(1) Overview
Repository location
Context
Marion Scott exemplifies persistent challenges in music historiography: systemic undervaluation of women’s contributions (Adams, 2011) despite abundant archival evidence of their central roles in cultural production. Scott’s career aligns with broader patterns where women achieved professional recognition while facing institutional barriers, challenging assumptions about male dominance in the music profession during the transformative early twentieth-century period (Fuller, 2018). These challenges reflect deeper epistemological issues about whose knowledge and networks are preserved and deemed worthy of scholarly attention – concerns central to feminist data science approaches that interrogate how data collection can perpetuate or challenge existing power structures (D’Ignazio & Klein, 2020).
The “network turn” in humanities scholarship has provided new frameworks for understanding cultural production as fundamentally relational rather than individually focused (Ahnert et al., 2020), aligning with musicological movements beyond score-based analysis toward understanding music as performance and social practice (Cook, 2013). Musical culture emerges through complex webs of collaboration, mentorship, and institutional relationships (Crossley et al., 2014). Database-driven methods have proven particularly effective in revealing hidden patterns, as Willen shows with Stockholm’s musical life through network analysis mapping interactions between musical actors and venues (Willen, 2024), while research on European ladies’ salon orchestras demonstrates how large datasets can meaningfully address questions of social class and gender that traditional methods struggle to illuminate (Koivisto, 2019).
However, musicology has been notably slower than other humanities disciplines in adopting computational approaches, creating significant gaps in understanding how musical culture operates through networks rather than individual achievement. While automated approaches for extracting relational data from unstructured texts have proven successful in other domains (Graham et al., 2020), historical musical sources require careful manual curation due to the nuanced nature of professional relationships and specialized knowledge needed to interpret musical contexts. This dataset therefore serves both as a resource for understanding specific historical figures and as a methodological intervention in how music historians approach women’s cultural influence when traditional metrics may not fully capture their contributions. Specifically, the dataset enables investigation of key research questions including: How did women musicians circumvent institutional barriers through informal networks? What role did Scott play as a network broker between marginalized composers and mainstream institutions? How did the Society of Women Musicians function as an alternative professional network to male-dominated institutions?
(2) Method
Steps
Archival Survey and Material Selection
The project conducted a comprehensive survey across three major archival collections at the Royal College of Music to capture Marion Scott’s multifaceted role as a network broker in London’s musical community. We examined: (1) the Marion Scott Collection (87 catalogue entries) containing personal correspondence and papers; (2) the Society of Women Musicians archive (526 catalogue entries), where Scott served as co-founder and key organizer; and (3) the Royal College of Music Magazine archive (264 entries from issues where Scott served as editor). From this comprehensive survey, 76 documents were strategically selected for the dataset based on their evidence of Scott’s role as a broker connecting different musical communities – from marginalized composers to institutional gatekeepers, from women musicians to mainstream publications. This approach adapted established historical data extraction principles to meet network analysis requirements (Whitfield, 2023).
Relationship Identification and Documentation
Structured collection protocols focused on documenting collaboration patterns and mutual support among women musicians – relationships that traditional scholarship has only recently begun to investigate (Mathias, 2021; Hamer, 2021; Head & Wollenberg, 2024). Following insights that research proves most revealing when examining what is hidden, excluded, or marginalized (Barget & Schreibman, 2025), each document was systematically examined for evidence of professional, social, and collaborative relationships. Particular attention was given to interactions challenging canonical assumptions, including editorial advice, concert organization, committee work, mentorship, advocacy relationships, social gatherings with professional implications, correspondence about professional matters, collaborative performances, and institutional affiliations.
Data Structure and Classification System
Each identified relationship was recorded with comprehensive metadata including connected individuals’ names and unique identifiers, relationship nature and directionality, relevant dates, documentary evidence type, archival references, geographic locations, event contexts, and interaction directions (Figure 1). A standardized weighting scale from 1–5 captured relationship intensities: 1 (figures referenced with no direct personal relationship); 2 (professional acquaintances or single documented interactions); 3 (teachers, mentors, regular professional contacts); 4 (direct peer-level interactions and sustained professional engagement); 5 (intimate professional partnerships such as long-term collaborators or co-founders).
Figure 1
Dataset excerpt from the Marion Scott project (1906–1953), mapping her professional and academic ties to composers and musicians in early 20th-century London, with details of relationship type, archival references, metadata, and event contexts.
Expert Interpretation and Contextual Assessment
Given that relationship type classification and weight assignment necessarily involve subjective interpretation of historical evidence, these assessments were guided by Dr. Christina Guillaumier’s specialized musicological expertise to ensure classifications reflected appropriate understanding of early twentieth-century musical contexts and professional practices.
Temporal Coverage and Data Considerations
The resulting dataset spans 1906–1953, with highest concentration during 1910–1940, corresponding to Scott’s most active professional period. Particular density appears around key events, including the Society of Women Musicians’ 1911 founding, Scott’s Haydn research in the 1920s–1930s, and her advocacy for Ivor Gurney. Date fields require preprocessing for analysis due to varying temporal specificity in archival documents, following approaches used in relational data modelling of textual corpora (Wills, 2015).
Quality control
Each entry was verified against the original documents through a systematic manual review, with all data undergoing double-checking between the curators and the compiler. Disputed classifications were resolved through collaborative examination of the original archival materials. Text normalization included removing non-standard characters and encoding all files in UTF-8 to ensure cross-platform compatibility. Metadata was validated for consistency across both dataset files.
(3) Dataset Description
Repository name
Zenodo
Object name
Marion Scott Professional Networks Dataset 1906–1953
Format names and versions
CSV, Excel (.xlsx)
Creation dates
2025-01-17 to 2025-08-20
Dataset creators
Dr. Antonina Puchkovskaia (King’s College London): primary curator, methodology design, and database architecture; Dr. Christina Guillaumier (Royal College of Music): secondary curator with musicology expertise; Luciana Perc (Royal College of Music): data compiler
Language
English
License
Creative Commons Attribution 4.0 International License (CC BY 4.0)
Publication date
2025-08-25
(4) Reuse Potential
While this dataset is specifically structured for social network analysis applications, its rich relational data offers significant benefits across multiple humanities disciplines. The systematic documentation of connections between women musicians, critics, composers, and institutions creates opportunities for cultural historians to trace previously invisible networks of influence in London musical culture. Feminist studies scholars will find empirical evidence for examining women’s agency within patriarchal institutions, with the documented relationships providing concrete examples of strategies women employed to circumvent institutional barriers and build collaborative networks of support and influence.
For music historians, the dataset enables revision of established narratives about British musical modernism and the discipline’s development beyond traditional score-based analysis toward performance and social contexts (Cook, 2013). The data advances understanding of how musical culture operated through networks of personal and professional relationships, providing empirical foundation for network-based approaches to musical culture (Crossley et al., 2014). The documentation process itself revealed other notable women musicians such as Liza Lehmann and Katherine Eggar as central network nodes, identifying them as priority subjects for future systematic research across these intersecting fields.
The dataset’s primary strength lies in its suitability for computational social network analysis. As demonstrated in studies of music collaboration networks, computational approaches can reveal structural patterns invisible through traditional analysis (Gienapp et al., 2021). Researchers can apply network centrality measures to identify key brokers in women’s musical networks, revealing how figures like Scott occupied crucial positions in cultural transmission despite lacking formal institutional power. The dataset can be analyzed using R or Python graph visualization packages, or imported directly into Gephi for interactive network exploration and analysis.
The manual extraction approach, while labor-intensive compared to automated methods for structured texts (Graham et al., 2020), ensures accurate interpretation of musical and historical contexts that require domain expertise. This methodology serves as a replicable model for similar projects focused on underrepresented figures in cultural history, with documentation protocols and analytical frameworks that can be adapted by scholars working at the intersections of feminist studies, cultural history, and musicology. The structured foundation specifically designed for network analysis thus supports both computational and traditional humanities approaches, making it valuable for researchers regardless of their methodological preferences or disciplinary backgrounds.
Limitations and Future Development
The documentation necessarily reflects archival preservation patterns, which embody historical biases about whose papers were preserved. Additionally, Years_Active data for some women remains incomplete due to name changes upon marriage and systematic gaps in historical record-keeping for women’s professional activities – limitations that paradoxically highlight the urgent need for further research to recover these lost trajectories. Despite these limitations, the dataset provides empirical foundation for reassessing both Scott’s individual significance and broader questions about measuring cultural influence through social network analysis.
Acknowledgements
I am grateful to the Royal College of Music for granting institutional access to the Marion Scott Collection and for the invaluable assistance of the RCM archivists throughout this research. I wish to extend particular thanks to Dr. Christina Guillaumier for initiating this collaboration, offering essential musicological insights, and contributing specialist expertise in the interpretation of historical musical contexts. I also acknowledge Luciana Perc for her meticulous work in the systematic compilation and verification of data.
Competing Interests
The author has no competing interests to declare.