Table of Contents
1. Introduction: Film Music Narratology
This book positions itself at the intersection of film musicology and narrative theory. It acknowledges the rapid development of film music studies while grounding its approach in narratological concepts from the 1970s-1990s. The central aim is to apply a refined toolkit of narrative levels—extrafictional, diegetic, nondiegetic, and metadiegetic—to analyze how film music functions not just as an emotional underscore, but as an active agent in storytelling. The author, Guido Heldt, frames the work as a tentative contribution in a dynamic field, seeking to address persistent questions about music's narrative status.
2. The Conceptual Toolkit: Music and Levels of Narration
This chapter establishes the core theoretical framework for analyzing film music through narratology.
2.1 Fictional Worlds and the Filmic Universe
Explores the foundational concept of the diegesis—the story world inhabited by the characters—and its boundaries. It distinguishes between what is internal to the fictional reality and what is external, setting the stage for understanding where music is positioned relative to this world.
2.2 The 'Historical Author': Extrafictionality and the Title Sequence
Examines music that exists outside the fictional narrative altogether, such as title sequence music. This music is attributed to the "historical author" (the composer/filmmaker) and speaks directly to the audience, establishing tone and genre expectations before the story proper begins.
2.3 Extrafictional Narration and Audience Address
Discusses instances where music breaches the "fourth wall," directly acknowledging or addressing the audience. This creates a layer of communication separate from the character's world.
2.4 Nondiegetic and Diegetic Music
The heart of the theoretical model, delving into the traditional binary and its complexities.
2.4.1 Narratology, the Diegesis and Music – Some Considerations
Critically reviews the application of the diegetic/nondiegetic distinction from literary narratology to film music, noting its limitations and the need for a more nuanced model.
2.4.2 Nondiegetic Music and Narrative Agency
Argues that nondiegetic music (the "score" unheard by characters) is not neutral but is a form of narration. It is an agent that comments, emphasizes, and interprets the story, often aligned with an "implied author" or overarching narrative voice.
2.4.3 Diegetic Music: Storyworld Attachment and Narrative Agency
Analyzes music originating within the story world (e.g., from a radio). While anchored to the diegesis, it can still serve narrative functions beyond mere realism, influencing mood and character perception.
2.4.4 Diegetic Commentary and the Implied Author
Explores how diegetic music can be ironically or pointedly used to comment on the action, a technique where the choice of source music implies an authorial intelligence behind the scene's construction.
2.4.5 Diegetic Music: Further Options
Examines other functions, such as establishing setting, period, or character identity through their musical tastes.
2.4.6 Transitions, Transgressions and Transcendence
Investigates borderline cases: "displaced diegetic" music (seemingly sourced but ambiguously placed) and "supradiegetic" music (as in musicals, where characters sing in a heightened reality). These are "steps across the border" that challenge simple categorization.
2.5 Music on My Mind: Metadiegetic Narration and Focalization
Introduces the concept of "metadiegetic" music—music that represents a character's internal state, thoughts, memories, or fantasies (e.g., a character's mental soundtrack). This ties music closely to focalization, showing the story through a specific character's subjective experience.
3. Breaking into Song? Hollywood Musicals (and After)
Applies the framework to the musical genre, where the boundaries between diegetic and nondiegetic are inherently fluid and often deliberately transgressed.
3.1 Supradiegesis
Defines the "supradiegetic" space of the musical number, a realm beyond ordinary reality where characters express emotion through song and dance.
3.2 Superabundance: Top Hat and the 1930s
Analyzes early musicals where the shift into song is often abrupt and unmotivated by the plot, celebrating pure spectacle and performance.
3.3 The Classical Style
Examines the integrated musical (e.g., Singin' in the Rain), where numbers are more plausibly motivated by the plot (putting on a show, rehearsing), creating a more seamless, though still distinct, narrative level.
3.4 Transcendence Lost and Regained
Traces the evolution of the musical after its classical peak, where later films either parody the convention or struggle to integrate it into more realist narratives.
3.5 The Next-to-Last Song
Looks at self-reflexive or deconstructive uses of the musical form in films like Dancer in the Dark, where the tension between musical fantasy and harsh diegetic reality is central to the drama.
4. Things That Go Bump in the Mind: Horror Films
Explores how horror films use music to manipulate audience perception and create fear, often blurring the line between objective threat and subjective anxiety.
4.1 Of Implied Authors and Implicit Contracts
Discusses the narrative "contract" in horror: the audience accepts that nondiegetic music (stingers, creepy atmospheres) signals danger or the presence of the monstrous, even though characters cannot hear it. This music is the voice of an "implied author" guiding—or misleading—audience expectation.
4.2 ... and Thirteen Examples
Provides detailed analyses of specific scenes from horror films, demonstrating how music operates across different narrative levels to build suspense, reveal character psychology, or deliver shocks.
5. Beyond the Moment: Long-range Musical Strategies
Examines how music creates narrative meaning not just in individual scenes but across entire films through leitmotifs, thematic development, and structural repetition.
5.1 Music and Memory in Once Upon a Time in America
A detailed case study of Sergio Leone's film, analyzing how Ennio Morricone's score uses recurring themes to connect past and present, mirroring the protagonist's fragmented memory and sense of loss. It traces the evolution of this technique from Leone's earlier Westerns.
5.2 Life's Troubled Bubble Broken: Musical Metalepses in The Truman Show
Analyzes how music in The Truman Show operates on multiple, conflicting levels: as part of the diegetic artificial world of Seahaven (composed by Christof), as a nondiegetic score for the television audience, and as a tool that occasionally hints at the truth to Truman, creating a narrative metalepsis (a breach between levels).
5.3 Far from Heaven, Breakfast at Tiffany's, and Hollywood Melodrama
Explores the use of pre-existing music and score in melodrama. It examines the "retrospective prolepsis" in Far from Heaven, where an anachronistic 1950s-style score comments on the film's 1950s setting from a contemporary, critical perspective. Compares this to the function of "Moon River" in Breakfast at Tiffany's.
6. The Future's Not Ours to See: Outlook
The concluding chapter reflects on the provisional nature of the study in a rapidly evolving field. It suggests that the narratological framework offered is a tool for ongoing analysis, not a final word, and hints at future directions for research at the intersection of music, film, and narrative theory.
7. Core Insight & Analysis
Analyst's Perspective: A Four-Step Deconstruction
Core Insight: Heldt's work is a crucial, if belated, intervention that successfully weaponizes classical narratology to demystify film music's narrative agency. Its core value isn't in new data, but in a powerful new organizing framework. It moves beyond the simplistic "diegetic/nondiegetic" binary—a staple of film studies since Claudia Gorbman's seminal work—and introduces a multi-level model (extrafictional, diegetic, metadiegetic, supradiegetic) that finally accounts for music's complex narrative positioning. This is akin to applying a more sophisticated neural network architecture (like a Transformer's attention mechanism) to a problem previously tackled with simpler models, revealing deeper patterns of meaning.
Logical Flow: The book's logic is impeccably structuralist. It first builds the theoretical machine (Chapter 2), then runs different filmic "genres" through it to test its outputs (Chapters 3 & 4), and finally examines complex, longitudinal cases to demonstrate its analytical power (Chapter 5). This methodology mirrors rigorous computational analysis: define your parameters, run standard tests, then stress-test with edge cases. The choice of musicals and horror films is brilliant—they represent the two poles of narrative integration, from overt, celebratory boundary-breaking to covert, manipulative ambiguity.
Strengths & Flaws: The primary strength is the framework's explanatory power. Concepts like "metadiegetic music" provide a precise vocabulary for phenomena critics have long described vaguely as "subjective scoring." The extended analyses (e.g., Once Upon a Time in America) are masterclasses in close reading. However, the flaw is its hermetic classicism. The analysis is deeply rooted in auteur-driven, score-centric cinema (Leone, Sirk, classical Hollywood). It has less to say about the post-modern, sample-based, pop-saturated soundscapes of contemporary film (e.g., the work of Edgar Wright or the use of music in the Guardians of the Galaxy franchise), or the narrative role of sound design that blurs into music. It also, as the author admits, leans heavily on 70s-90s theory, somewhat sidelining more recent cognitive and embodied approaches to film music (like those advocated by Annabel J. Cohen or published in journals like Music and the Moving Image).
Actionable Insights: For scholars: Adopt and adapt this leveled model as a baseline for analysis, but expand its scope. Use it to analyze interactive narratives (video games), where the player's agency adds another narrative layer, or streaming series, where musical themes evolve over dozens of hours. For practitioners (composers, directors): Use this framework as a conscious tool for script breakdown. Intentionally assign narrative levels to musical cues during pre-production. Ask: "Is this music the character's memory (metadiegetic), the director's irony (extrafictional), or the world's ambiance (diegetic)?" This deliberate planning can elevate a score from mere accompaniment to foundational narrative structure. The field's next step must be a synthesis: merging Heldt's precise narratological mapping with the cognitive insights of how audiences actually process these levels of musical information.
8. Technical Framework & Analysis
8.1 The Narrative Level Model
The core analytical framework can be formalized as a hierarchical set of narrative levels $N$, where each level $L_i$ is defined by its relationship to the diegetic story world $D$ and its communicative address.
- Extrafictional ($L_e$): $L_e \cap D = \emptyset$. Music originates from the historical author(s) and addresses the audience directly. Function: Establish contract, credit sequence.
- Nondiegetic ($L_n$): $L_n \cap D = \emptyset$, but $L_n$ comments on $D$. Attributed to an implied author/narrator. Function: Commentary, emphasis, emotional guidance.
- Diegetic ($L_d$): $L_d \subset D$. Music has a source within the story world. Function: Realism, setting, character.
- Metadiegetic ($L_m$): $L_m \subset \text{Subjectivity}(C)$ where $C$ is a character. Music represents a character's internal state. Function: Focalization, subjectivity, memory.
- Supradiegetic ($L_s$): $L_s \supset D$ or $L_s$ is a transformation of $D$. A heightened, performative reality. Function: Spectacle, emotional expression (musicals).
Transitions and Ambiguity: The model's power lies in analyzing transitions, where $Music(t)$ at time $t$ belongs to level $L_i$, and at $t+1$ belongs to $L_j$. A smooth transition (e.g., orchestral swell into a diegetic radio song) can create cohesion. A jarring or ambiguous transition can create irony, confusion, or metalepsis.
8.2 Analysis Framework Example: A Scene Breakdown
Case: Analyzing the opening of The Truman Show.
- 00:00-01:00 (Title Sequence): Grand, optimistic orchestral music. Level: Extrafictional ($L_e$). Establishes the film's "show" premise for the audience.
- 01:01-05:00 (Truman's Morning): Light, repetitive, slightly synthetic music. Level: Ambiguous. Presented as nondiegetic ($L_n$) but is later revealed to be Diegetic ($L_d$)—composed by Christof for the show. This is a deliberate misdirection.
- 05:01 (Spotlight Falls): A dramatic, dissonant sting. Level: Nondiegetic ($L_n$) for the film's audience, but potentially Metadiegetic ($L_m$) for Truman, representing his sudden spike of anxiety and suspicion.
Result: The analysis reveals how the score constructs two simultaneous narratives: the seamless, cheerful world of Seahaven (diegetic score) and the omniscient, manipulative perspective of the show's creator (extrafictional/nondiegetic), with Truman's subjectivity caught between them.
9. Future Applications & Directions
The narratological framework outlined by Heldt has significant potential beyond traditional film analysis:
- Interactive Media & Video Games: The model must expand to include a Ludic Level, where music responds to player action. Is the combat music nondiegetic (dramatic emphasis) or a diegetic part of the game world's logic? Games like Hellblade: Senua's Sacrifice use voices as metadiegetic music, representing the protagonist's psychosis.
- Extended Reality (XR): In VR/AR narratives, the spatialization of sound becomes a narrative tool. A character's theme that moves with the user occupies a different level than ambient world music. This requires a Spatio-Diegetic parameter.
- Algorithmic & AI-Generated Scoring: As AI begins to compose adaptive scores (e.g., for dynamic video games or personalized streaming content), the narrative level must be a key input parameter. An AI needs to "understand" whether to generate $L_d$ (source music) or $L_m$ (subjective music) based on scene context.
- Transmedia Storytelling: How does a musical theme function when it moves from a film ($L_n$) to a streaming series ($L_n$) to a theme park ride ($L_e$ or $L_d$)? The framework can track narrative consistency and audience perception across platforms.
- Integration with Cognitive Science: The next frontier is empirical validation. Using EEG or fMRI, researchers could test whether audiences show different neural responses to music identified as $L_m$ (subjective) vs. $L_n$ (commentary), potentially quantifying the "focalization effect."
10. References
- Heldt, G. (2013). Music and Levels of Narration in Film: Steps Across the Border. Intellect.
- Gorbman, C. (1987). Unheard Melodies: Narrative Film Music. Indiana University Press.
- Chatman, S. (1978). Story and Discourse: Narrative Structure in Fiction and Film. Cornell University Press.
- Genette, G. (1980). Narrative Discourse: An Essay in Method. Cornell University Press.
- Kassabian, A. (2001). Hearing Film: Tracking Identifications in Contemporary Hollywood Film Music. Routledge.
- Neumeyer, D. (2015). Meaning and Interpretation of Music in Cinema. Indiana University Press.
- Cohen, A. J. (2013). "Film Music from the Perspective of Cognitive Science." In The Oxford Handbook of Film Music Studies. Oxford University Press.
- Mera, M., & Sadoff, R. (Eds.). (2019). The Routledge Companion to Screen Music and Sound. Routledge.
- Whittington, W. (2020). "Sound and Narrative in the Digital Age: From Diegesis to Design." Journal of Sonic Studies, 20.
- International Association for Cognitive Musicology (IACM). (n.d.). Research Publications. Retrieved from [Hypothetical authoritative database link].