DATAFICATION AS MASTERY
Across policy, industry and academic discourse, datafication is widely promoted as a means of rendering social life more efficient, predictable, and governable. From early-warning systems capable of detecting infectious disease outbreaks with unprecedented speed, to smart city platforms that optimise urban flows, to digital identity systems promising frictionless access to public services, datafication is often framed as a benign extension of technological progress. Yet this optimistic narrative sits uneasily alongside growing evidence that the same infrastructures that generate public value also deepen inequalities, enable new forms of surveillance, and consolidate power in the hands of a small number of institutions and corporations. The puzzle, then, is how to account for this dual character: why does datafication produce genuine public benefits while simultaneously generating structural harms?
This tension resonates with the theme of mastering nature: the idea that modernity has long sought to make the world legible, calculable, and manipulable as a means of ordering both nature and society. As scholars of science and technology studies argue, projects of “mastery” operate by simplifying complex phenomena into manageable objects, whether forests, labouring bodies, or, today, datafied populations. Datafication extends this modernist project by treating human behaviour and social relations as extractable and governable resources to be measured, predicted, and optimised, thereby transforming sociality itself into a domain of technical control.
Building on work in science and technology studies, surveillance studies, and political economy, this essay argues that datafication’s benefits are structurally dependent on, rather than merely accompanied by, logics of surveillance, classification, and colonial extraction. While much of the existing literature analyses these dynamics in isolation (emphasising either efficiency gains, surveillance harms, or data colonialism), this essay shows how they operate together as a coherent political project of mastery. The first section examines the instrumental case for datafication, acknowledging its contributions to public health, governance, and civic participation while identifying the limitations that optimistic accounts obscure. The second section moves beyond instrumental evaluation to analyse how data infrastructures function as regimes of surveillance and ontological politics, actively constructing the categories and visibilities through which governance occurs. The third section situates these mechanisms within the longer historical trajectory of colonial political economies, arguing that contemporary datafication reproduces patterns of extraction, infrastructural domination, and epistemic imposition. The essay concludes by considering what non-colonial forms of datafication might require.
1 The Benefits and Limits of Datafication
Proponents of datafication emphasise its capacity to enhance public welfare, improve institutional decision-making, and expand opportunities for civic participation. At this instrumental level, datafication appears to fulfil a longstanding modern ambition: rendering complex social phenomena legible and governable. Nowhere is this more apparent than in public health, where data infrastructures have enabled earlier detection of infectious disease outbreaks and more targeted interventions. Meckawy et al.’s (2022) systematic review of Early Warning Systems (EWSs) synthesises sixty-eight studies across diverse contexts and concludes that syndromic surveillance systems often outperform traditional reporting channels, detecting anomalies days or even weeks earlier. Sankoh’s (2013) work on demographic surveillance in low- and middle-income countries similarly shows that population-based data reveal mortality trends and service gaps that would otherwise remain obscured. Together, these examples demonstrate how datafication can materially strengthen institutional capacities to monitor, predict, and respond to public health challenges.
Datafication also underpins broader governance innovations. Kitchin (2014) notes that real-time data streams, sensor networks, and algorithmic analytics enable forms of “smart governance” centred on continuous monitoring and reactive optimisation. Urban administrations increasingly rely on dashboards and predictive analytics to prioritise resources and inform planning decisions (Tenney & Sieber 2016). Such developments echo earlier technocratic efforts to master natural environments through quantification and calculation, now applied to social environments through dynamic modelling and prediction.
However, even the optimistic literature acknowledges significant limitations. First, the benefits of data-driven governance depend heavily on data quality, representativeness, and institutional capacity, conditions unevenly distributed across and within societies. As Meckawy et al. (2022) observe, EWS performance varies widely depending on resources, standardisation, and training; in low-income settings, improving existing infrastructures is often more effective than deploying complex new systems. Datafication thus tends to amplify existing institutional strengths and weaknesses rather than resolve them.
Second, claims that datafication expands democratic engagement often overlook how platforms structure participation. While Koc-Michalska et al. (2016) identify increased opportunities for civic engagement facilitated by digital media, these benefits accrue disproportionately to individuals with greater digital literacy and socioeconomic advantage. Tenney & Sieber (2016) similarly highlight how algorithmic filtering and interface design privilege certain forms of political expression while obscuring others, narrowing rather than widening the participatory sphere. Datafication therefore shapes not only the capacity for engagement but also who becomes legible to political institutions and on what terms.
Third, the epistemological assumptions underpinning optimistic accounts have been widely challenged. Stevens et al.’s (2018) discourse analysis reveals that over 60 per cent of healthcare editorials frame data as a neutral resource that “captures” reality, while only a small minority adopt critical-interpretive perspectives. This dominance of naïve empiricism obscures the constructed nature of data: what is measured, how categories are defined, and which phenomena become legible all depend on human decisions embedded within institutional and political contexts. As Frické (2015) and Kitchin (2014) caution, treating data as inherently objective can produce misleading or harmful conclusions.
Finally, emerging research demonstrates that even well-intentioned uses of datafication can entrench inequality. Mann and Matzner (2019) show that algorithmic profiling produces opaque forms of discrimination that evade traditional regulatory frameworks, while van Eijk (2017) illustrates how risk assessment tools encode socioeconomic marginality as “risk,” reinforcing punitive outcomes for poorer individuals. Rather than eliminating structural biases, datafication may instead obscure or automate them.
These limitations signal the need to move beyond instrumental evaluation toward analysing the deeper mechanisms through which datafication orders social life. The next section therefore examines how data infrastructures function as systems of surveillance and world-making that classify, sort, and shape populations in ways that are neither neutral nor evenly distributed.
2 Surveillance, World-Making and the Ontological Politics of Datafication
While Section 1 outlined the instrumental benefits of datafication, these advantages reveal little about the deeper political mechanisms through which data infrastructures operate. This section argues that datafication functions as a regime of surveillance and world-making, producing new forms of visibility, legibility, and governance that reconfigure relations between states, citizens, and corporations. Examining these mechanisms clarifies why the benefits identified earlier are unevenly distributed and why datafication often intensifies inequality despite its promise of improved governance.
2.1 Surveillance as Social Sorting
Surveillance studies demonstrate that contemporary surveillance is not merely observation but the systematic production and use of data for classification and control. Lyon (2010) describes surveillance as an “organising practice” that sorts populations into categories (desirable or risky, eligible or excluded) based on data profiles. These categories shape access to rights, services, and opportunities; surveillance is therefore distributive, allocating social goods and burdens according to data-driven logics.
The shift from watching individuals to sorting data has profound implications for governance. Lyon’s (2014) analysis of the Snowden revelations shows how Big Data enables states and corporations to engage in continuous monitoring, risk assessment, and prediction, blurring boundaries between public and private authority. Platforms and governments increasingly operate through shared infrastructures: security agencies rely on platform-generated data, while platforms depend on state regulation to legitimate their extraction and processing practices.
Ruppert, Isin, and Bigo (2017) conceptualise this entanglement as “data politics.” Datafication does not merely document existing populations; it produces “data citizens,” transforming individuals into assemblages of attributes and behavioural traces that circulate across bureaucratic, commercial, and algorithmic systems. Decisions about eligibility or risk rest not on lived experience but on abstractions encoded in digital profiles, profiles that individuals cannot fully access or contest.
The distributive consequences are substantial. As noted in Section 1, van Eijk (2017) demonstrates that sentencing algorithms encode socioeconomic marginality as “general risk,” resulting in harsher outcomes for poorer defendants, while Mann and Matzner (2019) show that algorithmic profiling creates discriminatory effects that elude traditional legal frameworks because harms arise from inferences rather than explicit categories. These are not malfunctions but structural features of surveillance architectures. Even when deployed for ostensibly beneficial purposes (such as healthcare triage or social service prioritisation) data-driven sorting reproduces disadvantage because classification itself embeds assumptions about risk, worth, and eligibility. For example, welfare fraud detection systems deployed in several European countries use algorithmic risk scores to flag recipients for investigation, disproportionately targeting low-income and racialised populations while rendering middle-class fraud statistically invisible (van Eijk 2017).
2.2 Datafication as Ontological Politics
Beyond sorting, datafication reshapes the social world by constructing categories, identities, and phenomena rather than neutrally representing them. This is the realm of “ontological politics”, a concept from science and technology studies that refers not to abstract metaphysical claims about being, but to the ways socio-technical systems actively enact particular versions of reality. In this sense, data infrastructures participate in world-making by determining what entities exist, which attributes matter, and how relations become visible to authority.
Latour (1996) argues that socio-technical systems (such as laboratories, sensors, and databases) function as “centres of calculation” that transform complex realities into simplified, mobile representations. These representations underpin governance, expertise, and intervention, yet they are not mirrors of an independent world; they are products of networks, inscription devices, and institutional practices. Gorur and Dey (2020) extend this insight to digital platforms, showing how interfaces do not merely mediate interaction but actively define what a “user” is, which actions are possible, and which behaviours become visible. Seemingly mundane decisions about data fields, algorithmic parameters, or interaction flows thus enact specific social realities while marginalising others.
Bowker and Star’s (2005) analysis of classification infrastructures deepens this argument. Categories do not pre-exist data systems; they are made by them. Classification embeds moral, political, and economic assumptions that, once operationalised, come to appear natural and objective. The power of datafication lies in this naturalisation: contestable decisions become infrastructural facts, shaping how people are known, governed, and valued.
The ontological effects of datafication are further reinforced by the material infrastructures that sustain it. As Pickren (2018) shows, contemporary data systems rely on a “global assemblage of digital flow” composed of data centres, fibre-optic cables, energy systems, and urban landscapes. These infrastructures do not simply support data-driven governance; they actively constitute socio-technical realities by reorganising space, labour, and resources in ways that determine which forms of life are rendered legible and governable. Ontological politics therefore operates not only at the level of representation and classification, but through the infrastructural conditions that enact datafied worlds.
Taken together, these insights show that datafication does not merely capture the social world but actively constructs it through classificatory schemes and technical architectures embedded in material infrastructures. Data-driven governance thus rests on ontological decisions about what, and who counts.
2.3 From Description to Critique
Surveillance and ontological politics illuminate how datafication operates, but they do not fully explain why these systems consistently benefit certain actors over others. Latour’s actor-network theory (ANT) is valuable for tracing how socio-technical systems come together, yet, as Couldry (2020) argues, its flat ontology cannot account for structural power, inequality, or domination. ANT describes how networks are assembled but not why technology firms, security agencies, and wealthy states occupy privileged positions within them.
This limitation matters because datafication is not merely a socio-technical arrangement but a political-economic system. Understanding why certain actors hold disproportionate control over data infrastructures requires analysing the historical and material conditions that organise extraction, ownership, and global asymmetry. Surveillance and world-making reveal the mechanisms; they do not explain the structure that gives those mechanisms their direction and beneficiaries.
The next section therefore situates datafication within the longer trajectory of colonial political economies, showing how contemporary data infrastructures reproduce patterns of extraction, dependency, and epistemic domination characteristic of colonial rule.
3: Datafication as Digital Colonialism
The mechanisms examined in Section 2 reveal how datafication structures social life, but not why these structures consistently privilege certain actors while marginalising others. To understand this pattern, datafication must be situated within a broader historical and political-economic trajectory. This section argues that contemporary datafication constitutes a form of digital colonialism: a mode of extraction, appropriation, and domination that extends the logic of historical colonial projects through digital infrastructures. Viewing datafication through this lens clarifies not only its harms but also the conditions enabling its expansion. Far from a neutral or inevitable technical evolution, datafication depends on structural asymmetry, infrastructural dependency, and the cheap appropriation of life, features characteristic of earlier forms of colonial domination.
3.1 Cheap Data as the New “Cheap Nature”
Moore’s (2015) theory of cheap nature provides a powerful framework for understanding capitalism’s expansion through the appropriation of new domains of life as low-cost inputs. In Moore’s world-ecology perspective, capitalism does not simply exploit nature; it produces nature as an externalised realm to be mastered, quantified, and extracted. Turning forests into timber or bodies into labour required infrastructures that made them legible and exploitable.
Datafication operates through an analogous logic. Couldry and Mejias (2019) argue that digital platforms engage in the continuous appropriation of human life (behaviour, relationships, attention) as raw material for extraction. Their concept of data colonialism describes the systematic capture and conversion of human experience into data that can be analysed, monetised, and controlled. Just as colonial enterprises treated land, minerals, and labour as free or cheap, contemporary data infrastructures treat behavioural traces as a resource available for unilateral extraction.
Zuboff (2015) shows how technology firms convert such “behavioural surplus” into predictive products sold to advertisers, insurers, and governments. This extraction is non-consensual and structurally embedded: users cannot meaningfully refuse participation, and value overwhelmingly accrues to a small cluster of corporations. Taylor’s (2016) account of data capitalism reinforces this analysis, demonstrating how platforms unilaterally determine what is collected, how it is processed, and who benefits. Individuals become sources of data rather than claimants to its value, paralleling colonial arrangements in which populations supplied raw materials for imperial markets without access to profits or decision-making.
Through these processes, datafication extends the modernist project of mastering nature. The object of mastery, however, is no longer the external environment but human sociality itself, recast as cheap data.
3.2 Infrastructural Imperialism
Digital colonialism is not merely metaphorical; it is embedded in the infrastructures that enable data extraction. Kwet (2019) shows that U.S.-based technology firms control vast portions of the global digital ecosystem (operating systems, cloud platforms, undersea cables, standards, and social media) and that this concentration constitutes an imperial architecture through which a handful of corporations shape global communication and governance.
This infrastructural dominance produces forms of dependency closely resembling colonial economic arrangements. India’s Aadhaar system, the world’s largest biometric identification programme, relies extensively on foreign technology providers for hardware, software, and cloud infrastructure. Although framed as a tool for inclusion and efficient service delivery, Aadhaar has raised significant concerns about surveillance, exclusion, and data vulnerability (Anand 2021), demonstrating how infrastructural dependency can limit sovereignty and policy autonomy. Likewise, Facebook’s Free Basics initiative offered zero-rated internet access across the Global South but was widely criticised for creating a walled garden that positioned Facebook as a gatekeeper to the internet, channelling users toward its services while foreclosing alternatives (Sambuli 2016).
Similar dynamics appear at the urban scale. Tenney & Sieber (2016) show that city governments frequently outsource sensing and analytics systems to private vendors, embedding proprietary standards and centralising data flows in corporate hands. Lips (2010) illustrates how digital identity systems implemented through contracted technologies reshape citizen–state interactions in ways that privilege those who can be easily datafied while excluding those who cannot.
These forms of infrastructural power reproduce colonial logics: control over the means of communication and representation, extraction of value from peripheries, and imposition of socio-technical regimes aligned with the interests of dominant centres.
3.3 Epistemic Colonialism
Beyond economic and infrastructural domination, digital colonialism operates at the level of epistemology, privileging particular ways of knowing while marginalising others. Ricaurte (2019) argues that datafication extends the colonial matrix of power by universalising Western, quantification-oriented epistemologies. The assumption that the world is best understood through discrete, measurable data points reproduces colonial tendencies to render diverse life-worlds commensurable and governable through abstract categories.
When these epistemic assumptions become embedded in global data infrastructures, they reinforce asymmetries in whose knowledge matters. Indigenous, relational, or non-quantifiable ways of knowing are rendered invisible or unintelligible within systems designed around Western ontological assumptions. Lewis et al.’s (2024) work on “abundant intelligences” shows how dominant AI architectures exclude Indigenous epistemologies centred on relationality and interdependence, reproducing epistemic colonialism even in contexts positioned as sites of innovation. Classification thus becomes a means of asserting authority, defining what counts as real, and determining who is legible to power.
3.4 Counterarguments
Critics caution that characterising datafication as colonial risks metaphorical overstretch, since digital infrastructures entail neither territorial conquest nor direct administrative rule. Yet, as Couldry and Mejias (2019) emphasise, the salient continuity lies not in institutional form but in function: the large-scale appropriation of human life for economic gain, legitimised through narratives of progress and development.
A second objection concerns the agency of Global South states. Some governments voluntarily adopt platform infrastructures to expand connectivity or enhance service delivery. However, such choices often occur under conditions shaped by structural constraints (market monopolisation, lack of domestic capacity, and absence of viable alternatives) limiting meaningful agency (Kwet 2019). Finally, as Section 1 established, datafication does produce genuine public benefits. The argument here is not that datafication is inherently harmful, but that its current organisation is structurally unequal and historically patterned. These benefits do not negate its colonial dimensions; they often rely on the same extractive infrastructures that generate dependency and asymmetry.
Conclusion
This essay has argued that datafication’s benefits are structurally dependent on colonial logics of extraction, and that surveillance and classificatory world-making are the mechanisms through which this dependency is reproduced. Although datafication demonstrably enhances public health surveillance, streamlines service delivery, and expands opportunities for engagement, these advantages are inseparable from the processes through which data systems render social life legible. Such processes do not merely facilitate efficient governance; they construct new forms of visibility and vulnerability that systematically redistribute power. By transforming behaviour, identity, and social relations into quantified objects of analysis, datafication extends the modernist ambition of mastering complexity, an ambition historically entangled with modes of domination that this essay has traced.
Situating datafication within this trajectory clarifies why its effects cannot be adequately understood through the language of innovation or optimisation. Contemporary data infrastructures reproduce global hierarchies through the extraction of behavioural surplus, the imposition of proprietary technological standards, and the privileging of quantification-oriented epistemologies. Individuals and states alike are positioned as sources of raw material for distant centres of power. The benefits that datafication delivers are real, but they remain conditional on asymmetries embedded in its foundational architectures.
Recognising datafication as a form of digital colonialism does not imply rejecting data-driven governance altogether. Rather, it underscores the need to reimagine the ownership, governance, and epistemological foundations of data systems. The CARE Principles for Indigenous Data Governance (2020), emphasising collective benefit, authority to control, responsibility, and ethics, provide one meaningful alternative to extractive data relations. Initiatives such as the Māori Data Sovereignty Network demonstrate how Indigenous communities can assert control over how data about them is collected, stored, and used, grounding governance in relational accountability rather than unilateral appropriation. Emerging experiments in data cooperatives and community-controlled infrastructures likewise gesture toward alternative socio-technical arrangements.
These efforts remain limited in scope and face significant structural constraints, yet they illustrate that non-colonial data futures are not merely aspirational but practically imaginable. Realising them requires more than technical reform: it demands a fundamental rethinking of what it means to treat human life as a resource to be mastered, measured, and monetised, and a commitment to building data systems and political imaginaries that reject mastery as the organising principle of technological governance.
Bibliography
Anand, N., 2021. New principles for governing Aadhaar: Improving access and inclusion, privacy, security, and identity management. Journal of Science Policy & Governance, 18(01).
Bowker, G. & Star, S.L., 2000. Sorting Things Out: Classification and Its Consequences. Cambridge, MA: MIT Press.
Carroll, SR, et al. 2020. The CARE Principles for Indigenous Data Governance. Data Science Journal, 19: 43, pp. 1–12.
Couldry, N. 2020. Recovering critique in an age of datafication. New Media & Society, 22(7), 1135-1151.
Couldry, N. & Mejias, U.A., 2019. The Costs of Connection: How Data Is Colonizing Human Life and Appropriating It for Capitalism. Stanford: Stanford University Press.
Frické, M., 2015. Big Data and Its Epistemology. Journal of the Association for Information Science and Technology, 66(4), pp.651–661.
Gorur, R. & Dey, P., 2020. Making the User Friendly: The Ontological Politics of Digital Data Platforms. In: Digital Platforms and the Social.
Kitchin, R., 2014. Big Data, New Epistemologies and Paradigm Shifts. Big Data & Society, 1(1), pp.1–12.
Koc-Michalska, K., Lilleker, D., & Vedel, T., 2016. Civic Political Engagement and Social Change in the New Digital Age. Media, Culture & Society, 38(5), pp.793–810.
Kwet, M., 2019. Digital Colonialism: US Empire and the New Imperialism in the Global South. Race & Class, 60(4), pp.3–26.
Latour, B., 1996. On Actor-Network Theory: A Few Clarifications. Soziale Welt, 47(4), pp.369–381.
Lewis, J.E., Arista, N., Pechawis, A. & Kite, S., 2024. Abundant Intelligences: Reimagining AI through Indigenous Epistemologies. Placing AI within Indigenous knowledge frameworks. Ai & Society 40(4) 2141-2157.
Lips, A.M.B., 2010. Rethinking Citizen–Government Relationships in the Age of Digital Identity: Insights from Research. Information Technology & People, 23(4), pp.348–367.
Lyon, D., 2010. Surveillance Studies: An Overview. Cambridge: Polity Press.
Lyon, D., 2014. Surveillance, Snowden, and Big Data: Capacities, Consequences, Critique. Big Data & Society, 1(2), pp.1–13.
Mann, M. & Matzner, T., 2019. Challenging Algorithmic Profiling: The Limits of Data Protection and Anti-Discrimination in Responding to Algorithmic Harms. Big Data & Society, 6(2), pp.1–11.
Meckawy, A., et al., 2022. Early Warning Systems for Infectious Diseases: A Systematic Review. Frontiers in Public Health, 10, pp.1–15.
Moore, J.W., 2015. Capitalism in the Web of Life: Ecology and the Accumulation of Capital. London: Verso.
Pickren, G. 2016. ‘The global assemblage of digital flow’: Critical data studies and the infrastructures of computing: Critical data studies and the infrastructures of computing. Progress in Human Geography, 42(2), 225-243.
Ricaurte, P., 2019. Data Epistemologies, the Coloniality of Power, and Resistance. Television & New Media, 20(4), pp.350–365.
Ruppert, E., Isin, E. & Bigo, D., 2017. Data Politics. Big Data & Society, 4(2), pp.1–7.
Sambuli, N., 2016. Challenges and opportunities for advancing Internet access in developing countries while upholding net neutrality. Journal of Cyber Policy, 1(1), pp.6174.
Sankoh, O., 2013. Population Surveillance in Low- and Middle-Income Countries: The INDEPTH Network. Global Health Action, 6(1), pp.1–8.
Stevens, G., et al., 2018. Conceptualizations of Big Data and Their Epistemological Claims in Healthcare: A Discourse Analysis. BMC Public Health, 18, pp.1–11.
Taylor, L., 2016. The Ethics of Big Data as a Public Good: The Case of Data Capitalism. Philosophy & Technology, 29(4), pp.327–345.
Tenney, M., & Sieber, R. (2016). Data-Driven Participation: Algorithms, Cities, Citizens, and Corporate Control. Urban Planning, 1(2), 101-113.
van Eijk, G., 2017. Risk Assessment and the Criminal Justice System: Encoding Social Marginality. Critical Criminology, 25(3), pp.393–410.
Zuboff, S., 2015. Big Other: Surveillance Capitalism and the Prospects of an Information Civilization. Journal of Information Technology, 30(1), pp.75–89.