Daejeon, July 27–31
Fashion and clothing have long been recognized across multiple disciplines as potent signifiers of identity, personality, psychological states, and socio-economic positioning within society. In psychology, clothing serves as a form of non-verbal communication and self-presentation (Kaiser, 1990); in cultural studies, it functions as a text encoding social relations and historical transformations (Barthes, 1967); in anthropology, garments constitute material culture that embodies collective memory and cultural transmission (Küchler & Miller, 2005). However, within the specific domain of African francophone literary studies, scholarship investigating the semiotic and narrative functions of fictional characters' fashion choices remains surprisingly limited. This research gap is particularly striking given the centrality of sartorial practices in African societies undergoing rapid social, political, and cultural transformations from colonial domination through postcolonial nation-building to contemporary globalisation.
This study pioneers an innovative methodological approach by applying Quantitative Narrative Analysis (QNA) and Michael Gavin’s literary mathematics theory to systematically map and analyse sartorial practices across a century of African francophone novelistic production. Our research interrogates how clothing functions not merely as descriptive detail but as a sophisticated narrative device encoding ideologies, power relations, and identity negotiations within the African literary imagination. By examining 50 purposefully selected novels spanning from 1921 to 2021, representing over 23 Francophone African nations, we trace the diachronic evolution of sartorial perceptions and representations across three distinct literary-historical periods: colonial (1921-1960), postcolonial (1961-1990), and contemporary/globalised (1991-2021).
Experts, globally, have long treated clothing as more than elegant physical appearance or an aesthetic expression. It is rather a language, an embodiment of cultural semiotics. One of the most influential voices in the 20th century is Barthes (1967), who asserted that fashion operates as a structured semiotic system through which societies express values, hierarchies, and ideologies. In the same vein, Bourdieu (1979) demonstrated how taste, including dress, is tied to socio-economic distinction and cultural capital. Evidently, what characters wear in novel or play or painted in poems is not totally accidental; it mirrors deeper social standing.
Foremost psychological perspectives, such as Flügel (1930), already theorised clothing as an extension of the self, mediating identity and desire. Later, Kaiser (1990) expanded this into a social psychology of elegance, showing how clothing becomes a tool for negotiating togetherness and difference as a form of social homeostasis. These fundamental ideas facilitate our understanding why sartorial details in novels carry narrative weight, exporting cultural civilisation.
In literary sartorial studies, this insight has evolved into the idea that clothing or fashion is not beauty but functional within narration. Deming and Kambara (2017), notably, present clothing as part of the making of the story shaping character evolution and narrative axes. More recently, Auguste (2024) and Morata (2022) accentuate how clothing encodes disjointed identities in contemporary fiction, especially in contexts marked by conflict or transformation.
As for sartorial practices in African contexts, clothing becomes even more coated. It is not only about identity, but also about history, power, and quest for survival. Classical works such as Fashioning Africa (Allman, 2004; Mustafa, 2019) and Rovine (2015) show how African dress traditions are influenced by colonial contacts, globalisation, and Africanistic creativity.
Clothing in Africa has often been caught in a tension between imposed practices and lived modernity. Rabine (2002) and Sylvanus (2016) emphasise how garments travel across frontier, physically and symbolically, creating mix identities. Hansen’s (2000) work on second-hand clothing in Zambia further illustrates how global economies reshape local sartorial meanings.
Importantly, feminist and gender-based studies expose how clothing intersects with power structures. Scholars such as Nfah-Abbenyi (1997) and Oyěwùmí (1997) opine that dress plays a role in constructing and contesting gender identities in African societies. Badejo (1996) similarly put clothing within larger strategies of survival and resistance.
Within literary analysis, Ajah and Egege (2017) show how sartorial practices in Aya de Yopougon reflect contemporary African identity and beauty politics. Their work proves how clothing becomes a narrative placeholder where modernity, femininity, and cultural negotiation intersect.
Furthermore, most of these studies rely on close reading of individual texts. What remains largely missing is a wide, long-term mapping of how sartorial practices evolve across multiple works and decades.
Francophone African literature offers a particularly rich terrain for analysing clothing because it spans colonial, postcolonial, and globalised contexts. During the colonial period, clothing often singled assimilation or resistance, who adopts European dress, and who refuses it, becomes a political statement (Mamdani, 1996; Cooper, 2002). In post-independence literature, sartorial excess or parody frequently reflects disillusionment with new elites. Studies of Alain Mabanckou’s work, for instance, show how clothing can become a satirical device exposing social contradictions (Dumont, 2020; Johnson, 2021).
In more recent novels, clothing increasingly reflects mobility, migration, and hybrid identities. Covington (2022) describes this as “fashioning the self,” where dress becomes a strategy for navigating fragmented postcolonial realities. Morata (2022) similarly points to a crisis of appearance, where identity is no longer stable but continuously performed.
Across these periods, clothing shifts from being a marker of fixed identity to a dynamic and often unstable signifier. Yet, this transformation has not been systematically analysed across a large corpus, another gap your study addresses. This is where digital humanities come in, and things get really interesting.
Quantitative Narrative Analysis (QNA), as developed by Franzosi (2010), provides a way to transform narratives into structured data. Instead of only interpreting texts, researchers can now track patterns: who wears what, when, and in what context enshrined in literary mathematics theory.
This approach aligns with wider developments in digital literary studies. Moretti (2005) introduced “distant reading,” encouraging scholars to step back and look at largescale patterns rather than individual texts. Jockers (2013) extended this through computational analysis of literary history, while Underwood (2019) demonstrated how machine learning can reveal long-term shifts in literary language.
Even more striking is the work of Michel et al. (2011), who analysed millions of digitized books to uncover cultural trends, a reminder that literature can be studied at scale. Similarly, Jacobs and Kinder (2017) show how machine learning can be used to analyse metaphors, opening the door to computational interpretations of meaning.
Despite these advances, African literature, and especially Francophone African novels, remains underrepresented in such studies which is why our research is crucial in closing the gap in the previous studies.
Digital humanities projects focusing on fashion have largely cantered on visual archives and material culture. These include digital collections of garments, textile databases, and machine-learning analyses of fashion imagery. Manovich (2020) describes this broader approach as cultural analytics, where large datasets are used to study cultural patterns. Visualization tools, as discussed by Drucker (2014), allow scholars to map trends over time, making abstract patterns visible.
Network analysis also plays an important role. Using frameworks developed by Newman (2010) and Easley and Kleinberg (2010), researchers can model relationships between elements, characters, objects, or themes. In this case, clothing is treated as nodes within a narrative network, revealing how it connects to identity, class, or power.
However, most existing digital humanities projects focus on Western fashion systems or visual data (Campbell, 2017; Fashion Institute of Technology, 2015; Moir, 2021; Sweeney, 2016). There is very little work that combines African literature, Sartorial studies, and computational methods.
Critical overview across the literature, a few patterns become clear. Initially, studies of clothing in literature are rich but mostly qualitative. Next, digital humanities offer powerful tools, but these have rarely been applied to African literary corpora. Lastly, Francophone African novels, despite their importance, remain underexplored in computational studies.
Our research is anchored in an interdisciplinary theoretical framework that bridges literary studies, fashion theory, and digital humanities. We employ semiotics as our primary analytical lens, following Barthes' conception of fashion as a system of signs that communicates complex social meanings (Barthes, 1967). This is complemented by literary mathematics theory (Gavin, 2023) built from relational theory, which examines meaning construction through patterned interplays between textual elements, social contexts, and historical moments (Deleuze & Guattari, 1987). Additionally, we integrate postcolonial theory, particularly concepts of hybridity (Bhabha, 1994) and cultural translation, to understand how sartorial practices negotiate between indigenous traditions and external influences.
Methodologically, this study implements a novel Quantitative Narrative Analysis (QNA) protocol specifically adapted for literary-sartorial analysis. Our QNA framework encompasses four interconnected analytical dimensions:
4. Corpus Construction and Data Processing
Our research corpus comprises sartorial narrative extracts from 51 novels, and one Malagasy novella selected through stratified purposeful sampling to ensure representativeness across geographical regions, historical periods, genders of authors, and literary movements. The corpus includes works by foundational figures (Ahmadou Kourouma, Mariama Bâ, Sony Labou Tansi), contemporary voices (Alain Mabanckou, Léonora Miano, Nétonon Noël Ndjékéry), and spans genres from historical fiction to magical realism to autofiction.
Data extraction follows a rigorous annotation protocol developed specifically for this project. Each novel undergoes multiple reading passes where trained annotators identify and code:
All annotations are stored in a relational database with standardised controlled vocabularies for sartorial terms, enabling consistent querying and analysis across the corpus. Inter-annotator agreement is maintained at κ > 0.85 through regular calibration sessions (see Figure 1).
Our initial analyses reveal several significant patterns in the representation of sartorial practices across the century of African francophone literature:
5.1. Historical Evolution of Sartorial Representation
We identify three distinct phases in the literary treatment of clothing:We identify three distinct phases in the literary treatment of clothing:
5.2. Regional Variations in Sartorial Imagination
Network analysis reveals distinct regional "sartorial signatures":Network analysis reveals distinct regional "sartorial signatures":
5.3. Gender and Sartorial Representation
Statistical analysis shows significant gender differences in sartorial description:Statistical analysis shows significant gender differences in sartorial description:
6. Innovative Contributions
By combining semiotics, postcolonial theory, and quantitative methods, our study does something new: it treats clothing not just as symbolism, but as data that can be mapped, measured, and analysed over time in the form of cultural analytics. Again, this study chosen allow us track how sartorial practices evolve across a century, identify patterns that are invisible to close reading alone as well as reconstruct clothing as a dynamic narrative variable.
In nutshell, this research makes several significant contributions to both Digital Humanities and African Literary Studies:
6.1 Methodological Innovation
We develop and validate a novel QNA protocol for literary sartorial analysis that can be adapted to other cultural contexts and literary traditions. Our integration of computational methods with nuanced literary interpretation represents an advance in humanities computing methodologies.
6.2. Theoretical Advancement
By applying quantitative methods to sartorial analysis, we move beyond anecdotal observations to identify statistically significant patterns in literary representation. This enables more robust claims about the relationship between literary form and social reality.
6.3. Empirical Findings
Our century-spanning analysis provides the first comprehensive mapping of sartorial representation in African francophone literature, revealing previously undocumented patterns in how clothing encodes historical experience, regional identity, and social transformation.Our century-spanning analysis provides the first comprehensive mapping of sartorial representation in African francophone literature, revealing previously undocumented patterns in how clothing encodes historical experience, regional identity, and social transformation.
6.4. Digital Resources
The research produces several publicly available digital resources:The research produces several publicly available digital resources:
7. Implications and Future Directions
7.1. Literary Studies Implications
Our findings challenge period-based approaches to African literature by revealing continuities and transformations in sartorial representation that cut across conventional period boundaries. They also demonstrate how "minor" descriptive details like clothing choices can encode major historical and ideological shifts.
7.2. Cultural History Applications
The research provides a novel source for cultural historians studying African fashion history, complementing archival and ethnographic sources with literary evidence of how clothing practices were perceived, valued, and contested.The research provides a novel source for cultural historians studying African fashion history, complementing archival and ethnographic sources with literary evidence of how clothing practices were perceived, valued, and contested.
7.3. Pedagogical Applications
The visualizations and datasets produced can enhance teaching of African literature by making visible the material and symbolic dimensions of literary representation.The visualizations and datasets produced can enhance teaching of African literature by making visible the material and symbolic dimensions of literary representation.
7.4. Future Research Directions
We identify several promising directions for future research:We identify several promising directions for future research:
8. Conclusion
This study demonstrates the rich analytical potential of applying Quantitative Narrative Analysis to the study of sartorial practices in African francophone literature. By systematically mapping clothing representations across a century of novelistic production, we reveal how garments function as complex narrative devices encoding historical experience, negotiating cultural identity, and commenting on social transformation. Our research shows that clothing in these novels is never merely decorative coverup but constitutes a sophisticated semiotic system through which authors grapple with the central questions of African modernity: tradition and change, local and global, individual and collective identity.
Clothing has always mattered in literature; however, it has often been read in rubbles. Digital humanities now offer the tools to see the bigger picture. By bringing together fashion studies, African literary criticism, and computational analysis, this study moves beyond interpretation toward systematic mapping of cultural meaning. It opens a new way of reading Francophone African novels, one where fabric, identity, and history are not only described, but also quantified and visualised.
The methodological framework developed here, offers a model for future digital humanities research that seeks to bridge quantitative and qualitative approaches to cultural analysis. As we prepare to present these findings at DH2026, we particularly look forward to discussing how such methods can foster cross-cultural and comparative research in the digital humanities, and how they might be adapted to other literary traditions and cultural contexts.
8. Technical Specifications
This research utilizes Python for data processing and analysis, with packages including NetworkX for network analysis, Matplotlib/Seaborn for visualization, Pandas for data manipulation, and Scikit-learn for statistical modelling. Visualization outputs include D3.js, mermaid.js for time series, interactive web displays for conference presentation. All code and data will be made available through a GitHub repository following publication.
Keywords: Quantitative Narrative Analysis (QNA), sartorial studies, African francophone literature, digital humanities, computational literary analysis, fashion semiotics, postcolonial studies, data visualization, narrative networks, cultural analytics.