Abstract
This essay argues that Latin American scholarship and movement practice are key to understanding the dynamics of the datafied society and countering its inequities. Examining the sources of inspiration of a frontrunner seeking to decolonize the datafied society – the Big Data from the South Initiative (BigDataSur) – we review Martín-Barbero's ontological shift from media to mediations, Freire's methodology centring individual agency and empowerment as a structural task of society, Mignolo's invite to take decoloniality as a praxis rather than merely an idea, Rodríguez's first-hand engagement with technology at the margins, Escobar's autonomous design for the pluriverse, and the critical ecology of eco-social movements. We engage with a new generation of Latin American thinkers who turn their gaze to core problems of today's systems of knowledge production, be they media or academia. Learning from these scholars, we warn against decolonial reductionism, namely the trend to evoke decolonial ideas and theories without fully committing to putting them into practice. We maintain that to decolonize datafication, we ought to also change how we generate knowledge about the datafied society. We outline three practical strategies that foster an open-ended dialogue on alternative approaches to datafication and scientific practice: multilingualism, public scholarship, and mentorship.
This article is a part of special theme on Critical Data Studies in Latin America. To see a full list of all articles in this special theme, please click here: https://journals.sagepub.com/page/bds/collections/critical_data_studies__in_latin_america
The history of Latin America is as much of a chronicle of aggressive colonization and pervasive dispossession as it is awash with social struggles and grassroots rebellion. In the 20th century alone, the great revolutions of Mexico (1910s), Bolivia (1952), Cuba (1959), and Nicaragua (1979) have brought about radical structural change that inspired activists across the world. The Zapatista struggle that emerged in the 1990s in Chiapas, rural Mexico, has fuelled the radical imagination of movements globally. Simón Bolívar (1783–1830), Emiliano Zapata (1879–1919), Ernesto ‘El Che’ Guevara (1928–1967), and the iconic Subcomandante Marcos (1957–) have epitomized the social justice ideals of their times. Noteworthily, uprisings and movements have gone hand in hand with a dynamic conceptual production that has changed the way we think about culture (Pasquali, 1963), education and empowerment (Freire, 1968), cultural imperialism (Somavia, 1977), media production and ‘mediations’ (Martin-Barbero, 1993), (de)colonization (Mignolo, 2000) – and counting. The legacy of Latin American thinkers is tangible and thriving and subverts our contemporary understandings of the datafied society, too. Because ‘ideas are also weapons’ (Marcos, 2004: 502), this essay revisits key sources of inspiration for the decolonization project applied to datafication. Examining one of the frontrunners seeking to decolonize the datafied society – the Big Data from the South Initiative, or BigDataSur – it outlines three practical strategies that foster an open-ended dialogue on alternative approaches to datafication.
Against decolonial reductionism: BigDataSur as open-ended dialogue
BigDataSur was launched in July 2017 by Stefania Milan and Emiliano Treré following the one-day workshop ‘Big Data from the South: From media to mediations, from datafication to data activism’ in Cartagena de Indias, Colombia. Denouncing how our understanding of the datafied society is predominantly shaped by Western research, the workshop interrogated the universalism of datafication from an ‘epistemology of the South’ perspective (Milan and Treré, 2017). Because ‘there is no way out from the coloniality of power from within Western (…) categories of thought’ (Mignolo, 2011: 45), we wanted to pay the way ‘to a different place, to a different beginning (…), to spatial sites of struggles and building’ (Mignolo, 2011: 45) within pervasive datafication. We aimed at practice ‘decolonial delinking’ (Quijano, 1991) or ‘epistemic disobedience’ (Mignolo, 2011: 45), foregrounding ‘local histories’ and ‘subaltern knowledges’ (Mignolo, 2000) and centring users’ practices of engagement and the agency of the damnés de la terre (the wretched of the earth) (Maldonado-Torres, 2016).
The workshop made apparent the extent to which such conversation was sorely needed. Thus, we launched BigDataSur to support an inclusive, open-ended dialogue on ‘de-Westernizing’ datafication (cf. Waisbord and Mellado, 2014). The goal was to create a space that would both highlight the importance of decolonization in the study and practice of the datafied society and actively push back against oversimplified views of it – what we term decolonial reductionism. This concept indicates the trend to evoke decolonial ideas and theories without fully committing to putting them into practice. But Mignolo and Walsh (2018) warned that decoloniality is not merely a new paradigm or mode of critical thinking, but a standpoint, an analytic, and above all a praxis. With much current research superficially touching on decolonial ideas, decolonial reductionism undermines efforts to radically change the way we generate knowledge about the datafied society – a step that, we argue, is as important as the type and nature of the knowledge we generate. Put differently, the sole reading of a socio-technical problem in a decolonial light is not enough: how we create such interpretation matters, as we demonstrate later.
Over time, BigDataSur came to include a mailing list with 400+ members globally, a multilingual blog, and academic publications, including two special issues (Milan and Treré, 2019; Treré and Milan, 2021). It promoted events engaging both the public and researchers from media studies to area studies, from critical informatics to development studies and law. The spin-off blog ‘COVID-19 from the Margins’ was launched in 2020 to interpret the crisis from the perspective of marginalized communities such as gig workers, migrants, and the poor, and later became an open-access, multilingual volume with the same title (Milan et al., 2021). Despite the intermittent activity due to insufficient funding, BigDataSur remains a significant reference point for those discontented with datafication.
Latin America thinking and BigDataSur
Latin American scholars and movements have permeated BigDataSur from the start – and not only because Colombia is the symbolic birthplace of the network. Concepts such as the Andean indigenous social philosophy of buen vivir (‘good living’; Barranquero and Baeza, 2017), which emphasizes a community-centred, ecologically balanced, and culturally sensitive way of life, have sparked alternative imaginaries of a more equitable and sustainable data-driven society (Milan and Treré, 2021b). Continuing the conversation inaugurated by the special issue on ‘Latin American Perspectives on Datafication and Artificial intelligence’ (Palabra Clave, 2021), we explore the central tenets of the BigDataSur Southern epistemology as it is moulded on Latin America theoretical production. The list is all but exhaustive (Table 1). It illustrates how authors who do not focus on datafication can inspire fresh perspectives for exploring alternatives to exploitative data systems and scientific practices.
Key Latin American theorists for decolonizing the data-driven society.
The work of Jesus Martín-Barbero (1987) has been pivotal for cultural and media studies scholars: his ontological move ‘from media to mediations’ brought under the spotlight popular appropriations of media and their messages. More than three decades later, Martín-Barbero's call inspired the conceptual shift at the heart of BigDataSur: centring grassroots agency within critical data studies, promoting an ontological move from data as an artefact to how data are (re-)mediated. Paraphrasing Castro-Gómez and Grosfoguel (2007: 20), attention should go to ‘the forms of interstitial, hybrid knowledge emerging in subversive complicity with the system’ (Milan and Treré, 2019: 328), departing from purely functionalist analyses of information and technology. Re-centring agency shifts the gaze towards social actors and their situated, affective responses to datafication (e.g. Reilly and Morales, 2023), instead of exclusively examining institutions and the ‘colonial’ forms of power enshrined in massive data collection (Couldry and Mejias, 2019). BigDataSur, however, goes one step forward, marrying the ontological move inspired by Martín-Barbero to an epistemological one, stressing the value of socially engaged scholarship grounded on its environments as opposed to research removed from the needs of the people on the ground. Researching data practices and mediations of social actors in the South(s) unlocks horizons of possibility, experimentation, and hope which allow us to imagine new ways of changing the world through and/or with data. This epistemic operation resonates with Latin American scholarship on ‘lo popular’ intended as ‘people's ways of narrating, of exposing their bodies and politicizing their lives from and in communication’ (Rincón and Marroquín, 2019: 43; for an application to the data-driven society, see Siles et al., 2020).
Brazilian educator Paulo Freire is known as the theorist of liberation pedagogy, or pedagogy of autonomy (1968). Influenced by Marxism and liberation theology, Freire criticized the structural inability of the education system to empower the poor to overcome their conditions. He proposed an educational approach based on dialogue and horizontal relationships between learners and teachers, promoting the role of individuals as active agents in transforming their world. Liberation pedagogy nurtures a ‘critical consciousness’ (conscientização in Portuguese) which is ‘an intrinsic part of cultural action for freedom’ and supports the promotion of social change. Liberation pedagogy inspires the work of BigDataSur for three reasons: the active role it attributes to individuals and communities in shaping their futures, consistent with Martín-Barbero's ‘mediations’; the unity of praxis (e.g. engagement with data) and theory (e.g. values) in the struggle for social change (Milan and Treré, 2021b), and its rigorous methodology to intervene in society – his pedagogy (Suzina and Tufte, 2020).
Another key reference for BigDataSur is Argentinian decolonial thinker Walter Mignolo. Scrutinizing the effect of imperial domination over the continent, Mignolo demonstrated that the very idea of Latin America is a colonial product, where hegemonic narratives of Western civilization worked to eradicate marginal knowledge and praxis. Mignolo emphasized the importance of decolonial inquiry within and beyond academic settings, aiming to break free from dominant modes of existence and knowledge. As noted above, jointly with Walsh (2018), Mignolo exposed how decoloniality should be primarily a praxis. These observations resonated strongly within the BigDataSur agenda, positing that there is not a singular way to analyse datafication and its dynamics of dispossession. Coupled with the notion of decolonial ‘border thinking’ as epistemology from a subaltern perspective (Mignolo, 2000), these ideas have encouraged BigDataSur to engage not only with decolonizing data, but also with the transformation of knowledge production that is the necessary first step towards decolonization. Accordingly, our engagement with decoloniality must be flexible, agile, and contextual if we are to avoid the temptation to use the (de)colonial framework as a lens that aspires to include and explain datafication in relation to domination (Treré, 2022) – in other words, a way of merely gesturing towards a decolonial reality without working for its realization.
The writings by Clemencia Rodríguez, known for her seminal work on ‘citizen media’ (2001), have encouraged us to attend to critical imagination exercised through first-person engagement with technology. Mimicking her notion of ‘media from the margins’ (2017), we have elaborated the idea of ‘data from the margins’ (Milan and Treré, 2021a) to complement our notion of the plural South(s), too often flattened to the ‘Global South’, making room for a journey ‘into the land of otherwise’ (Rodriguez, 2017: 49). Rodríguez conceives the margin as ‘a shortcut to speak of complex dynamics of power inequality. Processes of asymmetrical access to material and symbolic resources shape differentiated and unequal access to the public sphere’ (2017: 56). Focusing on data at the margins allows us to illuminate the plurality of voices silenced by the hegemonic discourses of ‘data at the centre’. The margins are sites of struggles where datafication often looks different. There imaginaries about alternative data futures are concocted that shake the foundations of what is taken for granted and cast as inevitable and immutable – to the point that changes become possible (cf. Rodriguez, 2001).
Colombian anthropologist Arturo Escobar's critique of development (e.g. 1995) and his writings on the ‘pluriverse’ (i.a., 2018), have greatly influenced BigDataSur activities. Firstly, the Zapatista-inspired notion of the pluriverse as ‘a world where many worlds fit’ (2018: xiii), has encouraged us to include and give voice to the many data realities, each with its own ontological and epistemic grounding. Pluriversal understandings of data and justice challenge the ‘data universalism’ that flattens social realities (Milan and Treré, 2019). As a result, BigDataSur views the Majority World as a multi-layered place of, and a proxy for, resistance, subversion, and creativity. Countless Souths, in the plural – situations of disempowerment but also imaginative agency – can be found also in the Global North and the wealthy West, whenever people resist injustice and fight for better life conditions within data capitalism. Secondly, within ‘COVID-19 from the margins’, Escobar's notion of ‘autonomous design’, involving a concern for social justice, radical equality, and non-hierarchy of all beings, has provided a valuable perspective for critiquing technological solutionism in the datafied society (Milan, 2020), and to productively imagine post-pandemic data futures (Milan et al., 2021).
Escobar's observations on the pluriverse evoke another important Latin American legacy to engage with the critical junctures of the present, such as the climate crisis, to which the data-driven society is a major contributor: the critical ecology tradition of biodiversity and ecology preservation advanced by eco-social movements such as the Brazilian Movimento dos Trabalhadores Rurais sem Terra and the international peasants’ movement Vía Campesina. In an ideal dialogue with these movements, in our exercise to imagine how a ‘buen vivir with data’ would look like, we identified three axes: the fusion of the social with the ecological question, a dialogic and participatory approach to decision-making, and a ‘localized, relationship-oriented’ practice of community care and solidarity based on the recognition of ontological difference and commonalities’ (Milan and Treré, 2021b: 101).
The possibilities for ‘innovative theoretical arguments that capitalize on the wealth of empirical lines of research and the unique intellectual tradition in the region’ (Waisbord, 2014: 12) are endless. In no particular order, we highlight the work of Juan Díaz Bordenave, Carlos Monsiváis, María Cristina Mata, Néstor García Canclini, Immacolata Vasallo de Lopes, Luis Ramiro Beltrán, Alfonso Gumucio Dagron, Rossana Reguillo, Bolívar Echeverría, Juan Díaz Bordenave, Ismael de Oliveira, Renato Ortiz, Arlindo Machado, Mario Kaplún, Valerio Fuenzalida, Guillermo Orozco, Jorge González, Gloria Feliciano, Jorge Huergo, Orlando Fals Borda, Rosa Maria Alfaro, Carlos Scolari, Muniz Sodré, and Cicilia Peruzzo. A new generation of Latin American scholars is also leaving a mark. To name but a few, Dorismilda Flores-Márquez (2021) has illustrated how Latin American communication thinking has developed as a way of showing a resistant face to domination in both practical and conceptual terms. Grohmann et al. (2022) have exposed how asymmetries in the algorithmic management of labour in Brazil contribute to our understanding of fraud and deception in the datafied society. Tironi and Valderrama (2021) have positioned app prototyping as a method of inquiry simultaneously critical and decolonial. Resonating with BigDataSur's commitment to multilingualism, Ana Cristina Suzina (2020) has denounced how the predominance of English as the lingua franca of academia is part of a project of epistemological domination. The ‘decolonisation of scientific thought’ requires making room for other idioms – as experimented with BigDataSur from the start. Our multilingual conversations were often untranslated because there is value in experiencing occasional discomfort when not understanding the language of others – it is the discomfort that the dispossessed often suffer (Milan et al., 2021).
Looking ahead: Three commitments to fight decolonial reductionism
This brief excursus has illustrated the value of Latin American scholarship and movement practice for understanding the datafied society, despite the ‘(in)visibility of Latin American scientific production’ in academic publishing (Suzina, 2020: 171). Regrettably, scientific publishing ‘is not a level playing field’ and ‘we continue to cause minoritized scholars to experience injustice’ when their first language is not English (Nature Human Behaviour, 2023: 1019). Editorial and reviewer bias plays a role in keeping barriers high (Smith et al., 2023). Against this backdrop, we advocate for three tangible measures to advance the decolonization of knowledge production about the datafied society – in other words, the ‘how’ alongside the ‘what’.
Firstly, our community should embrace multilingualism, because language is not a neutral vehicle of meaning but is part of the meaning itself and as such allows us to make and reproduce worlds. Valuing linguistic diversity in scientific production means not only being inclusive and open to (and regard) other voices, but also making room for alternative ways of understanding society. Secondly, we ought to encourage the promotion of public scholarship as an effective form of intervention. This entails the endorsement of open access and open science, but also the experimentation with other formats of knowledge sharing that are palatable to the public beyond academia (e.g. podcasts, art, and popular writing). Thirdly, those in a position of privilege should engage in mentorship, passing on skills and professional norms to those who may lack access to the expertise required for academic writing. An approach to mentorship based on the Freirean ideal of (recursive) dialogue is crucial for empowering individuals and communities at the margins as well as early career researchers to share their narratives in their own voices, rather than having them co-opted by others. In practical terms, this could entail providing editorial support as a means of fostering inclusivity, as successfully experimented within BigDataSur.
Many questions remain to be addressed. How can we put Latin American theories and perspectives in dialogue with other ‘peripheral’ traditions? Can we aspire to ‘decolonize academia’, while, for example, publishing in conventional venues or accepting funding from questionable sources? How can we decolonize not only our epistemologies, but also our methodologies? Only by collectively embracing these inquiries can we prevent falling prey to the superficiality of decolonization as a metaphor devoid of genuine commitment, merely promoting a façade of ‘settler moves to innocence’ (Tuck and Yang, 2012).
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
