Conference Agenda

The ability of music to evoke specific extramusical semantic content has long been a topic of rich debate among music theorists. While behavioral research has established that music evokes many kinds of extramusical thought, including fictional narratives and autobiographical memories, the contents of these thoughts were long believed to be driven by the idiosyncrasies of individual listeners rather than systematic relationships to musical features. Recent research, however, indicates that fictional narratives evoked by instrumental music show similarities in semantic content for within-culture participants, and that the content of these narratives appears to be driven by musical features such as tension, repetition, timbre, dynamics, and more.

I conducted a series of experiments investigating whether personal experiences with music in specific contexts might lead to a similar intersubjectivity in music-evoked autobiographical memories (MEAMs) reported by people with broadly similar backgrounds. University undergraduates (n=233) gave free-response descriptions of memories evoked by short clips of popular songs in three genres (pop, hip-hop, and country) from time periods matching the listeners’ primary and secondary reminiscence bumps. I used tools from natural language processing to examine the semantic contents of the memories, revealing effects of both genre and release date. A k-means clustering analysis in semantic embedding space showed that memories for country songs tended to be similar regardless of song release date, while memories for pop and hip-hop songs were similar within each time period. A Latent Dirichlet Allocation (LDA) topic model confirmed these results, revealing significant differences in the distribution of certain topics with respect to genre and time period.

Two additional experiments suggest that these results were not solely due to similarities in lyrical content between songs of the same genre/time period. Participants described memories evoked by key words from the lyrics (n=122) and short narrative summaries of the lyrics (n=129), and the resulting memories did not follow the same patterns of semantic similarity as those evoked by the corresponding songs. Altogether, these results suggest ties between: (1) musical features such as emotionality, genre, singability, and groove, (2) the listening behaviors they afford, and (3) the semantic content of MEAMs recounting those behaviors.

Ever since the affective turn took hold of the humanities and social sciences beginning in the mid-1990s, music-theoretical scholarship has had to grapple with new ways of thinking about some of its most fundamental attributes: structure, systematicity, category, sign, and narrative. Affect theory, by stark contrast, is said to resist all these things—what Roger Mathew Grant (2020) describes as its “negative ontology”—placing emphasis instead on forms of feeling that reside outside the purview of language, signification, and conscious reflection all together.

A prime example of music scholarship’s embracing of affect is the Arab concept of ṭarab (Figueroa 2022; Hirschkind 2006; Racy 2003; Seeman 2025; Sprengel 2019). Most often described in English as a form of “musical ecstasy,” ṭarab is largely understood by these scholars as a heightened state of affective engagement that eludes semantic explication and emotional specificity. In this view, “emotion” is seen as a binding force for capturing the ecstatic state’s elusive boundaries and resistance to conceptual order.

This paper aims to tell a different story about ṭarab through the words and lived experiences of Arab Americans. I argue that Arab Americans (particularly those living in Dearborn, Michigan and Brooklyn, New York) experience ṭarab in ways that challenge the idea of affect as lacking structure and opposing emotion. Supplementing musical analysis with ethnographic methodologies, I make two critical interventions. First, in response to Grant’s (2020) call for a renewed attention to the objects that generate affects in subjects, I reveal a distinct homology between musical (particularly rhythmic) structure and the experiential characteristics of ṭarab. Drawing on the work of Eugenie Brinkema (2014), I discuss the ecstatic effects of what I call “rhythmic-temporal disruptions” to show how they function as a direct sonic materialization of the ṭarab experience. My second intervention makes a case for the association of rhythmic-temporal disruptions, entrainment, and nostalgia. I contend that such forms of entrainment (e.g., participatory clapping) are not only inscribed within the collective muscle memory of its practitioners, but when practiced in diasporic contexts, activate a form of kinesthetic memory that renders nostalgia an accessible emotion for those of the Arab diaspora.

De Souza’s (2015) reframing of texture as dynamic and emergent synthesizes Berry’s (1987) numerical approach to textural independence with principles of auditory stream segregation (Bregman 1990) and melodic integration (Huron 1989; Duane 2015). Textural dynamics play an important role in ensemble coordination, especially during phrase boundaries.

This paper examines how textural dynamics affect ensemble singers’ listening and coordination, particularly at phrase boundaries where greater temporal and expressive freedom heightens coordination demands (Bishop et al. 2019). Keller’s theory of prioritized integrative attending argues that performers predominantly focus on their own part while flexibly monitoring the group in response to changing musical needs (2008). We propose that changes in texture at phrase boundaries shape these listening and coordination behaviors.

In a mixed-methods study, 18 vocal trios performed two madrigals. After each performance, singers indicated in real time whom they were listening to most, provided descriptions of their listening and coordination, and completed surveys assessing their use of coordination strategies and sense of agency. These data were mapped onto textural analyses based on Berry’s and De Souza’s models to investigate correlations between predicted textural streams and observed performer behaviors.

Findings reveal connections between predicted textural streams and performer listening behaviors and suggest that homophonic phrase boundaries encourage cross-part listening and greater collective agency. In an analysis of a King’s Singers performance, phrase boundaries affected the relationship between performer listening and textural streams. Moving into the final cadence, singers shifted from listening within their respective streams to listening across streams, ending in a global listening mode for the final chord. This pattern of listening allowed singers to shift between performance goals of unifying expression within their streams, attending to key voices when approaching a phrase boundary, and coordinating timing and text shaping in the final chord. Across groups, singers reported listening to each other more, focusing on their own parts less, and experiencing stronger ensemble integration when phrase boundaries were predominantly homophonic.

By examining how performers allocate their attention, these findings offer new analytical perspectives on emergent textural streams, revealing the influence of textural dynamics on ensemble coordination and performers’ subjective experiences.