse (‘‘focus’’) for the exploration of the different
dimensions of oral and literate modes of poetic texts.

4.2. Procedure

All interviewees were sent the printed lyrics before the interview along with the instruction to
read them thoroughly and not to listen to the song. When we met, I presented them with the song
on CD and finally with the performance of the song on video, an aspect I cannot address in this
paper.3 The order of presentation aims to capture the ways in which the understanding of the
lyrics is actualized and possibly altered through their musical implementation. In this way, both
the more durable reading experience and the ephemeral listening experience were accessible to
the interviewees. Interview questions, usually one closed and one open-ended question per topic,
were roughly structured along the two sides of Schmidt’s media model (‘‘cognition’’ and
‘‘communication,’’ Fig. 1) and focused on psychological and social dimensions of the media-
specific reception of the lyrics.
The main questions were concerned with the perception of the lyrics in song and print, the
elaboration of the meaning, and the emotional, cognitive and conative effects of the three media
offerings (printed text, song and video sequence). Furthermore, I asked about background
knowledge of genres of popular music as well as of aesthetic and social conventions of popular
culture and of Laurie Anderson as a performer. These general dimensions are derived from the
reception theories of Schmidt (1991/1980) and Groeben and Vorderer (1988). However, since
both of these theories were developed for the literary reading of printed texts, neither offers a
detailed discussion of the manifestation and the functions of different media modalities of lyrics.
I therefore added the following media-specific categories.
On the side of cognition, I asked questions about the specific perception of the media modality
as such. This category is concerned with sensory aspects of reception, such as the specific
qualities of sound and voice in listening or the quality of layout in reading. The perception of
media modality also deals with the specific modes in which information is processed in a certain
medium. It addresses, for example, the ways in which the interviewees enact the intonation or the
silent reading of a text.
On the side of communication, I addressed the metaelaborations of media. This category is
concerned with metalinguistic statements and with the reflexive observation of specific

3
Since the analysis of audiovisual communication requires a theoretical framework on its own, the comparison
between reading, listening, and viewing goes beyond the scope of this paper. The analysis of the video reception of
‘‘Kokoku’’ is in the making and will be presented in a separate follow-up article.
 S. Moser / Poetics 35 (2007) 277–300 285

psychological (‘‘cognitive’’) and social (‘‘communicative’’) functions of oral and written

linguistic practice. Thus, metaelaborations of media helped to profile oral and literate modes of
the cognition and communication of lyrics.
The interviews took place in a personal setting and were recorded on video. Since most
subjects are used to speaking and/or performing in public contexts, they agreed to the recording
and the public use of the interviews.

4.3. Participants

In order to specify the theoretical dichotomies at stake, I decided to extend my understanding

of media modalities in conversation with of a group of recipients who were trained to engage in
the exploration of the phenomenon in question: creative professionals, who are familiar with
popular cultures, and who are specialists in the practical use of particular media such as print,
music or multimedia technologies. Accordingly, the theoretical sample is structured along three
media modalities of lyrics – oral, literate and audiovisual – each of which is represented by
artistic or art-oriented professionals in their respective fields. I contacted, for example, literary
authors, cultural critics and translators as experts for print; performance artists, singers and
musicians as experts for oral communication; and artists, filmmakers and architects as experts for
visual and/or audiovisual media. The interviewees are nine Canadians and nine Austrians. Two to
four Canadian and Austrian professionals represent each media modality. The dimensions of the
theoretical sample allow exploring whether linguistic competence (English native and non-native
speakers) and expertise (media-specific professions) have an impact on the perception, the use
and the social definition of media modes of poetic reception.

4.4. Method of data collection: expert interviews

Since the conversations are qualified as expert interviews, the question arises, how the notion
of ‘‘expertise’’ and the associated notion of knowledge is conceptualized in this study. According
to constructivist media theory, media expertise encompasses cognitive as well as communicative
dimensions; the idea of embodied cognition extends expertise into the field of tacit knowledge,
such as aesthetic experience and practical know-how. This understanding partly exceeds the
definition of ‘‘expert’’ in the framework of mainstream cognitivism, where the term refers to ‘‘a
person with special skills and knowledge acquired through experience, rather than inherent
talent’’ (Peskin, 1998:237). In accordance with this definition, most of my interviewees could
name a variety of genres of popular music and had background knowledge of Laurie Anderson’s
work, most had an understanding of what ‘‘avant-garde’’ meant to them, most of them had insight
into the aesthetic conventions and communicative rules of popular culture.
According to the constructivist notion of expertise, I consider my interview partners to be
aesthetic experts in various ways, each corresponding with a specific level of ‘‘observation.’’ On
background of the outlined constructivist model of cognition, these expert observations are
defined as the ability to draw cognitive and communicative ‘‘distinctions’’ and to distinguish
between distinctions themselves, that is, to reflect on cognitive and communicative semantic
differences:

 First-order observation: All my interviewees have an intense history of active participation in

popular cultures; most of them are familiar with contemporary art, most of them know Laurie
Anderson and have personal experiences either with her music or her art work. In this sense,
286 S. Moser / Poetics 35 (2007) 277–300

they are experts of aesthetic experience, they have an interest in and a readiness to expose
themselves to this rather unconventional form of popular music. This aspect of expertise
represents the ‘‘aesthetic, pleasure oriented’’ dimension of the interviewees’ media
competence (cf. Schreier and Rupp, 2002:254). In categories such as the ‘‘perception of
media modality,’’ the interviewees formulate a concrete phenomenology of the way in which
they experience the mediality of texts. Since they do not consciously distinguish between
reading and listening but formulate experienced differences, I label these distinctions as ‘‘first-
order observations.’’
 Second-order observation: Within the category ‘‘metaelaboration of media,’’ the interviewees
explicate specific knowledge which guides their professional work with media and therefore
meets with the traditional understanding of expertise. However, creative professionals not only
embody technical knowledge and knowledge about production processes, but also enact
‘‘interpretative knowledge about subjective relevance, rules, perspectives and interpretations’’
(cf. Bogner and Menz, 2002:43–44). Thus, the second-order observations labeled as
‘‘metaelaboration of media’’ explicate the ‘‘technical, instrumental, analytical-critical and
productive-elaborative’’ dimensions of media competence (cf. Schreier and Rupp, 2002:254).
 Third-order observation: Finally, the interviewees comment on the interview situation and the
setup of the media presentation itself. This aspect is relevant for the evaluation of the interview
setup.

This constructivist research design emphasizes the mutual observation of researcher

and subjects and is therefore dedicated to the ecological validity of the data. My basic
assumption is that media researchers can learn from media professionals in general and from
artists in particular. Thus, I use the interviewees’ statements in order to inductively
differentiate the theoretical dichotomies at stake. In this sense, the interviews are part of an
interactive process of deductive-inductive theory construction. Distinctions elaborated in the
mutual process of conversation are used to develop a better theoretical understanding of
different media modes of poetic reception and to stimulate the development of hypotheses. In
order to capture the distinctions between orality and literacy, questions are systematically
asked in a comparative way, which confronts the reception of the printed text with the
reception of the song. The following interview section demonstrates this comparative style of
conversation as well as the way in which interviewees formulate their experience of different
media modalities. My interview partner below is Monica, 28, a spoken word performer from
Toronto:

I (Me): Could you give an example for the expansion of the reception experience while
listening?
M: Example of how it goes beyond?

I: Yeah, just, . . .

M: Just simply by adding instrumentation, and you know the source of music and background
effects that take you right into the piece, take you into a place, and then her voice which sort
of, just by speaking alone sets the mood, sets the atmosphere kind of you know that sort of
leads you to where you’re supposed, where she wants you to be when you hear it, as opposed
to the page which you can pretty much be in any place you want to when you read it. The
music forces you to be in a certain place.
 S. Moser / Poetics 35 (2007) 277–300 287

I: Are there also experiences with the text which are lost while listening to the lyrics?

M: I don’t think so, not with this particular piece.

[. . .]

I: There’s not any advantage or disadvantage in reading as such?

M: You know this is one of those pieces I’d probably like to hear, you know if I heard it on a CD
I’d be like ‘‘Hey, that’s kinda cool! I’m gonna read it!’’ Like it would be one of those things
I’d read after the fact I’d hear it and get drawn into it and its concepts and stuff and then I
might read it later for the deeper meaning as opposed to how we did it. Because I think more
power lies in it in its audio form than in its written form.

I: Right, right.

M: So it’s written form, I would almost, in my personal experience, in my personal world, I

would use sort of to supplement its audio, to kind of go deeper into it. As opposed to, you
know, using the audio to kind of go deeper into it.

Monica mentions a wide range of criteria that profile her understanding of the media
modalities of print and song (see underlined words); some of them are in accordance with, while
others contradict media theories of orality and literacy. In my data presentation, I will focus on
the inductive differentiation of the category perception of media modality and on the cognitive
and communicative functions which were assigned to the reading of lyrics and to song listening
in the category metaelaboration of media. I will furthermore introduce results concerning de
Kerckhove’s distinction of oral and literate listening.

4.5. Method of data analysis: computer aided qualitative content analysis

In order to deal with the ‘‘entropy’’ produced by these extensive communications on listening
and reading, I analyzed the transcripts (about 900 pages) with the TAMS-Analyzer, a freeware for
qualitative content analysis developed by Matthew Weinstein.4 TAMS-Analyzer allows for the
definition of codes and code sets and offers a range of hypertext like operations. It facilitates both
the qualitative coding of interviews and a tentative quantitative structuration of the material.
Coding units are the answers to the questions of the interview thread. The analysis was guided by
the following questions: How do the interviewees characterize media modalities? Which aspects
of reception seem to stand out for them? Which functions do they assign to the reading of lyrics
and to song listening? Thus, I am primarily interested in how the interviewees characterize media
modalities and the corresponding media modes. However, this qualitative exploration implied
two quantitative aspects as well. The number of interviewees who mention a subcategory
indicated to what degree the respective dimension of a media modality seems to represent a rather
common feature of a media mode. The distribution of nationality and profession among each

4
For details see http://educ.kent.edu/mweinste/tams/; TAMS-Analyzer for Macintosh System OSX can be down-
loaded from www.sourceforge.net for free.
288 S. Moser / Poetics 35 (2007) 277–300

subcategory indicated to what degree the perception of media modalities seems to depend on
selective and specialized observations. Thus, the quantitative structuration stimulated the
generation of hypotheses about the impact of different variables on the enactment of media
modalities and media modes respectively.

5. Results: distinction of the media modes reading and listening

5.1. Perception of media modality: listening to Kokoku

When listening to the song Kokoku the most important subcategories in the category
‘‘perception of media modality’’ were the following (Table 1).
Text performance: Table 1 shows that all 18 interviewees (100%) brought up the specific ways
in which the performers intonate the text. All of them mentioned the category several times,
emphasizing that the text performance vitally influences their interpretation as well as the effect
of the text. The importance of concrete performance becomes also apparent in the rich
phenomenology describing Laurie Anderson’s voice. Observations ranged from more technical
aspects such as ‘‘timbre’’ and ‘‘sonority’’ to the assignment of semantic qualities such as ‘‘sexy,’’
‘‘an inner voice,’’ ‘‘toxic,’’ ‘‘parental,’’ ‘‘a medium,’’ ‘‘surreal,’’ ‘‘prophetic,’’ ‘‘authoritative,’’ ‘‘a
voice of memory,’’ ‘‘artificial and distant’’ and ‘‘a techno whisper.’’
Text fade out: Fourteen of the interviewees (78%) mentioned that they temporarily or even
fully ignore the text while listening, or that they had problems to acoustically discern the words –
among these also seven of the English native speakers. Interestingly, all six of the (audio-)visual
professionals mentioned this category.
Relation of text and music: Twelve of the interviewees (67%) brought up the fact that not only
the voice interprets the text but also the music. In Kokoku the content of the lyrics (frames of
reference were, for example, ‘‘war,’’ ‘‘evolution’’ and ‘‘cosmos/science fiction’’) were
experienced in contrast to the soothing music by seven of the interviewees. The sound of
the Japanese language and the Japanese chorus were characterized as complementing each other
and going together well with the instrumentation by seven interviewees.
Interference with reading: Twelve of the interviewees (67%) made a topic out of the inability
to separate the listening experience from reading the lyrics beforehand (third-order observation).
Text fragmentation: Ten of the interviewees (56%) perceived the lyrics in fragments and
focused only on selected lines or words – among these were also four English native speakers.

Table 1
Listening to Kokoku – distribution of subcategories of ‘‘perception of media modality’’ among nationalities and media-
specific professions
All (18) CA (9) A (9) Oral (6) Print (6) Audiovisual (6)
Text performance 18 (100%) 9 (100%) 9 (100%) 6 (100%) 6 (100%) 6 (100%)
Text fade out 14 (78%) 7 (78%) 7 (78%) 4 (67%) 4 (67%) 6 (100%)
Relation text/music 12 (67%) 7 (78%) 5 (56%) 3 (50%) 5 (83%) 4 (67%)
Interference reading 12 (67%) 5 (56%) 7 (78%) 5 (83%) 4 (67%) 3 (50%)
Text Fragmentation 10 (56%) 4 (44%) 6 (67%) 2 (33%) 4 (67%) 4 (67%)
Resonance 8 (44%) 5 (56%) 3 (33%) 4 (67%) 1 (17%) 3 (50%)
Sound technology 7 (39%) 4 (44%) 3 (33%) 3 (50%) 2 (33%) 2 (33%)
Sound Japanese 5 (28%) 3 (33%) 2 (22%) 1 (17%) 2 (33%) 2 (33%)
 S. Moser / Poetics 35 (2007) 277–300 289

Magda, 28, a visual artist from Toronto, addressed the fragmentation and fade out of text as
follows:
[. . .] Hearing the meaning of the words was a little bit difficult, so again it wasn’t like I was
just listening to what she was saying with the music as the background, I found my mind
focusing more on the background and then a word would float up and I would say, oh, well
that’s an interesting line, you know so, I guess I didn’t, I wasn’t listening to the song in
terms of listening to the text of the song, I was just listening to the song not so much through
time, like from beginning to end, but just, more like a space that you’re in, so you enter the
space at the beginning of the song and you leave the space at the end of the song, but then
you’re just kind of in it, not so much of a journey, I guess.
Resonance: Eight of the interviewees (44%) responded to the question whether there is such as
thing as a ‘‘resonating volume’’ in listening; three interpreted this term concretely in terms of
technical resonance and two in terms of bodily resonance. Four interviewees emphasized that
listening to the song opens up a mental space to them. One interviewee mentioned the holistic
nature of resonance, that is, the interplay of body and mind in listening. Interestingly, only one of
the print-oriented professionals could make sense of the concept of resonance.
Sound technology: Seven of the interviewees (39%) mentioned that the specific sound
technology, in this case the hi-fi and the quality of the CD, has an impact on their oral perception
of language. Listening to a CD differs, for example, from listening to the piece via headphones,
on a vinyl record, or live in a concert hall – a point that was particularly made by proponents of
musical professions.
Sound of Japanese: Five of the interviewees (28%) described concretely how the foreign
language sounded to them. For example, Japanese was perceived by one subject as ‘‘ornamental’’
and by another as ‘‘Asian-Japanese sounding.’’ The Japanese text provided about a third (35%) of
the acoustic key words, phrases or lines (that is words, phrases or lines that received attention for
the way they sound). I expected the interviewees to talk more about the perception of a language
which they do not understand semantically. On the other hand, most of the interviewees (83%)
thought that in general words in songs are often chosen for their sound as well. In the
metaelaboration of the function of the Japanese lines, the majority (65%) of the functions which
were assigned are non-verbal and musical such as the (a) structuration of the song (the Japanese
lines indicate the alternation between chorus and stanzas), (b) the emotionalization (evocation of
feelings and moods) and (c) the use of the voice as an instrument. This functional characterization
suggests that the specific sound of language can be a means of expression and might therefore
serve as a relevant subcategory of text performance.

5.2. Perception of media modality: reading Kokoku

In general, I found fewer categories which refer to the specific modalities of texts in reading,
both in the theoretical literature which informed my questions and in the interviewees’ answers.
The latter might be partly due to the sequence of stimulus presentation which encouraged a
fresher memory of listening. Yet, all interviewees had the printed text at hand and could therefore
have made a topic of its specific media materiality. It seems that indeed in print we tend to forget
about the mediality of language while at the same time being more aware of its structural
dimensions. That is, print seems to foreground the structure of language while masking its
sensory dimension. Print enhances, as Olson (1991) has pointed out, metalinguistic activities,
such as a focus on the morphology of words and the syntax of sentences. This assumption is
290 S. Moser / Poetics 35 (2007) 277–300

Table 2
Reading Kokoku – distribution of subcategories of ‘‘perception of media modality’’ among nationalities and media-
specific professions
All (18) CA (9) A (9) Oral (6) Print (6) Audiovisual (6)
Layout structure 15 (83%) 8 (89%) 7 (78%) 6 (100%) 4 (67%) 5 (83%)
Inner listening 10 (56%) 4 (44%) 6 (67%) 3 (50%) 3 (50%) 4 (67%)
Reading aloud 4 (22%) 3 (33%) 1 (11%) 1 (17%) 2 (33%) 1 (17%)
Text fragmentation 3 (17%) 2 (22%) 1 (11%) 1 (17%) 1 (17%) 1 (17%)
Sound fade out 2 (11%) 2 (22%) 0 (0%) 0 (0%) 2 (33%) 0 (0%)
Interference listening 2 (11%) 1 (11%) 1 (11%) 0 (0%) 1 (17%) 1 (17%)
Reading tempo 1 (6%) 0 (0%) 1 (11%) 0 (0%) 0 (0%) 1 (17%)

backed up by the fact that few of my interview partners referred to the syntax or the narrative
structure of the sung lyrics, whereas linguistic structures played a vital role in the discussion of
the printed piece, especially for print-oriented professionals and for performers who work with
written versions of their performance texts. In the perception of media modality the following
dimensions played a role in reading Kokoku (Table 2).
Layout (visual perception of text): Table 2 shows that 15 of the interviewees (83%) perceived
the layout as a means of structuration and referred to italics and/or paragraphs, some also
mentioned the way the title was set.
Inner listening (inner vocalization): Ten of the interviewees (56%) described that they
assigned a voice to the text or that they practiced the inner vocalization of syllables and words.
Reading aloud (outer vocalization): Four of the interviewees (22%) had read the lyrics aloud,
a result that is in line with the media historical reconstruction of reading as a process of increasing
silence. Three of them were English native speakers, two of whom work in print professions.
Text fragmentation: Three of the interviewees (17%) brought up the selective perception of
lines or words while reading.
Sound fade out: Two of the interviewees (11%) mentioned the total ignorance of sound in
reading, both of them print professionals.
Interference with listening: Two of the interviewees (11%) articulated the self-observation
that having heard the song they cannot move back to reading without imaging the song; this
observation was shared by some of the interviewees as a dimension of listening to songs in
general: once they have a printed text they are interested in they move back and forth between
reading and listening.
Reading tempo: One interviewee (6%) made a topic out of the experience of reading speed in
contrast to the temporal experience of the listening experience, an aspect which was taken up by
four other interviewees in the metaelaboration of print. In general, more statements on the media
perception of print were made in the metaelaboration of the reading experience (see Table 4,
profile ‘‘reading lyrics’’ below).

6. Why do songs have texts?5

If lyrics are often perceived as fragments or even neglected at all, then why do songs have
texts? First hints are given by the exploration of the category metaelaboration of media, which

5
This is a title from an essay by Simon Frith from 1988, whose main insights are summarized in Frith (1996:158–182).
 S. Moser / Poetics 35 (2007) 277–300 291

covers the explicit reflection on cognitive and communicative potentials and limitations of
actions enabled by media. While Table 3 shows an overview of functions that were assigned to
song listening by the interviewees, Table 4 summarizes functions that were assigned to the
reading of lyrics. The subcategories were derived from (a) statements on lyrics/song Kokoku, (b)
other examples of lyrics/songs and (c) lyrics/songs in general. Note that the tables are not
structured in a strictly complementary way. The fact that a listener enacts a cognitive potential
such as ‘‘modal complexity’’ does not necessarily imply that her or his reception is determined by
a cognitive limitation such as ‘‘nonverbal determination.’’ Limits and potentials are options that
can but do not have to occur together in concrete reception processes. The same holds true for
cognitive and communicative functions. The cognitive potential of ‘‘emotionalization,’’ for
example, can but does not have to go along with the communicative potential of ‘‘emotional
communication.’’ In the following, I will elaborate on the most prominently mentioned functions
in each profile.

6.1. Metaelaboration: functional profile ‘‘listening to songs’’

Cognitive possibilities offered by songs are coined by their nonverbal potentials and their lack
of semantic specificity. Table 3 shows that 10 of the interviewees (56%) pointed out songs’
potential for emotionalization. The experience of songs offers an immediacy that often evokes
intense feelings and moods. This affective potential goes hand in hand with the opportunity to
experience texts in the framework of a multimodal complexity, a quality mentioned by seven of
the interviewees (39%). Songs provide recipients with the simultaneous experience of musical
instrumentation and text performance thereby creating complex multimodal text receptions.
Accordingly, cognitive limitations pertain to perceptual limits, a category mentioned by nine
of the interviewees (50%). In songs, texts are often hard to understand and cannot be analyzed in
detail. Other limits are the nonverbal determination of reception, which refers to the fact that the
music often overwhelms the level of verbal meaning construction and tends to direct the
emotional response. This category was mentioned by seven of the interviewees (39%). Moreover,
the musical implementation controls the tempo of text reception, a limit mentioned by five of the
interviewees (28%).
The most prominently mentioned communicative potential of songs is the construction of
collective meaning through participation in shared pop cultural groups, a potential mentioned

Table 3
Functional profile ‘‘listening to songs’’
Cognition Communication
Potentials (‘‘enactment’’) Emotionalization 10 (56%) Collective meaning construction 13 (72%)
Modal complexity 7 (39%) Emotional communication 9 (50%)
Accessibility 6 (33%) Wide distribution 6 (33%)
Semantic openness 5 (28%) Interpretative freedom 3 (17%)
Intensified meaning 4 (22%)
Nonverbal imagination 3 (17%)
Limitations (‘‘selection’’) Perceptual limits 9 (50%) Social distinction of popular cultures 6 (33%)
Nonverbal determination 7 (39%) Conformity / cultural industry 5 (28%)
Tempo of reception 5 (28%) Individualization of reception 2 (11%)
Loss of control 3 (17%)
Disturbance of concentration 3 (17%)
292 S. Moser / Poetics 35 (2007) 277–300

by 13 of the interviewees (72%). Songs evoke cultural frames of reference by either drawing
on thematic universals such as ‘‘love’’ or, as the example Kokoku demonstrates, the
conveyance of complex issues by simple means. As some interviewees pointed out these
shared meanings depend on the respective pop musical genre. While, for example, Kokoku is
typical for the ambiguity of ‘‘avant-garde pop,’’ folk or protest songs tend to tell more
traditional stories. Additionally, songs enable emotional communication (mentioned by nine
interviewees (50%)) and often create the atmosphere for social events such as shared dinners
or parties. Six of the interviewees (33%) made a topic of the wider distribution of songs
through technical reproduction, a communicative potential that facilitates to reach wide and
diverse audiences.
Limits of communication through songs pertain to the social distinctiveness of popular
cultures, a category mentioned by six of the interviewees (33%). Here, interviewees elaborated
on social distinctions such as ‘‘mainstream versus subculture,’’ ‘‘intense listeners versus
superficial audiences’’ and ‘‘professional versus non-professional listeners.’’ Five of the
interviewees (28%) pointed out the conformity of popular music and criticized the stereotypes
and clichés communicated through songs, a critique often accompanied by an emphasis of the
capitalist orientation of the cultural industry. Again, this critique was formulated relative to
particular pop musical genres. While, for example, protest songs of the 1960s were deemed to
pursue a critical agenda, typical pop songs of the pop charts were seen more likely to foster
cultural stereotypes.

6.2. Metaelaboration: functional profile ‘‘reading lyrics’’

Table 4 shows that the prevailing subcategory for the characterization of reading lyrics was
the control of reception, a category mentioned by 12 of the interviewees (67%). Reading enables
the recipient to move back and forth in the text, to repeat the perception of key lines and to self-
determine the tempo of text processing. The text visually presents itself as a structured unit and
can be reconstructed as a coherent whole. This control furthermore enables ‘‘deeper’’
interpretation, a term by which I summarize operations traditionally associated with
hermeneutic procedures of interpretation, such as the reflection on the message of the text
or the interpretation of narrative details. The category ‘‘deeper’’ interpretation was mentioned
by 11 of the interviewees (61%). Reading facilitates the assignment of a personal meaning and
often goes along with textual analysis (metalinguistic activities), a cognitive potential
mentioned by nine of the interviewees (50%). In line with this functional description, 12 of the
interviewees (67%) recognized stylistic means of repetition in the example, and 14 (78%)

Table 4
Functional profile ‘‘reading lyrics’’
Cognition Communication
Potentials Reception control 12 (67%) Critical discourse 4 (22%)
(‘‘enactment’’) ‘Deeper’ interpretation 11 (61%) Relation performer/audience 3 (17%)
Textual analysis (metalinguistic) 9 (50%) Symbolic capital of poetry 2 (11%)
Semantic openness 3 (17%)
Verbal imagination 2 (11%)
Limitations Unimodality 5 (28%) Social distinction of literary system 3 (17%)
(‘‘selection’’) Pressure to interpret 3 (17%) Individualization of reception 1 (6%)
Linguistic barriers 3 (17%)
 S. Moser / Poetics 35 (2007) 277–300 293

commented on the unconventional narrative structure of Kokoku, applying narratological

criteria for stories to the piece such as ‘‘causal connection of events’’ or ‘‘having a beginning and
an end.’’
Cognitive limitations of the mode of reading were seen in the uni-modality of the printed text,
a limit mentioned by five of the interviewees (28%). Some interviewees saw the reading of a song
text as an aesthetic paradox which deprives the song from its genuine sensual, musical pleasures
and results in a reduced experience. Furthermore, interpreting the lyrics through reading evoked
the impression of forced interpretation (mentioned by three of the interviewees (17%)), the
pressure to find meaning in the text. Finally, three of the interviewees (17%) pointed out that
reading can also fail due to linguistic barriers, in case readers do not know the respective
language well enough.
Communicative potentials of reading lyrics were seen in critical discourse, such as the
discussion of songs with friends or on fan websites. This potential was mentioned by four of the
interviewees (22%). Three of the interviewees (17%) pointed out that reading contributes to the
establishment of a relation between performers and audiences. Finally, reading potentially adds
the symbolic capital of high art to lyrics and associates them with the reading of poetry in the
literary system. This communicative potential was mentioned by three of the interviewees (17%).
Yet, this very possibility of ‘‘literary elevation’’ was criticized as the distinctiveness of the
literary system by three of the interviewees (17%) as well. The social expectation to find meaning
in a text through reading was characterized as an inappropriate expectation, even more so since it
exposes song texts to the institution of literary criticism, which applies the normative criteria of
literary analysis to lyrics and tends to question their poetic quality.

6.3. Metaelaboration: oral versus literate listening

The differentiation of media modalities and the functional profiles of reading and listening
raise the question whether qualities of linguistic processing mentioned for song reception can, as
de Kerckhove suggests, be generalized into a specific mode of listening. Given that the
interviewees had read the lyrics before the interview and heard the song under the conditions of
secondary orality, we would expect them predominantly to enact qualities associated with
‘‘literate listening,’’ a mode of listening defined as ‘‘specialized, selective, interested in verbal
meanings, linear, temporal and logo-centric’’ (cf. de Kerckhove, 1995:104).
However, contrary to this strict distinction, many statements suggested that the reception of
songs could imply qualities associated with literacy, such as a specialized focus on lines and
words (exemplified through the categories ‘‘text fragmentation’’ and ‘‘key words’’), as well as
qualities associated with McLuhan’s ‘‘acoustic space,’’ such as the diffuse, immersive quality of
listening. Some of the interviewees referred to this space indirectly in their discussion of the
subcategory ‘‘resonance.’’ Confronted with de Kerckhove’s distinction, seven of the interviewees
(39%) classified their listening experience as ‘‘oral as well as literate,’’ while six of the
interviewees (33%) saw themselves as ‘‘oral’’ listeners (Table 5). Thus, despite the fact that they
had read the lyrics of Kokoku, Table 5 shows that 13 of the interviewees (72%) did not identify
with an exclusively literate mode of listening, nor did the majority identify with an exclusively
oral mode of listening. Interestingly, three of the interviewees from oral professions (50%)
labeled themselves as ‘‘literate’’ listeners. The self-observation of listening modes might be
correlated with professional specializations. None of the print professionals classified their
assessment of Kokoku as exclusively ‘‘oral,’’ none of the (audio-)visual professionals identified
themselves as ‘‘literate’’ listeners only.
294 S. Moser / Poetics 35 (2007) 277–300

Table 5
Self-classification ‘‘oral versus literate listening’’
Professional specialization
Oral (6) Print (6) (Audio-)Visual (6) All (18)
‘‘Oral listener’’ 2 (33%) 0 (0%) 4 (67%) 6 (33%)
‘‘Literate listener’’ 3 (50%) 2 (33%) 0 (0%) 5 (28%)
‘‘Oral and literate listener’’ 1 (17%) 4 (67%) 2 (33%) 7 (39%)

7. Discussion: feedback between data and corporeal media theory

In order to make contact with our social environment, or, in McLuhan’s words, to ‘‘massage’’
our communication partners (cf. McLuhan and Fiore, 1996/1967), we do not have any other
means to mold media materials into signs. The mediality of signs connects cognitive systems
with sociocultural environments. The specific ways in which this ‘‘structural coupling’’ of
cognition and communication takes place define what social interactions are about. Sensual
qualities of semiosis are therefore essential to textual signification, both on the surface
appearance of semiotic materials and on the level of the semantic meaning.
Accordingly, on the basis of a concrete aesthetic experience, the analysis exemplifies the
central assumptions of a corporeal media theory that takes the sensing body as the first and
foremost source of meaning. Media modalities such as print and song seem to determine modes
of the embodied experience of language independently from differences in linguistic
competence. Austrian as well as Canadian experts experience nonverbal, performative
dimensions of pop lyrics and reflect on media-specific functions of reading and listening in
similar ways. This congruence becomes most palpable in the fact that text fragmentation and text
fade out occurred in English-native speakers as well. Differences between perceptions of media
modality and the metaelaboration of media modes occurred mainly between professions but not
between nationalities. The perception of media modality seems to be partly a function of concrete
media enactment and thus of media competence. Here, the expertise of practical know-how and
background knowledge of media systems is crucial: print professionals focused more on
structural dimensions of language, oral professions showed a higher sensitivity for performative
nonverbal aspects of language, and (audio-)visual experts had less interest in language in general.
The role of professional identity may also account for the fact that more interviewees identified
with an unambiguous mode of listening rather than with ‘‘oral and literate’’ listening.
Yet, the majority of non-professional recipients do not have the specialized media competence
of artists or cultural critics to draw from in their experience and interpretation of songs. The self-
classifications of some print professionals as ‘‘literate listeners’’ can therefore not, as Great
Divide theories suggest, be generalized into the assumption that they represent members of an
overall ‘‘literate’’ culture or society. It seems to be more likely that qualities of oral and literate
text processing occur in various combinations relative to background knowledge, aesthetic
genres and media-specific reception situations. The likelihood of a diversity of practices is also
supported by the fact that many subcategories in the functional profiles ‘‘listening to songs’’ and
‘‘reading lyrics’’ were coined by selective emphasis and seemed to have a low degree of
intersubjectivity. Only a few interviewees mentioned most of the listed functions, a fact which
suggests that recipients implement media modalities with different functional emphasis.
Furthermore, the dynamic interplay of reading and listening in the reception of lyrics has to be
 S. Moser / Poetics 35 (2007) 277–300 295

taken into account. Crucial conditions for this interplay seem to be a specific interest in a
particular song and the identification with the content of the song and/or the performer as a star
with a public persona.
In summary, the qualitative content analysis of the expert interviews suggests that the
following dimensions are at work in the reception of lyrics.

7.1. Nonverbal versus verbal dimensions of language

The prominence of text performance and the partial text fade out suggest that nonverbal
dimensions of language such as sound and rhythm play a crucial role in the reception of lyrics.
Accordingly, Langer characterizes the specificity of texts in the framework of musical media
schemata as follows:
When words and music come together in song, music swallows words; not only mere words
and literal sentences, but even literary word-structures, poetry. Song is not a compromise
between poetry and music, though the text taken by itself may be a great poem; song is
music. (Langer, 1953:152)
In songs, the musicality of language, which is constitutive for poetic language in general, is
integrated with music directly and therefore often figures as an instrument. The fact, that many
interviewees emphasized repeatedly that they experience lyrics in nonverbal dimensions has to be
taken into account in studies on the effects of lyrics. Theories on the embodiment of language
suggest that linguistic media offerings are concretely embodied through rhythms (cf. Iverson and
Thelen, 1999), sounds and conceptual metaphors (cf. Danesi, 1993; Ruthrof, 2000). In the same
spirit, Frith (1996:164) argues against a content analysis of song lyrics that focuses on propositional
meanings, noting pointedly: ‘‘Song words are not about ideas (‘content’) but about their
expression.’’ In aesthetic media schemata such as popular songs, this mediality or ‘‘materiality’’ of
language is implemented reflexively through intermedia strategies of linguistic expression. Songs
play with the sound of language, the expressiveness of voice and the rhythm of texts. Zumthor
points out that this level of nonverbal expressiveness is in fact the crucial feature of oral poetic
communication, supporting his argument with a cross-cultural, ethnographic panorama:
In Tibet, the epic song loses its force, so says a singer, if one does not intone the formula
ala-tha-la, stripped of sense, three times repeated: is this an effect of the magic power of
voice? The Fuegians salute the arrival of a guest with joyous vocalizations, abandoning
themselves to this pure pleasure, like French teenagers I used to hear singing a popular
American song without understanding a single word of English. (Zumthor, 1990:128)
Accordingly, the interviewees formulated a range of possible cognitive functions for the
Japanese lines of the example Kokoku, which none of them understood semantically. Moreover,
the qualities of the voice of Laurie Anderson and the background singers of Kokoku as well as the
particular ways in which the lyrics were intonated all represented individual meanings to them in
their own right. Different styles of vocal performance (e.g., reciting versus melodic singing)
might therefore influence cognitive and communicative effects of thematically similar lyrics.

7.2. Text fragmentation versus text coherence

Studies with adolescents suggest that lyrics are often understood either wrongly, only partially
or that they are not perceived at all (cf. Frith, 1996:164; Ballard et al., 1999:477f.). This
296 S. Moser / Poetics 35 (2007) 277–300

prevalence of text fade out and text fragmentation suggests that research on the reception of lyrics
should focus on significant lines or words and the overall relevance of a song rather than asking
for a coherent, metalinguistic interpretation of the lyrics as a whole. Accordingly, Hansen and
Hansen (2000:178) point out that semantic gaps are often filled with schematic ways of
processing.
The focus on textual coherence is derived from a model of literary interpretation which
fundamentally relies on the awareness of the structural features of language (Olson, 1991). These
metalinguistic interpretations indicate a highly specialized approach to texts, one that deploys
skills that are based on technologies of print in the framework of literary expertise. In contrast,
the oral mode of ephemeral perception and selective comprehension corresponds with the
temporalization and the spatial nature of the reception experience mentioned by some
interviewees as a cognitive function of song listening. In accordance with McLuhan’s
assumptions on secondary orality, this temporalization opens an acoustic space of experience
which, despite its selective construction of meaning, does not necessarily result in the
construction of a coherent text. Sound technologies distribute music into various cultural spheres,
a process by which songs become part of a sonic ambience that is permeated by voices and verbal
particles. Thus, lyrics are often part of the musical ‘‘background’’ (Frith, 1996:236) of various
activities and social interactions. What DeNora notes for musical listening practice then may
hold true for the reception of lyrics as well: ‘‘Respondents . . . engage in various bricolage
activities with regard to music, mobilizing, picking and choosing in magpie fashion musical
‘bits’ . . .’’ (DeNora, 1999:44).
It can therefore be hypothesized that different lines and key words of lyrics are selected in
reading and listening. Moreover, given the different status of the conceptual dimension of texts in
reading and listening, the emotional and cognitive rating of lyrics might differ significantly
between these media modes.

7.3. Interplay of lyrics, text performance and music

Communicative functions of ‘‘listening to songs’’ demonstrate that lyrics have to be examined

as part of pop musical genres such as folk songs, protest songs, classical pop songs, (progressive,
alternative, heavy metal, etc.), rock songs, ambient pieces or avant-garde pop songs. This result
concurs with Frith’s detailed analysis of the cultural meanings of voice and music in the
interpretation of songs (Frith, 1996:75–95). Both the style of text performance as well as the
relation between text and music suggests that the reception of lyrics is determined by knowledge
about musical genres. Pop musical media schemata deploy modalities of language in relation to
musical forms of expression. The interplay of text and sound (voice, intonation, instrumentation,
etc.) represents a semiotic dimension in its own right. It corresponds with the question which
social roles a performer embodies with her/his voice in a song and how the audience receives text
and music on background of their aesthetic and sociocultural expectations. In the interviews, for
example, the fact that Laurie Anderson had been associated with the context of artistic
performance contributed to the fact that the lyrics of Kokoku were interpreted as polyvalent and
rather opaque. Neither the textual nor the musical composition of the piece fit neatly into a
standardized production scheme of pop songs. Interviewees who associated Anderson with the
art system and situated her text in the tradition of the historical avant-garde were more tolerant of
his fact than those who did not. Thus, talking about Kokoku helped to gain insight into aesthetic
expectations and value judgments that seem to guide modes of listening. The association of
Anderson with high art stimulated the interviewees to reflect on the distinction between popular
 S. Moser / Poetics 35 (2007) 277–300 297

music and avant-garde and held responsible for the classification of her music as ‘‘avant-garde
pop,’’ ‘‘performance based pop’’ or ‘‘art rock.’’ The discussion of the piece therefore showed that
pop musical genres vary according to textual complexity and the degree to which they deploy
nonverbal dimensions of language.
The central question for an ecologically valid investigation of song texts must therefore be
how verbal meanings are ‘‘enacted’’ by musical means. Sellnow and Sellnow (2001:396)
formulate this programmatically: ‘‘any method designed to analyze music as a rhetorical form
must consider the dynamic interaction between lyrics and score to capture a full meaning of a
message.’’ One hypothesis here would be that different styles of vocal performance (e.g., reciting
versus melodic singing) correlate with different genres (e.g., avant-garde rock versus traditional
pop). Furthermore, the degree of textual opacity in a genre may correlate with the importance of
nonverbal textual features in this genre. Another hypothesis is concerned with the question of
‘‘congruence/incongruence’’ (ibid.:399) as expressed through the subcategory relation text/
music. As the emphasis on the irritating gap between its disturbing text and the soothing music in
Kokoku suggests, the contrast or harmonious interplay of text and music may have different
effects on the reception of lyrics.

7.4. Intermediality: from listening to reading and back

In the reception of lyrics, moving back and forth between reading and listening seems to
represent a third media modality quite distinct from either reading or listening in themselves.
Intermedia reception of lyrics is indicated by distinctions such as ‘‘deeper’’ interpretation in the
metaelaboration of reading as well as the prominence of the subcategory interference with
reading in the perception of the media modality of the song. Here, some interviewees’
commentaries on the order of stimulus presentation, exemplified through Monica’s statement
above, must be interpreted as a valuable third-order observation on the reception situation of the
interview. In everyday life, lyrics do in fact more often occur as songs first; indeed, if they are
important at all, they are in most cases read subsequent to the initial listening experience. This
supplementation of listening through reading supports the constructivist assumption that media
systems gain their specific profiles in relation to each other. In contrast to de Kerckhove’s strict
distinction of oral and literate listening, the reception of a song can imply ‘‘literate’’ qualities
such as a focus on selected verbal meanings as well as ‘‘oral’’ qualities such as the spatial nature
of experience. The rejection of an exclusive oral mode of listening by most proponents of oral
professions might point to the fact that they are aware of this intermedia condition of today’s
linguistic practice.
In general, the fact that the interviewees had read the lyrics before the interview may have
limited the immediacy and emotional quality of their listening experience. It may well be that in
natural listening situations the conceptual meanings of lyrics are most often completely lost to the
power of pure sound. However, as some interviewees’ statements suggested as well, selected
lyrics do present relevance and meaning to their fans in case they intentionally choose to focus on
them and explore them through reading. This perspective of intermediality opens the researcher’s
eyes – and her ears – for the most ‘‘natural’’ – that is, ecologically valid – situation for the
reception of lyrics: the investigative reading of a CD booklet while listening to a newly purchased
CD. It remains an open question what becomes of this situation under the conditions of electronic
music distribution. I assume that when listening to copied CDs or downloaded Mp3 files, listeners
retrieve the lyrics of their favorite songs from the Internet, where a number of well-equipped lyric
databases are now available. In general, the interviewees’ emphasis on intermediality suggests
298 S. Moser / Poetics 35 (2007) 277–300

that reception studies on the impact of lyrics and music should provide recipients with either
printed lyrics, songs or, as a third option, with both of them.

8. Conclusion

Sousou’s cognitivist study, which was criticized at the beginning of this paper, demonstrates in
an exemplary fashion that different paradigms of cognition determine research designs in
empirical aesthetics. By ignoring the embodiment of linguistic meaning, cognitivist projects miss
out on the fact that poetic text reception is mediated through perceptual structures as well as
through communication technologies and their respective uses. Cognitivist assumptions of
reading models should therefore be reexamined in the light of oral communication, the salient
features of which might prove germane to print communication as well. The subcategory of inner
listening/inner vocalization in reading suggests, for example, that nonverbal dimensions of
language may persist in the reception of printed texts.
The explorative study presented deployed constructivist ways to theorize and describe media
modalities of language and their corresponding modes of reception. Its key strategy was to
implement the relational definition of observation as ‘‘distinction’’ through the systematic
comparison of media modes of poetic reception. This comparison has furthermore led to the
formulation of hypotheses on the impact of media modalities and media modes on the reception
of lyrics. The validation of these hypotheses requires experimental research designs, which
introduce media modalities as independent variables and measure their effects on the basis of
‘‘media ecologically’’ valid questionnaires. Experimental designs that investigate the poetic
reception of texts should actively deal with the complexities of mediality in order to avoid the
construction of theoretical artifacts. Print-oriented questions will produce print-oriented answers,
thereby ultimately telling us little about the pleasures of inner listening (print), ‘‘outer’’ listening
(song) and the semantics of sound.

Acknowledgements

I am especially indebted to Margrit Schreier who guided me through the research process with
generosity and precision as well as to the participants of the two conferences and the anonymous
reviewers of Poetics for their instructive commentaries. Stuart Murray’s rhetorical competence
improved the text significantly and made it ready for print. Moreover, the paper is owed to the
expertise of my fabulous interviewees in Austria and Canada.

References

Alkalay-Gut, K., 2000. Literary dialogues: Rock and Victorian poetry. Poetics Today 21 (1), 33–60.
Ballard, M.E., Dodson, A.R., Bazzini, D.G., 1999. Genre of music and lyrical content: expectation effects. The Journal of
Genetic Psychology 160 (4), 476–487.
Biakolo, E., 1999. On the theoretical foundations of orality and literacy. Research in African Literatures 30 (2), 42–66.
Bogner, A., Menz, W., 2002. Das theoriegenerierende Experteninterview. Erkenntnisinteresse, Wissensformen, Inter-
aktion. In: Bogner, A., Littig, B., Menz, W. (Eds.), Das Experteninterview. Theorie, Methode, Anwendung. Leske +
Budrich, Opladen, pp. 33–70.
Burns, L., Lafrance, M., 2001. Disruptive Divas. Feminism, Identity & Popular Music. Routlege, London/New York.
Chandler, D., 1994. Biases of the ear and eye: ‘The Great Divide’ theories, phonocentrism, graphocentrism & logcentrism
<http://www.aber.ac.uk/media/Documents/litoral/litoral.html> (25.3.2003).
Danesi, M., 1993. Vico, Metaphor and the Origin of Language. Indiana University Press, Bloomington/Indianapolis.
 S. Moser / Poetics 35 (2007) 277–300 299

de Kerckhove, D., 1995. The Skin of Culture. Investigating the New Electronic Reality, Somerville, Toronto.
DeNora, T., 1999. Music as a technology of the self. Poetics 27, 31–56.
Finnegan, R., 1977. Oral Poetry. Its Nature, Significance, and Social Context. Cambridge University Press, Cambridge.
Frith, S., 1996. Performing Rites. On the Value of Popular Music. Harvard University Press, Cambridge.
Groeben, N., Vorderer, P. 1988. Leserpsychologie. Lesemotivation – Lektürewirkung, francke, Weinheim.
Hansen, Ch.H., Hansen, R.D., 2000. Music and music videos. In: Zillman, D. (Ed.), Media Entertainment: The
Psychology of Its Appeal. Lawrence Erlbaum Associates, Mahwah, pp. 175–196.
Havelock, E., 1991. The oral-literature equation: a formula for the modern mind. In: Olson, D.R., Torrance, N. (Eds.),
Literacy and Orality. Cambridge University Press, Cambridge, pp. 11–27.
Iverson, J.M., Thelen, E., 1999. Hand, mouth and brain: the dynamic emergence of speech and gesture. In: Núñez, R.,
Freeman, W.J. (Eds.), Reclaiming Cognition. The Primacy of Action, Intention and Emotion. Journal of Conscious-
ness Studies 6 (11/12), Imprint Academic, Thorverton, pp. 19–40.
Langer, S.K., 1953. Feeling and Form. A Theory of Art Developed from Philosophy in a New Key. Macmillan, New York.
McLuhan, M., 1962. The Gutenberg Galaxy. The Making of Typographic Man. University of Toronto Press, Toronto.
McLuhan, M., 1997/64. Understanding Media. The Extensions of Man with a New Introduction by Lewis H. Lapham. The
MIT Press, Cambridge/London.
McLuhan, M., Fiore, Q., 1996/1967. The Medium is the Massage. An Inventory of Effects. Ginko Press, Corte Madera.
Middleton, R., 2000. Popular music analysis and musicology: bridging the gap. In: Middleton, R. (Ed.), Reading Pop.
Approaches to the Textual Analysis of Popular Music. Oxford University Press, Oxford/New York, pp. 104–121.
Núñez, R., Freeman, W.J. (Eds.), 1999. Reclaiming cognition. The primacy of action, intention, and emotion. Journal of
Consciousness Studies 6 (11/12), Imprint Academic, Thorverton.
Olson, D.R., 1991. Literacy as metalinguistic activity. In: Olson, D.R., Torrance, N. (Eds.), Literacy and Orality.
Cambridge University Press, Cambridge, pp. 251–270.
Olson, D.R., Torrance, N., 2001. Conceptualizing literacy as a personal skill and as a social practice. In: Olson, D.R.,
Torrance, N. (Eds.), The Making of Literate Societies. Blackwell Publishers, Massachusetts/Oxford, pp. 3–18.
Ong, W.J., 2002/1982. Orality and Literacy. The Technologizing of the Word. Routledge, London/New York.
Peskin, J., 1998. Constructing meaning when reading poetry: an expert–novice study. Cognition and Instruction 16 (3),
235–263.
Roth, W.M., 2002. From action to discourse: the bridging function of gestures. Cognitive Systems Research 3/3, 535–556
(Special Issue: Situated and Embodied Cognition).
Ruthrof, H., 2000. The Body in Language. Cassell, London/New York.
Schmidt, S.J., 1991/1980. Grundriß der Empirischen Literaturwissenschaft. Mit einem Nachwort zur Taschenbuchaus-
gabe, Suhrkamp, Frankfurt/M.
Schmidt, S.J., 1996. Media: The coupling of cognition and communication. In: Kreuz, R.J., MacNealy, M.S. (Eds.),
Empirical Approaches to Literature and Aesthetics. Ablex Publishing Corporation, Norwood, pp. 35–49.
Schmidt, S.J., 2000. Kalte Faszination. Medien, Kultur, Wissenschaft in der Mediengesellschaft, Velbrück, Weilerswist.
Schön, E., 1987. Der Verlust der Sinnlichkeit, oder, Die Verwandlungen des Lesers: Mentalitätswandel um 1800. Klett-
Kotta, Stuttgart.
Schreier, M., Rupp, G., 2002. Ziele/Funktionen der Lesekompetenz im medialen Umbruch. In: Groeben, N., Hurrelmann,
B. (Eds.), Lesekompetenz. Bedingungen, Dimensionen, Funktionen. Juventa Verlag, Weinheim/München, pp. 251–
274.
Sellnow, D., Sellnow, T., 2001. ‘The illusion of life’ rhetorical perspective. An integrative approach to the study of music
as communication. Critical Studies in Media Communication 18 (4), 395–415.
Sousou, D.S., 1997. Effects of melody and lyrics on mood and memory. Perceptual and Motor Skills 85, 31–40.
Stockwell, P., 2002. Cognitive Poetics. An Introduction. Routledge, London/New York.
Storey, J., 2003. Inventing Popular Culture. Blackwell Publishing, Oxford.
Stratton, V.N., Zalanowski, A.H., 1994. Affective impact of music vs. lyrics. Empirical Studies in the Arts 12 (2), 173–
184.
Triebel, A., 2001. The roles of literacy practices in the activities and institutions of developed and developing countries.
In: Olson, D.R., Torrance, N. (Eds.), The Making of Literate Societies. Blackwell Publishers, Massachusetts/Oxford,
pp. 19–53.
Varela, F.J., Thompson, E., Rosch, E., 1991. The Embodied Mind. Cognitive Science and Human Experience. The MIT
Press, Cambridge/London.
Whitten, S., Graesser, A., 2002. The emotional impact of song lyrics depends on the reader and the type of text. In:
Proceedings of the VIII Congress of the International Society for the Empirical Study of Literature and Media, Pécs
21–24 August 2002, <http://www.arts.ualberta.ca/igel/IGEL2002/Proceedings.htm> (15.7.2005).
300 S. Moser / Poetics 35 (2007) 277–300

Ziemke, A., 2003. What’s that thing called embodiment? In: Proceedings of the 25th Annual Meeting of the Cognitive
Science Society, Erlbaum, Mahwah, NJ, pp. 1305–1310.
Zumthor, P., 1990. Oral Poetry. An Introduction (Kathryn Murphy-Judy, Trans.) University of Minnesota Press,
Minnesota (Foreword by Walter J. Ong).

Sibylle Moser is a lecturer of media studies at the Universities of Vienna and Innsbruck. She received a PhD in the
methodology of literary and media studies at the University of Vienna in 1999. Her research focus lies on cognition and
aesthetics, systems theory and constructivism, empirical studies in literature and gender studies. Currently, she is working
toward her postdoctoral project ‘‘Cognition of Aesthetic Media: Laurie Anderson’s Pop Lyrics,’’ a transdisciplinary study
which focuses on the embodiment of language in Laurie Anderson’s work. Website: www.sibyllemoser.com.