Data Colonialism

What data colonialism means

Data colonialism is the idea that today’s tech companies extract data from human life in much the same way that historical empires seized land, labour and resources, treating our everyday experience as territory to be claimed, owned and mined for profit. The term was given its fullest treatment by Nick Couldry of the London School of Economics and Ulises Mejias of the State University of New York, in their 2019 book The Costs of Connection. They define data colonialism as an emerging order for the appropriation of human life so that data can be continuously drawn from it for profit.

The comparison is deliberately uncomfortable. Couldry and Mejias are not saying that data collection is exactly the same as the violence of historical colonialism. They are arguing that it follows the same underlying logic. Where old empires claimed physical land and called it empty or ownerless, the new order claims human experience and treats it as a free resource that was simply lying around, waiting to be put to use.

Seen this way, the daily business of using apps and devices is not just a privacy issue. It is the front line of a much larger grab, in which the raw material being appropriated is human life itself.

How data colonialism mirrors historical colonialism

The power of the idea comes from a set of close parallels between the colonial past and the data present. Couldry and Mejias trace several of them.

Appropriation dressed up as improvement

Historical colonists rarely described themselves as thieves. They framed seizing land as bringing order, civilisation or progress to people who supposedly were not using it properly. Data colonialism uses a strikingly similar script. Gathering our data is presented as making life more convenient, more connected and more personalised, a gift rather than a taking. That framing, in which extraction is made to feel natural and beneficial, is a textbook example of cultural hegemony at work.

A new resource declared free for the taking

In 2011 a World Economic Forum report described personal data as a new kind of resource, as valuable to the modern economy as oil. The trouble with that phrase is that oil belongs to someone before it is extracted, whereas our data is treated as if it belongs to no one until a company collects it. The comparison also flatters the process by making it sound clean and technical, when what is being mined is the texture of people’s actual lives. This extends commodification into parts of life that were never bought or sold, turning conversations, friendships, footsteps and heartbeats into inputs for profit.

Dispossession and dependence

Colonialism did not only take things. It rearranged whole societies so they depended on the colonising power. Couldry and Mejias argue that data colonialism works the same way: once daily life runs through a handful of platforms, opting out becomes close to impossible. Our own activity is captured, processed and sold back to us as services we then cannot do without. There is a quiet alienation in this, where the product of our own lives is turned into something owned by others and used to manage us.

From labour relations to data relations

Couldry and Mejias make one further point that sets their argument apart. Industrial capitalism was built on labour relations: the deal, however unequal, in which people sold their time and effort for a wage. Data colonialism adds a new kind of bond on top, which they call the data relation. Here we hand over information about our lives for no wage at all, and often without noticing we are doing it. We are not employed by the platforms we feed; we are simply connected to them, and the connection itself is the source of value. This is part of why the arrangement can feel so harmless. Nobody clocks in, nobody is obviously exploited, and the extraction happens through the ordinary, even pleasant, act of staying in touch. The wage relationship made exploitation visible; the data relationship hides it inside convenience.

Data colonialism in everyday life

The clearest cases are the ones that feel most ordinary. A fitness app turns your runs into a profile sold to advertisers. A smart doorbell turns your street into a feed that police and companies can request. A free messaging service, offered at no charge in parts of the world where mobile data is expensive, becomes the gateway through which millions first meet the internet, with a single company sitting at the door and shaping what an open web looks like for an entire generation.

The idea did not arrive fully formed with one book. Earlier researchers, including the geographers Jim Thatcher, David O’Sullivan and Dillon Mahmoudi, were already describing data colonialism in 2016, drawing on the notion of accumulation by dispossession: the way capitalism expands by enclosing and privatising things that used to be held in common. Couldry and Mejias gave the idea its fullest and most influential treatment, but it sits within a longer line of thinking about how new resources get captured.

There is also a global dimension that the colonial framing makes visible. The AI systems trained on all this data depend on armies of low-paid workers, often in the Global South, who label images, moderate disturbing content and clean up messy datasets for very little money. The benefits flow largely to a small group of firms in a few countries, while much of the cost and risk lands on those with the least power. That uneven pattern of harm echoes structural violence, where damage is built into the system rather than caused by any single villain.

Because the deal so often feels like a fair exchange, it can produce a kind of false consciousness, in which we experience constant surveillance as a service we freely chose. The convenience is real, which is exactly what makes the arrangement so hard to see clearly.

Data colonialism, surveillance capitalism and technofeudalism

Data colonialism is one of three big attempts to name the same shift, and the three sharpen each other. Surveillance capitalism focuses on the business model, on how behavioural data becomes predictions that are sold. Technofeudalism asks what kind of economy all this extraction builds, and argues the answer looks more like feudal rent than ordinary profit. Data colonialism supplies the historical lens, insisting that we have seen this pattern of appropriation before and should learn from how it ended.

You do not have to accept every part of the colonial comparison to find it useful. Even as a question rather than a verdict, it reframes the choice in front of us. Instead of asking only whether our privacy settings are tidy, it asks who should own the data drawn from human life, on what terms, and whether a different arrangement is possible. Couldry and Mejias end their book not with despair but with a call for what they term a decolonial response: defending the parts of life that should not be turned into data at all, and insisting that people keep some say over the information their lives produce. Naming the pattern is the first move towards imagining an alternative to it, and history suggests that systems which look permanent can change once enough people stop treating them as natural.