Big Data from the South(s): Beyond Data Universalism

Abstract

This article introduces the tenets of a theory of datafication of and in the Souths. It calls for a de-Westernization of critical data studies, in view of promoting a reparation to the cognitive injustice that fails to recognize non-mainstream ways of knowing the world through data. It situates the “Big Data from the South” research agenda as an epistemological, ontological, and ethical program and outlines five conceptual operations to shape this agenda. First, it suggests moving past the “universalism” associated with our interpretations of datafication. Second, it advocates understanding the South as a composite and plural entity, beyond the geographical connotation (i.e., “global South”). Third, it postulates a critical engagement with the decolonial approach. Fourth, it argues for the need to bring agency to the core of our analyses. Finally, it suggests embracing the imaginaries of datafication emerging from the Souths, foregrounding empowering ways of thinking data from the margins.

Datafication has put new weapons in the hands of institutions and corporations in the business of managing people. And it seems to hit harder where people, laws, and human rights are the most fragile. Undocumented migrants crossing European borders become entries in biometric databases designed for “processing alterity” rather than solving a humanitarian crisis (Pelizza 2019). By 2020, every Indian will be enlisted in Aadhaar, the world’s largest identity database and a privacy-infringing “behemoth” (Prakash 2017). Drones and georeferential radars watch over the Mapuche indigenous communities opposing corporate land exploitation in the Chilean region of Araucanía (Parra 2016). The list could continue: the algorithmic power enshrined in “gig economy” platforms reinforces existing inequalities in the labor market of on-demand workforce (Chen 2017). With the advent of big data, development interventions in the Global South are becoming the mere “byproduct of larger-scale processes of informational capitalism” (Taylor and Broeders 2015, 229). But as this composite geography of disempowerment in the datafied society shows, the dark side of big data produces “algorithmic states of exception” (McQuillan 2015) that do not necessarily map into the known North–South dichotomy.

Western epistemic centers—especially those traditionally considered the interpreters and storytellers of technological development—have produced hyperbolic narratives of the “big data revolution” (see, for one, Cukier and Mayer-Schoenberger 2013). Counterbalancing this powerful imaginary, critical literature on the impact of datafication for individuals and communities has mushroomed over the past few years. Scholars of various disciplines have interrogated the profound consequences of this paradigm change on the social, cultural, and political domains (among others, Lyon 2018; Srnicek 2016; Zuboff 2019). Now we know that no matter how disruptive the paradigm change (Kitchin 2014) and how flamboyant the chronicles of digital positivism associated with it (for a critique, see Mosco 2014), the power dynamics embedded in information flows and knowledge structures have survived pretty much unaltered—or have been made worse, reiterating or exacerbating existing divides (Andrejevic 2014). Yet, these analyses often take as frame of reference the liberal democracies of the West, with their sociocultural substrate and long tradition of representative institutions, rule of law, and citizen involvement in public affairs—and their self-representation as wealthy, advanced democracies. The different, the underprivileged, the silenced, the subaltern, and the “have nots,” whose presence outdoes geographical boundaries, often remain in the blind spot.

Two uncomfortable observations arise. First, while the majority of the world’s population today resides outside the West, we continue to frame key debates on democracy, surveillance, and the recent automation turn by means of “Western” concerns, contexts, user behavior patterns, and conceptual frameworks (cf. Arora 2019), with “White masculinity” being still “the normative core of scholarly inquiry” (Chakravartty et al. 2018, 254). But theory tends to travel badly; more often than not, in our thrust to build on the shoulders of giants, we uncritically disregard the specificities of distinct geographies, cultures, and communities. Hence, the question, how does datafication unfold in countries with fragile democracies, flimsy economies, impending poverty? In other words, what about the particularities and idiosyncrasies of the so-called Global South? To take this challenge seriously, we ought to deal with embarrassing questions concerning not only the unequal distributions of power and resources across the globe but also the political economy of knowledge production. Western modernity has supplied the knowledge underlying colonialism, and later global capitalism: both historical processes have marginalized and devalued the knowledge as well as the specific ways of knowing of the Global South (Santos 2014). The relentless push toward datafication, with its deleterious consequences for people and communities on the ground, seems to exacerbate this process, prompting us to ask whether we are facing the next stage of what Santos (2014) denounced as the “epistemicide” of the South.

Second, inequality transcends boundaries and known geographies—and big data are no game-changer in this respect. Even within the so-called developed countries, we tend to take for granted certain “basic” rights that are however modulated on class, race, access to income, education, and citizenship and labor rights, underscoring the impact of datafication on the disadvantaged, the silenced, and the invisible (with some exceptions, see, for example, Gangadharan 2012; Mustafa 2016; Noble 2018). How, then, does datafication affect individuals and communities with sparse access to human rights protection, education, or income, and/or with nonmainstream visions of their future? In sum, what happens to those individuals and communities “at the bottom of the data pyramid” (Arora 2016), be it for class, racial, legal, or sociocultural reasons?

We fear that the conceptual and methodological toolbox available to datafication scholars is only partially able to grasp the obscure developments, the cultural richness, and the vibrant creativity emerging at the margins of the “empire.” We face a major ontological problem: does the fabric of datafication shift alongside with inequality, geography, class, race, and culture—in a nutshell, alongside with its publics and the various denominations within said publics, which are hardly ever homogeneous? If we accept this challenge, what is new, distinct, and inconsistent with “mainstream” theories? What questions, concepts, theories, methods would we embrace or have to devise if we are to “decolonize,” metaphorically and not, our thinking?

This article has the ambition to sketch the central tenets of a theory of datafication of and in the Souths, where the South is however not merely a geographical or geopolitical marker (as in “Global South”) but a plural entity subsuming also the different, the underprivileged, the alternative, the resistant, the invisible, and the subversive. The “big data from the South” agenda gives voice to distinct data practices and epistemologies emerging in the myriad of Souths, their specific challenges, and the associated demand for alternative models. But it also plays with the many geographical denominations still inhabiting our scholarship and our imaginaries, disaggregating familiar geographical oppositions and the interpretation of the respective poles as homogeneous entities, in view of making away with prevailing dichotomies (e.g., Western/non-Western, or developed/underdeveloped, cf. Castro-Gómez and Grosfoguel 2007). In so doing, the essay wants to probe the dominant geography of data flows and infrastructure and their charting on theory development. It calls for a much-needed de-Westernization of critical data studies and social science more in general. More than providing clear-cut answers, it generates questions and instills doubts, under the assumption that what we propose is a long-term, open-ended, and collective ride which can only be roughly sketched in the limited space of this contribution.

The essay is organized as follows. First, it situates the “Big Data from the South” research agenda as an epistemological, ontological, and ethical program. Then, it outlines five conceptual operations that we believe are key in shaping the research agenda of Big Data from the South, giving voice to fields of inquiry, conceptual lenses, and scholars that can assist us in our reconsideration of mainstream interpretations of datafication.

This article introduces the special issue exploring “Big Data from the South.” Three articles of this collection scrutinize tensions and possibilities in rethinking datafication from a Southern perspective, reinterpreting the notion of colonialism as a useful lens to understand the pressures of datafication on people and communities. Couldry and Mejias (2019) introduce the notion of data colonialism, a process that normalizes the exploitation of human beings through data. Writing from the “North in the South” vantage point of Australia, Mann and Daly (2019) investigate the country’s information imperialism and digital colonialism. Ricaurte (2019) develops a framework for analyzing data colonization and examines a case of citizen resistance to data colonialism and gender violence in Mexico. Arora (2019) criticizes the neoliberal ideology and the development paradigm inspiring privacy policies in the Global South, calling for decolonizing the approach to privacy by practicing epistemic disobedience and reimagining people and place. Chenou and Cepeda-Másmela (2019) zoom in on the case of the #NiUnaMenos mobilization in Argentina, discussing the alternative imaginaries associated with a bottom-up database of gender violence to reflect on how data activism translates into specific local contexts. The final commentary by Segura and Waisbord (2019) reflects on the key issues explored by this set of contributions, cautioning about two potential pitfalls: assuming that data citizenship in the South is necessarily different than in the North and falling into essentialist positions about “the South.”

Datafication and Its Discontents: Situating the Research Agenda

Important reflections on the pitfalls and the challenges of datafication for citizens and democracies have mushroomed in the last half decade. The expansion of data mining practices by both corporations and states gave rise to critical questions about systematic surveillance and privacy invasion (Lyon 2015). Various disciplines, at the intersection of social sciences, the humanities, and informatics, have called attention to the potential inequality, discrimination, and exclusion harbored by the mechanisms of big data (e.g., Eubanks 2018; Gangadharan 2012; Noble 2018). Scholars have reminded us that big data are not merely a technological issue or the flywheel of knowledge, innovation, and change, but a “mythology” and an “ideology” we ought to critically interrogate (boyd and Crawford 2012; van Dijck 2014). They have shown that big data, although imbued with modernization narratives, are not without risks and threats, as opaque regimes of population management and control have taken central stage (Aradau and Blanke 2017; Cheney-Lippold 2011). Colleagues have encouraged us to pay attention to “small” or “good enough” data produced by nonexperts (Gabrys et al. 2016; Kitchin and Lauriault 2015), to bottom-up practices (Couldry and Powell 2014), everyday forms of critical engagement with data (Kennedy and Hill 2017), and feminist approaches to data and their uses (D’Ignazio and Klein 2019; Fotopolou forthcoming). However, our understanding of datafication remains for the most part still sturdily anchored to the dynamics of Western democracy.

The nature of the problem is threefold. To start with, a “big data from the South” agenda presents us with an ontological question, which interrogates the fabric of datafication and its consequences not only on distinct sociocultural and geopolitical areas but also on differential access to rights and privilege. What do we mean by datafication, surveillance, and automation in the Souths? As we explain below, however, we do not argue for the idealization of the South and what it stands for. Although we should resist the tendency to romanticize the Souths, we should challenge analyses that implicitly assume Western democracies as the privileged setting where and from where to observe the dynamics of datafication, overlooking the richness and the diversity of meanings, worldviews, and practices emerging in the Souths, as well as their specific challenges and inherent problems.

Second, big data from the South is an epistemological puzzle in that it invites us to critically examine how we get to learn about datafication and its impact. This includes a methodological component, whereby some methods are considered more apt to foster empowerment of research subjects (see Freire 1968; Costanza-Chock 2018). But it speaks also to the way researchers relate to and interact with social actors and communities on the ground. Hence, this epistemological program calls for an “engaged” approach to research, whereby scholars, safeguarding the gold standards of scientific research, take communities as starting (and end) points of their work, whenever possible involving them in a process of co-inquiry (S. Milan 2010). Such approach to knowledge production considers communities as “skilled learners” and seeks to foster community building and knowledge sharing (C. Milan and Milan 2016).

Finally, big data from the South is also an ethical query, as it forces us to position ourselves as students of social phenomena, our motives, goals, and agendas, in relation to the situations of alterity, coloniality, inequality, or resistance we encounter in our research—and the agendas, desires, and needs of the social actors involved. This entails questioning research processes, practices, and relationships, as well as infrastructure in the context of ongoing power relations (cf. Kazansky et al. 2019). But it also means, as we shall see in the “Through the Decolonial Lens” section, practicing “semiotic resistance capable of giving new meaning to the hegemonic forms of knowledge from the point of view of the post-Eurocentric rationality of subaltern subjectivities” (Castro-Gómez and Grosfoguel 2007, 20, own translation; see also Mignolo 2007; Ong 2019). In this respect, our reflections on big data from the Souths fit within—and hope to feed—the broader process of epistemological repositioning of the social sciences (see, e.g., Bringel and Domingues 2017; Castro-Gómez 2007; Comaroff and Comaroff 2015; Connell 2014; Santos 2014). As Bhambra and Santos (2017, 9) observed, “if the injustices of the past continue into the present and are in need of repair (and reparation), that reparative work must also be extended to the disciplinary structures that obscure as much as illuminate the path ahead.” We argue that the sociotechnical dynamics of datafication should be understood in relation to, and measured against, “the historical processes of dispossession, enslavement, appropriation and extraction . . . central to the emergence of the modern world” (Bhambra and Santos 2017, 4)—pending the risk of perpetuating Western-centric interpretations and power dynamics. The same is to be said for the theoretical toolbox used to address datafication. Luckily, our effort is not entirely new. This agenda is well in the cords of critical data studies and other strands of research interrogating the centrality of “Western” interpretations.

Beyond the West? Five Conceptual Operations

Approaching the epistemic blind spots in our current analyses of datafication and surveillance, we now move to sketch the kernel of a conceptual map on big data from the South. We believe that five conceptual operations are key to question the domination of Western paradigms in our scholarship. These operations are profoundly interrelated and are here disentangled only for the sake of clarity.

First, we have to move past the “universalism” associated with the interpretation of datafication and its consequences. Second, we ought to understand the South as a composite, plural entity. Third, we have to critically engage with the decolonial approach. Fourth, we need to decenter data studies away from the focus on data, by bringing agency to the core of our analyses. Finally, we have to embrace the novel imaginaries of datafication emerging from the Souths, zooming on in the specific ways of thinking data from the margins. In what follows, we ground these claims.

Beyond Data Universalism

Investigating the multiple initiatives to “network” contemporary Peru, Chan (2013) challenged the myth of “digital universalism,” that is, the tendency to assimilate the cultural diversity of technological developments in the Global South to Silicon Valley’s principles. She showed how digital cultures and network politics emerging in the periphery—commonly considered merely subaltern to the “center”—do much more than replicate the technological future imagined in the mainstream centers of technological entrepreneurship.

Digital technology is far from being just “imported magic” (Medina et al. 2014), “travelling” unchanged from the North to the South. Instead, the social and cultural appropriations of technology in the Global South often unsettle hegemonic conceptions of innovation. They cast light on the context-specific and contentious nature of technological development as it is molded by the conflicting interests of an array of social forces, for example, governments, activists, and indigenous communities.

Similarly, we observe an impending data universalism infesting many of the readings of datafication today. A vestige of the early-day technological determinism (Chandler 2000), data universalism is the original sin of Western interpretations in particular, as it tends to assimilate the heterogeneity of diverse contexts and to gloss over differences and cultural specificities. Building on Chan’s (2013) analysis of digital universalism, we argue that the main problem with data universalism is that it is asocial and ahistorical, presenting technology (and datafication-related dynamics, we add) as something operating outside of history and of specific sociopolitical, cultural, and economic contexts. We thus suggest making away of these universalizing readings with their corollary of normalization and generalization, to the benefit not only of epistemic diversity but especially of epistemic justice. Epistemic justice does not aim simply at creating or voicing “otherness.” Rather, it includes a fundamental component of affirmative action, which takes inequality seriously and seek to work against it. As we shall see in the “Unleashing Novel Data Imaginaries” section, it is constitutive of enabling contestation (cf. Fricker 2013) but also of promoting the formation of alternatives.

South as a Plural Entity

The notion of the Global South has been charged with a multiplicity of often conflicting meanings, following geopolitical and also sociocultural shifts. Oftentimes it has embedded reductionist or folkloric nuances (visible, for example, in the notion of mestizaje or the process of mixing ancestries, cf. Cornejo-Polar 1994). Recent debates have displayed the political weight of the term, its ambivalent but also empowering nature, as well as the need not to take it too literally in terms of geographical boundaries (Wolvers et al. 2015). Various thinkers have urged us to understand the notion of the Global South as a dynamic construct that should be placed within the shifting contours and dynamics of globalization and global capitalism. Santos (2014) went as far as postulating the urgency of novel “epistemologies from the South” against the leveling enacted by neoliberalism (see also Huérfano et al. 2016). He sought to encapsulate the richness of the knowledge and experiences of/in the South, often silenced and excluded by an epistemically powerful North bearer of Eurocentric interpretations of grassroots knowledge production (cf. Flórez-Flórez 2007).

Following on these footsteps, we maintain a flexible, expansive, and plural definition of the South(s) that casts it as a place of (and a proxy for) alterity, resistance, subversion, and creativity, embracing the dynamism and the multiplicity of interpretations while going beyond the geopolitical denomination. On the basis of this definition, which goes beyond the restrictive notion of the “Global South,” we can find countless Souths also in what would normally be inscribed under the West, as long as people suffer discrimination, and/or enact resistance to injustice and oppression and fight for better life conditions against the impending data capitalism. With our definition, we engage in an exercise of disaggregation of the geographical dimension: while, on one hand, geography per se loses centrality in favor of a “broader” and more imaginative definition of the South as also metaphor and proxy, on the other hand, we recenter geography, recognizing the differences it harbors and especially the shaping roles of power inequalities, old and new, still visible today in, for example, digital and data infrastructure. Simultaneously, we want to avoid the temptation to romanticize the South and reify the North. None of these is intrinsically better: rather, they should be analyzed critically, their biases and incongruences exposed. This means for instance avoiding romanticizing indigenous knowledge but exploring it in all its contradictory aspects. Furthermore, the South is not necessarily distinct—often, processes and practices observed in the North replicate in the South with no particular differences. Said differences should not be assumed but thoroughly investigated. The North is not a single, unified block either: there are multiple and diverse Norths. We thus call for a much-needed dialogue, in search of new ways to unleash the richness of the knowledge emerging in the Souths, questioning the fact that most theory is a product of the North: while valuable and illustrative, it is also incomplete and problematic. ¹

Through the Decolonial Lens

To understand the complexity of data extraction, storage, and processing in the Souths, it is essential—as we argued in the introduction—to situate them within the historical processes of domination, extraction, exploitation, and oppression that are central to the modern world—and that continue to be predominant to this day. A decolonial lens—rather than a postcolonial one, of Anglo-Saxon origin—is particularly apt to critically explore the complexity of these processes. Decolonial thinking puts the colonial question at the core of the process of knowledge production (Mignolo 2000), urging us to make three important operations. First, going beyond a Eurocentric perspective, it foregrounds the periphery and the margins, questioning border thinking and the geopolitics of knowledge (Mignolo 2000). Second, it gives adequate consideration to the material (as opposed to merely the cultural) dimension of knowledge generation (see, for example, Castro-GómezGrosfoguel 2007). Finally, it centers the demand for reparative measures, anchoring the possibilities of the future to the injustices of the past and their consequences, such as power asymmetries, persisting in the present (Mustafa 2014).

Thinking of data capitalism as a novel manifestation of colonialism is gaining traction, as some of the contributions to this special issue make abundantly clear (Couldry and Mejias 2019; Mann and Daly 2019; Ricaurte 2019). Just like historic colonialism appropriated territory and resources for profit, data colonialism exploits human beings through data, across geographies, paving the way for the “capitalization of life without limit” (Couldry and Mejias 2019). It is thus imperative to unmask and deconstruct this complex process of colonization through data by dispossession (Thatcher et al. 2016; see also Halkort In press). We ought to understand this new rationality as an expression of the coloniality of power (Mignolo 2011; Quijano 2007), manifested as the violent imposition of ways of being, thinking, and feeling that systematically deny the existence of alternative worlds, and threatens life on earth (Barranquero and Baeza 2017; Escobar 2017). These reflections resonate with Casilli’s (2017) call for a “digital decolonial turn” in digital labor studies, to make visible that coloniality is produced on today’s global digital platforms relying on the exploitation of unrecognized and marginalized identities. They also resonate with the conceptual underpinnings of decolonial computing, which, taking the lead from decolonial and critical race theorists like Mignolo and Grosfoguel (2011) in centering the “margins/borders/periphery of the world system,” foregrounds “systemic racial concerns” and “questions . . . technologies in terms of both a geo-politics of knowledge and a body-politics of knowledge” (Mustafa 2014, 1, 4). ²

Yet, just as we should resist the tendency to idealize the Souths, we should not uncritically adopt alternative decolonial (and even less so, postcolonial) thinking either. As Asher (2013, 838) argued, “unless we engage with the modernity/coloniality/decoloniality research program critically, we risk resuscitating old binaries (theory vs. practice, structure vs. agency, and identity politics vs. anti-capitalist struggles) at best and simple reversals (modern bad and tradition good) at worst.” Furthermore, we believe that decolonial perspectives have their own blind spots and should more explicitly dialogue with other powerful conceptual lenses. Feminist interpretations, for example, center embodiment and situatedness in relation to disempowerment (see, for example, Suchman 2006; Wajcman 2010), although occasionally at the expense of other dimensions such as race (Schueller 2005). Costanza-Chock (2018) invites us to pay closer attention to intersectionality and the matrix of domination—concepts developed by Kimberlé Crenshaw and Patricia Hill Collins, respectively—when investigating data systems. Intersectionality points to the ways in which our life chances are shaped by our specific location within the intersecting fields of race, class, and gender; it can help us disentangle how capitalism, white supremacy, and heteropatriarchy are interlocking systems. This can, in turn, inform our understanding of the design, implementation, and development of data systems in the Souths (see also Escobar 2017). Finally, as we argue below, we are persuaded that a decolonial program should coexist with a constituting movement of holistic conceptual and methodological alternatives within a prefigurative agenda (Bringel and Domingues 2015). Although the former is making its way into critical data studies, the latter is still in its infancy. How can research on big data from the South help to generate this kind of shift?

Recentering Agency

Many readings of the role of data within contemporary societies are flawed by a techno-centric view of data devoting excessive attention to technical aspects to the detriment of appropriations, practices, and the human agency around and behind data. Recent literature, however, has urged scholars to focus on data practices, relocating agency at the center of the debate around the implications of datafication (Dencik In press; S. Milan 2018; In press; Rincón and Marroquín Forthcoming; Trere 2019). The Spanish Colombian philosopher Martín-Barbero’s (1993) epochal move from media to mediations is key for this operation (see also Radfahrer 2018). More than three decades ago, Martín-Barbero contributed to decenter and resituate the investigation of media within the cultural negotiations and social interactions of concrete actors in Latin America. Departing from the reductionism and media-centrism of North American functionalist analyses, he explored everyday practices of media appropriation through which social actors enacted resistance and resilience to domination and hegemony within their specific cultural matrixes. The move was inherently political: looking at how communication was being shaped in bars, squares, homes, and the like, it refocused the gaze away from media institutions toward people and their heterogeneous cultures, restoring the agency of social actors participating in the communicative process intended as a process of production and exchange of multiple meanings.

A similar approach should motivate contemporary analyses of Big Data from the South. We suggest making the move from datafication to data activism/data justice to examine the diverse ways through which citizens and the organized civil society in the Souths engage in bottom-up data practices and resist a datafication process that magnifies oppression and inequality. Hence, the focus should shift from data as technical artifacts to the multiplicity of data practices and appropriations that human actors develop in a variety of social and historical contexts, resurrecting the sociological (rather than the cultural studies) notion of political agency in action (cf. Emirbaye and Mische 1988). In other words, we should pay attention to the forms of interstitial, hybrid knowledge emerging in subversive complicity with the system (Castro-Gómez and Grosfoguel 2007, 20, our paraphrasis).

This move is tightly interwoven with an agenda of socially engaged scholarship and a kind of research that is strongly related to its environments, rather than enclosed within institutional walls. It is important to stress also that the focus on data as practice does not entail a disregard for the materiality of data and the importance of infrastructures since materiality is a constitutive, essential component of practice (and of injustice, too). Rather, it means casting such materiality within the fabric of social life and the several spaces where humans interact and appropriate data in multiple, often unpredictable ways. To be sure, critical data studies have inaugurated an operation similar to what Latin American scholarship did with Martín-Barbero. However, we need to go one step forward, connecting data practices to alternative epistemologies—an operation we attempt in the next paragraph.

Unleashing Novel Data Imaginaries

Existing attempts to foreground agency in data studies appear limited as they variably fail to connect bottom-up practices to the emergence of novel epistemologies and ways of thinking/feeling. In this section, we posit the notion of imaginaries as the connective tissue able to couple critical analysis with a prefigurative agenda of change inspired to decolonial thinking and working toward epistemic justice.

A sense of ineluctability pervades contemporary reactions to datafication. This new “information order” constitutes a “21st century imaginary” centered on the discursive and institutional normalization of surveillance infrastructure (Turow et al. 2015). Social actors—activists included—appear to have resigned to a sort of “surveillance realism,” where the normalization of data exploitation and surveillance forecloses our possibilities to even imagine possible alternatives (Dencik and Cable 2017). In this context, the radical imagination (Khasnabish and Haiven 2014) that lies at the core of social movements is voided and severely undermined. But a myriad of non-mainstream ways of imagining/thinking/feeling data emerges in the fringes, subtending to the creation of alternative data practices (while in turn being fed by such practices).

We believe that understanding Big Data from the South entails the engagement with a plurality of uncharted ways of actively (re)imagining processes of data production, processing, and appropriation. Imaginaries (cf. Lehtiniemi and Ruckenstein 2019) can be seen as the trait d’union between grassroots data practices and the emergence of alternative data epistemologies (C. Milan and van der Velden 2016). Paraphrasing Rodriguez (2001, 20), we consider data practices from the Souths as forms of intervention and transformation of the established order; these interventions contest “social codes, legitimized identities, and institutionalized social relations” and empower social actors “to the point where these transformation and changes are possible.” In other words, imaginaries have the ability to embed and embody prefigurative realities capable of producing change.

But sometimes imagining alone is not enough to act upon values and visions for alternative futures. We thus call for a novel alliance between the “skilled learners” on the ground and academic observers, who can together work toward a thorough “Southern” theory of change for the datafied society. Grasping how the socio-critical imaginaries around data inhabiting the Souths can break the spiral of data inevitability absorbing us, such theory of change could contribute to overcome current forms of injustice, forging spaces of autonomy, liberation, and resistance. But this exploration of alternative data imaginaries and epistemologies needs to be infused with mutual respect, constantly keeping watch to prevent the replication of dynamics of domination, extraction, exploitation, and oppression.

Conclusions and New Beginnings

With this contribution, we aim to stimulate a dialogue and move forward the critical debate around datafication, to promote a reparation to the cognitive injustice (Santos 2014) that fails to recognize nonmainstream ways of knowing the world through data. Much work remains to be done to integrate and extend the five axes that we have outlined into a full-fledged research agenda on Big Data from the Souths. In times of pervasive and unequal datafication, this is an urgent conversation if we are to remove the mask of data universalism that conceals and justifies the mistakes, the biases, and the dangerous assumptions of the past, while depriving the Souths of their knowledge, richness, and freedom. It is also paramount, however, to put research at the service of a broader project of creation of a comprehensive theory of change for the age of datafication—one inspired to decolonial thinking and that puts epistemic justice at the core. It is only by recognizing the “need of a body-politics of knowledge without pretense of neutrality or objectivity” that we can make the move toward the creation of “alternative forms of ethical rationalities and also new forms of utopias” (Castro-Gómez and Grosfoguel 2007, 21, own translation).