For over a century, global agriculture has been shaped by agricultural productivism. The ‘more is better’ mantra was fueled by 20th-century industrial farm technologies, US farm policies that promoted overproduction, and expanded international trade in ever cheaper commodities. Today, as digitalization reaches the agrarian world, another kind of productivism is coming into view – where the goal is pumping out as much data as possible.

We’ve described this phenomenon as ‘data productivism’. Data productivism is the idea that the production of data is intrinsically socially desirable and that all parties benefit from increased output. This vision is being driven by a convergence of factors, including the crises generated by industrial agriculture, new technological capacities afforded by artificial intelligence (AI) and Big Data, and the prospects of connecting previously siloed datasets across the food system.

Global development institutions now suggest that more and better data holds the keys to ending global hunger, fighting climate change, and safeguarding farmer livelihoods. Agribusiness corporations like Bayer/Monsanto, as well as tech firms like Microsoft and IBM, are keen to invest in digitalization projects that accrue profit and monopoly control of data.

In recent research, we looked at how prominent actors in international development – including the World Economic Forum, the Food and Agriculture Organization (FAO), the World Bank, the Organisation for Economic Co-operation and Development (OECD), and the Consortium of International Agricultural Research Centers (CGIAR) – are driving data productivism forward. We found that they are building institutional networks: initiatives including the 50×2030 Initiative, Data4SDGs, GODAN, and the UN Food Systems Summit are all designed to push data projects ahead in Global South countries.

We also found that these actors and networks are telling evocative stories about the futures of datafied food systems. Sustainability stories paint data as a carbon-neutral path to planetary health. Inclusive innovation stories position women and girls, indigenous people, and communities of color as central data stakeholders. Cutting across these stories, we discovered a key strategy for invoking ‘data gaps’.

Gaps narratives recite colonial mantras like “empower the poor with knowledge”, when smallholder, indigenous, and worker communities already ‘have knowledge’ – and should have the power to decide what kind of data is generated, in what form, how, and where data is amassed and used, by whom, and for what purpose.

The data gaps narratives come in several varieties. One focuses on policymaking. For example, Data4SDGs proposes: “Unreliable, patchy, and out-of-date data on agri-food systems means that many countries are ‘flying blind’ when it comes to developing evidence-informed policies for reducing hunger, improving food security and nutrition, and ensuring sustainable food supplies.”

This narrative is typically invoked in the context of Africa and other world regions said to lack basic survey data on which national statistical systems can be built. For instance, the 50×2030 Initiative – a 10-year, USD 500 million joint project between the FAO, the World Bank, and the IFAD – was launched explicitly to “close the agricultural data gap” by 2030.

Another variant of gaps narratives emphasizes the equity dimensions of data. Mark Holderness, the Executive Director of GFAR, says in the preface to a key Godan report: “Truly sustainable development must empower the poor with knowledge, realizing the benefits from data access and use and minimizing their risks, such that ‘no one is left behind’”.

These stories imply that many of the problems facing food systems stem from a lack of data. They are an open invitation to help fill these gaps, particularly in the Global South, where since colonization, the rhetoric of scarcity – of technology, of knowledge, of modernity, of adaptability to a changing world – has been mobilized to legitimize and facilitate newer forms of extraction.

Moreover, by fixating on data, these stories reveal a simplistic assumption: that data and data-driven technologies are unmitigated goods. Rather than confronting the well–documented risks of surveillance, appropriation, deskilling, discrimination, and more, the only ‘data equity’ problem is improving access to data’s unalloyed benefits. Gaps narratives recite colonial mantras like “empower the poor with knowledge”, when smallholder, indigenous, and worker communities already ‘have knowledge’ – and should have the power to decide what kind of data is generated, in what form, how, and where data is amassed and used, by whom, and for what purpose.

Narratives of gap-filling are not new. The contemporary push for more data in agriculture resembles older forms of ‘agricultural productivism’ which marshaled science for geopolitical ends. In the mid-20th century, a coalition of states and the private sector, most significantly the US, promoted filling ‘yield gaps’ as part of US farm policy. Through the Green Revolution, the US sought to reinforce its economic and political power by exporting its industrial agriculture approach, while also shedding its grain surpluses on foreign markets. Substantial research now shows that the emphasis on yield not only failed to end hunger, but also degraded environments and fostered dependency on corporate inputs and global markets.

Through the Green Revolution, the US sought to reinforce its economic and political power by exporting its industrial agriculture approach, while also shedding its grain surpluses on foreign markets. Substantial research now shows that the emphasis on yield not only failed to end hunger, but also degraded environments.

Today, the crises wrought by agricultural productivism are catalyzing widespread demands for food system transformation. At the same time, technical advances are supercharging the prospects for collecting and analyzing data at scale. In this context, those who once sought to promote agricultural productivism are now coalescing around a new resource to fuel the engines of capitalist growth.

Data productivism, unfortunately, seems to repeat the perils and contradictions of analog pasts. It doesn’t recognize the expensive environmental tolls of data – Google’s global data centers used over 4.3 billion gallons of water in 2021. It also conceals what is likely to be a new era of colonial plunder as data from the Global South gets siphoned off to corporations and states in the Global North.

Crucially, data productivism distracts us from acting upon what we already know. Decades of scholarship, building on millennia of indigenous and peasant knowledge, have called attention to structural and cultural changes needed to nourish human and ecological wellbeing in the long-term. We know key ingredients will be agroecological production, land redistribution, reforms to liberalized trade, revitalizing commons for knowledge, land, water, and seed.

Yet data productivism buries this wisdom. As Dr. Ruha Benjamin argues, “Demanding more data on subjects that we already know much about is, in my estimation, a perversion of knowledge – the datafication of injustice, in which the hunt for more and more data is a barrier for acting on what we already know.”

But data productivism is not inevitable. Calls for data justice and data sovereignty are emerging from a wide range of social movements across the Majority and Minority Worlds. Movements are increasingly addressing not only issues of privacy, surveillance, and the political economy of data, but also the ways datafication propagates colonization, hindering local development and fostering dependency on the data infrastructures of global tech corporations.

While experiments with digital public goods, public IP commons, and digital sovereignty often focus on urban spaces, scientific R&D, communications platforms, and state social services like healthcare have reignited attention to ‘the commons’, a concept with deep roots in pre-capitalist agrarian history. Common-field agriculture once enabled peasants across much of the world to cultivate a wide variety of plants and animals for food, fuel, fodder, housing, clothing, and medicine before being enclosed through capitalist private property rights. Today, inspired by movements such as the Zapatistas, land is increasingly being ‘commoned’, or governed according to collectively negotiated protocols for access and use. Such systems are now gaining traction in the digital sphere, as reflected in the words of Nandini Chami of IT for Change: “The collective interests of farmers are not reducible to individual farmer consent. Data is relational, therefore, we must rethink how we view and define ownership of data.”

“Demanding more data on subjects that we already know much about is, in my estimation, a perversion of knowledge – the datafication of injustice, in which the hunt for more and more data is a barrier for acting on what we already know.”

Indeed, in 2022, IT for Change, in partnership with SEWA Cooperative Federation (an organization representing self-employed women workers), and Vrutti Livelihood Impact Partners organized a policy roundtable in which farmers, policymakers, and civil society groups gathered to discuss a farmer-centric strategy for confronting the digitization of Indian agriculture. Among the solutions proposed were ‘common data platform models’: systems designed to address the monopolization of the data platform economy by Microsoft, Meta, and Google. Commonized platforms, participants explained, would provide an alternative to corporate-run platforms using open networks and protocols to facilitate inclusive and equitable access and use.

While the political economy of ‘open’ data remains a hotly debated topic, what is clear is that communities worldwide are already breaking ground on systems that link agrarian pasts to emancipatory data futures. Moreover, they recognize that while scientists, engineers, and policymakers can make important contributions, especially in the domain of agriculture, the expertises of indigenous communities, peasants and smallholders, immigrant and migratory workers matter more. Data will not ever feed the world, they insist. People in living landscapes feed people, and data are just the relations that bind us together.