Research data-as-practice

We approached sonification as a process for conveying research data through sound. However, we aimed to understand how sonification could facilitate more open-ended interpretations of research findings rather than serve as a means of translating knowledge. Creating multiple sonifications with the same data set allowed us to reflect on how research data is inherently subjective, personal, interpretive and messy. Sonification can accentuate this productively, drawing out how data is always already ‘cooked’.¹ Researchers can apply sonification to corroborate findings, make inconsistencies audible, and extract new insights into a phenomenon. To this end, we engaged with our research data as practice, not as an object.

We were influenced by Nick Couldry’s ‘media as practice’ framework. This media sociology approach, as Couldry explains it, is “concerned with the specific regularities in our actions related to media and the regularities of context and resources that make certain types of media-related actions possible or impossible, likely or unlikely” (2012 p. 33). Having arrived at Couldry’s framework through a consideration of Miriam Posner’s and Lauren Klein’s proposition of understanding data as media², we were interested in extending their work and giving thought to the ways critical media studies can enrich the field of data studies. We were particularly struck by Posner’s and Klein’s observations about how critical media studies approaches could allow us to engage “data’s underexplored textures” in conversation with “our own assumptions about what data should be or do”² (p. 5). In turn, we considered how, as researchers, what we do with data and the disciplinary and methodological conventions and norms chosen to analyze and interpret the data would yield particular data practices. Could a data-as-practice approach direct researchers to consider how research data is always coming into being rather than as finitely decipherable or knowable?

In Sensing Sound: Singing and Listening as Vibrational Practice, musicologist Nina Sun Eidsheim reimagines the ways we think about sound and sound-related practices like music and listening. Eidsheim’s proposition that sound, listening and music be reconceived from a knowable object to “an unfolding phenomenon that arises through complex material interactions”³ (p. 2) was crucial to our thinking. We came to sonification through critical data studies, and Eidsheim’s work gave us an expansive way to think about research data, data studies, and sound data. Eidsheim’s revisioning of sound energized us, and we similarly sought to reconceive research data’s ontological status as a knowable object.

The way we engaged with our textual data in this study emerges from an evolving set of experimental applications that emphasize sound and somatic comprehension. These practices continue to yield fresh insights for digital humanities. Hayles observes that the so-called “second wave” of digital humanities emphasizes “attention to complexity, medium specificity, historical context, analytical depth, critique and interpretation”⁴ (p. 26). Her analysis of the recursive structures in literary texts, for example, reveals the musicality of prose and poetry, uncovering the rhythmic patterns and anomalies that contribute to a deeper understanding of an author’s style and thematic emphasis⁴ (p. 245). Similarly, our methods aimed to demonstrate how sonification can extend beyond translation or representation. By inviting researchers to reconsider the fundamental nature and epistemological foundations of their data, as well as the discourses that inform us about data, we sought to highlight the implications of sonification as a critical and interpretive tool.

As a critical and experimental method, sonification enabled us to parse how our research data is subjective, multifaceted and even contradictory.³ After all, as Eidsheim reminds us, “[e]very listening practice and its attendant theory arises from and reinforces a particular set of values”³ (p. 5). How might listening to research data reveal and complicate some of our taken-for-granted assumptions and hang-ups about the appropriate ways to engage with, represent and disseminate research findings? We have presented our research data as in flux, subjective, and above all else, interpretive. But isn’t all research data to a certain extent?

Research data-as-practice asks us to consider research data and the attendant networks of institutions and technologies that enable and constrain the possibilities for ‘research data.’ This includes the scripting of ‘best practices,’ which software and tools are deemed appropriate, and how expertise is understood and granted. In other words, research data-as-practice can train our focus on understanding our research data in relation to the histories, cultures and contexts that give rise to it and how we choose to engage with and create meaning from it.² Such a process can open a critical space to ask important questions about the institutional, disciplinary and cultural norms that dictate evolving standards and models over what constitutes research data and how interdisciplinary and experimental research ought to be managed (or not).

---