There is a moment in Cory Doctorow's 2025 book Enshittification: Why Everything Suddenly Got Worse and What to Do About It where he pauses to pick a fight with Audre Lorde. He calls her "far smarter than I am about nearly everything," then insists that one of her most quoted observations, that the master's tools will never dismantle the master's house, is "manifestly wrong." His argument is that regulation, antitrust law, interoperability standards: these are the tools we need to dismantle the tech monopolies that have captured and degraded the internet.
I first encountered this weird argument on a Mastodon thread years ago, and it never sat right with me. It would be a defensible claim if that was the argument Audre Lorde was making in the first place. Something strange happens in that paragraph, something worth sitting with: a writer invoking a Black lesbian feminist theorist in order to dismiss an argument about tech platforms, which she never made. The phrase has been borrowed, stripped of its roots, and sent out to do labour it was never meant to do. And in the act of rebutting a meme, Doctorow becomes part of the meme's problem.
Doctorow treats Lorde's line as if it were a claim that instruments of established power can never be used against that power; he then rebuts that claim by pointing to antitrust, regulation, and interoperability. But that is not the argument Lorde was making in the speech he invokes.
Enshittification is a serious and important book, and the broader project of understanding platform decay as a structural, political phenomenon rather than a personal failing or a market accident is exactly the kind of thinking we need. I will argue two things in this blog post: first, that Lorde's "master's tools" speech is routinely misread in tech discourse as a general proposition about rejecting reform; second, that taking her actual argument seriously would not invalidate enshittification but deepen it, by asking whose experience of technological harm the framework centers, and whose it treats as supplementary.
Where the "Master's Tools" Actually Comes From
On September 29, 1979, Audre Lorde stood at a podium at New York University at a conference celebrating the thirtieth anniversary of Simone de Beauvoir's The Second Sex. She had been invited at the last minute to respond to a panel, the only panel in the entire three-day conference where Black feminists and lesbians had been given a voice. She had been listed on the programme not as a speaker but as a "consultant." The organizers had corralled the women of colour and the queer women into a single slot on the final day, the implied message being that these perspectives were supplementary rather than foundational.
Lorde's speech, later published in Sister Outsider as "The Master's Tools Will Never Dismantle the Master's House," reads in this context not as a claim that reform is impossible or that working within systems is always futile, but as a specific intervention directed at white feminist academia, delivered from a position of structural exclusion, about how the very structures of knowledge production (conferences, panels, theoretical frameworks, the categories of analysis) were reproducing the hierarchies they claimed to oppose.
In this reading, the "tools" Lorde was naming were not literal instruments but epistemic ones: the frameworks of thought, the methods of inquiry, the structures of inclusion and exclusion that had been built by and for a particular kind of subject (white, heterosexual, Western) and that continued to operate even within movements ostensibly committed to liberation. The questions she poses make this frame legible: what does it mean to conduct feminist analysis while systematically excluding the voices of poor women, Black women, Third World women, lesbians? What does it mean to theorize liberation using categories that treat those exclusions as incidental rather than structural?
"What does it mean," she said, "when the tools of a racist patriarchy are used to examine the fruits of that same patriarchy? It means that only the most narrow perimeters of change are possible and allowable."
Her words suggest that such analysis could produce, at best, incremental adjustments within the existing order. It could let you "temporarily beat him at his own game," but not the genuine change, the "dismantling of the house", that would require what the existing frameworks seemed structurally incapable of generating: a reckoning with difference, with the knowledge that comes from the margins, with the ways that interconnected systems of oppression cannot be addressed one at a time by theorists who treat their own particularity as universal.
This is not a comfortable argument, and most crucially, it was not addressed to tech companies or policy makers or open source developers. What strikes me most about it is who it was addressed to: white feminist scholars, the people ostensibly on the same side, named as part of the problem. That specificity is often absent from contemporary tech discourse citations of the phrase.
What Tech Discourse Does With the "Master's Tools"
From my perspective, in free software and adjacent communities, the phrase has taken on a life entirely detached from its origins. I myself have probably engaged with it that way, to be completely honest. Micah White described it as "the atomic bomb of discussion enders," a phrase so potent it can be applied to absolutely any argument about strategy and method, often by people who have never encountered the full speech. White's concern is not with Lorde's argument but with what has been done to it: a revolutionary provocation flattened into a reflexive shutdown. It shows up most often as a thought-terminating cliché: using corporate infrastructure, proprietary platforms, or mainstream legal mechanisms to advance liberation goals is self-defeating by definition. The master's tools. End of conversation.
And it shows up as a straw person waiting to be knocked down. This is, I believe, what happens in that Enshittification paragraph: cite the phrase, acknowledge the authority of the speaker, then dismiss the claim as "manifestly wrong" on the grounds that regulation and antitrust law demonstrably can constrain monopoly power.
I'm not the first person to point this out. A review in DesignWhine called the Lorde passage "a particularly discordant note," observing that Doctorow's rebuttal addresses tech platforms rather than the "frameworks of racial and gender oppression that Lorde was addressing."
Both the use of the phrase as a cliché and as a straw person share a common structure. They treat it as a proposition about strategy, a claim about whether instruments of power can be redirected, and then debate it as such. In doing so, they evacuate the phrase of its specific context. Read against the speech itself, Lorde's argument seems less concerned with whether antitrust law can break up monopolies and more with whose knowledge counts, who gets to define the problem, and what gets systematically erased when liberation movements reproduce the exclusions of the systems they are opposing.
The sociologist Boaventura de Sousa Santos coined the word epistemicide precisely to name that kind of destruction: the killing, silencing, or devaluing of a knowledge system, what he called the "death of the knowledge of the subordinated culture."
How Taking the "Master's Tools" Seriously Can Serve Enshittification
Doctorow's enshittification thesis, at its core, is about how platforms extract value from users and suppliers in a structured, predictable way once they have achieved sufficient monopoly power. It is a story about the concentration of capital, the elimination of competitive discipline, the regulatory capture that allows monopolists to write the rules of their own industries. It is, in important ways, a structural analysis of power.
But there is a structural analysis that Enshittification doesn't seem to be making, and that taking Lorde's speech seriously rather than as a meme might push us toward. Whose experience of platform decay is centered in the enshittification story? Whose experience is supplementary? When we talk about the internet "getting worse," what baseline of "better" are we implicitly invoking, and for whom was it better?
The enshittification story, at its most powerful, describes a process by which platforms that once served users well came to exploit them. But this framing assumes a prior state of genuine service, a golden age of the open internet, that was for many people never particularly golden. The early internet was structured around the assumptions of its architects: predominantly white, male, Western, educated, and abled. The exclusions weren't mere flaws; they were perhaps, if we're being generous, deeply structural tendencies that were never seriously interrogated.
Consider one concrete example of what the enshittification framing tends to make invisible. In Nairobi, Kenya, described by one journalist as ground zero in a battle over the future of content moderation, young workers from informal settlements were employed by Meta's outsourcing partner Sama to review graphic content including beheadings, child abuse, and torture, for wages of between $1.46 and $3.74 per hour. A clinical assessment submitted to Kenya's Employment and Labour Relations Court in late 2024 found that 81 percent of these workers had developed severe PTSD. 185 of them have since sued Meta and Samasource Kenya. In a 2024 townhall, Kenyan President William Ruto reportedly assured Sama: "Now we have changed the law, so nobody will take you to court again on any matter."
These workers do not fit neatly into enshittification's consumer-facing timeline of decline. The platforms had not "turned against" them after an initial period of serving them well. They were constitutive of the platform's operation from the start, absorbed into its labour structure precisely because Kenyan law offered weaker protections and Kenyan wages were lower. The "good" platforms for users in the Global North was built, in part, on conditions that were never good for workers in the Global South.
In The Palestine Laboratory, the journalist Antony Loewenstein documents how the Palestinian territories have functioned for decades as a testing ground for surveillance and control technology, which is then exported to governments worldwide. The Pegasus spyware developed by the NSO Group, the facial recognition systems deployed across Hebron, the AI targeting tools used in Gaza: none of these began as consumer products that later turned exploitative. They were designed from the outset as tools of occupation and population control, refined through use on Palestinians, and sold as battle-tested to authoritarian regimes.
This is not the same technological domain as consumer platforms, but it reveals a parallel point: for many users, digital systems did not begin in service and then decay into domination; domination was the design premise. This directly raises important questions about the enshittification timeline. The technology, more often than not, arrived already serving the interests of those who commissioned it. And crucially, the same dynamics that make the Global South available as a testing ground (weaker legal protections, cheaper labour, populations with less political power to push back) are what make the enshittification story possible in the first place.
This is not an argument to reject the enshittification analysis. It is an argument to extend it. A decolonial critique of technology is not simply "the internet was always bad." It is rather: the conditions that made the internet harmful to specific communities were never peripheral to its design; they're an integral part of it. And any politics that aims to restore something like the pre-enshittification internet without reckoning with those conditions is doomed to reproduce them.
What Engaging with the "Master's Tools" Actually Looks Like
The decolonial tradition in technology scholarship has been doing this work for many years, often cited but rarely absorbed in mainstream tech discourse.
In Race After Technology: Abolitionist Tools for the New Jim Code, Ruha Benjamin develops the concept of the "New Jim Code," her term for technologies that reproduce existing inequities while presenting themselves as more objective and progressive than the discriminatory systems they replace. Benjamin's argument is a methodological one about how racism operates through algorithms. She writes, "racism thus becomes doubled, magnified and buried under layers of digital denial." One cannot identify or begin to address that if their analysis begins from the assumption that the technical layer is neutral.
In Algorithms of Oppression: How Search Engines Reinforce Racism, Safiya Noble makes a related case about search. Her argument is against the fiction of their neutrality. "Google creates advertising algorithms, not information algorithms", she writes, a distinction that matters enormously when those algorithms are treated as a public information resource by schools, journalists, and governments. The criteria of relevance built into search are not neutral, and treating them as such reproduces the structures of erasure they encode.
In her 2020 paper "Algorithmic Colonization of Africa", cognitive scientist Abeba Birhane argues that while traditional colonialism used brute force, algorithmic colonialism operates through corporate agendas: "state-of-the-art algorithms" and "AI-driven solutions" to social problems that reduce complex cultural and political realities to things that can be measured, quantified, and fixed. This can be clearly seen in the Nairobi case: the content moderators' psychological crisis was not incidental to the platform's AI safety work, it was the mechanism by which that work was accomplished.
Nick Couldry and Ulises Mejias take long historical view in Data Grab: The New Colonialism of Big Tech and How to Fight Back. Their claim is that data colonialism is not metaphor: it is structurally continuous with historical colonialism, with data replacing land as the resource being seized. "Like historical colonialism," they write, "today's tech corporations have engineered an extractive form of doing business that builds a new social and economic order." This is perhaps the most direct bridge that extends the enshittification critique into the decolonial one. Doctorow asks how platforms extract value; Couldry and Mejias ask what kind of thing that extraction is in historical terms.
I believe that Benjamin, Noble, Birhane, Couldry, and Mejias are engaging, in Lorde's terms, in the epistemic tools: not only the legal mechanisms or the technical architectures, but the analytical frameworks that tell us what counts as a problem, what counts as evidence, and whose testimony counts as knowledge. And none of them are arguing for inaction. Even Benjamin's subtitle, Abolitionist Tools, directly says that analytical tools can and should be turned toward liberation. But those tools have to be built with different assumptions than the ones baked into the systems being analysed.
What links these works is not a common slogan but a common method: they refuse to treat technical systems as neutral containers and insist that the categories through which problems are defined already encode relations of power. In that sense, they extend Lorde's challenge from feminist theory to technology critique.
This is precisely what gets lost when the phrase "the master's tools" becomes a debate about whether antitrust law is effective. The question is not whether antitrust law can break up a monopoly. The question, the one the speech seems to me to be asking, is whether breaking up a monopoly by itself constitutes the "genuine change" Lorde invoked, or whether it simply rearranges the furniture in a house whose foundations remain unchanged.
The "Master's Tools" Is an Invitation, Not a Rejection
There is something particular that happens when figures from Black feminist, queer, or decolonial traditions are cited in tech discourse. The citation tends to operate as credentialing rather than engagement, a way of demonstrating political seriousness before proceeding to make the argument you were going to make anyway. In that passage, Lorde is simultaneously "far smarter than I am about nearly everything" and also wrong about the thing that matters for the argument at hand. The invocation and the dismissal occur in the same breath.
This is, structurally, a version of what Lorde was diagnosing in 1979. The conference organizers had invited her. They had put her on the programme. They had formally acknowledged that her perspective had a place in the conversation. But they had put her in the one slot reserved for people like her, assigned her to respond to work that had not engaged with her tradition, and positioned her contributions as supplementary to a theoretical apparatus that remained unchanged by her presence.
The parallel is not exact. But the structure (incorporate, cite, proceed) is recognizable. And recognizing it is not a reason by itself to reject or abandon the enshittification analysis. It is a reason to take Lorde's actual argument seriously enough to let it do something to the analysis, rather than neutralizing it with a compliment.
I would hate for this critique to be read as "enshittification gets it wrong." I believe that enshittification is doing something real and important, and the feminist and decolonial traditions have tools that could make it more powerful, more accurate, and more honest about whose experiences it centers. To reject it, rather than invite it to engage, would in my view be a betrayal of what Lorde stood for.
In the speech, Lorde tells us that the first patriarchal lesson is "divide and conquer," and that the task is to "define and empower" instead. It places definition (understanding the actual structure of the problem including the parts that are uncomfortable or that implicate people within the movement) as prior to empowering. And defining requires the kind of analysis that comes from the marginalised: not as a supplement to the real analysis, but as a correction to its blind spots.
The question is not whether to use antitrust law or interoperability standards to fight platform monopolies. Use them to their fullest extent. The question is whether we can also build the analytical framework that asks why the platforms were built the way they were, who paid the costs that made them profitable, and what "better" actually looks like when we account for the full range of people who will live in whatever comes next.
What Lorde left us with, is not the answers to those questions but something more valuable: the insistence that they have to be asked, and that asking them requires centering the people for whom the stakes are highest. The "master's tools" speech, in its context and practice, is the antithesis of a thought-terminating cliché. Let's stop treating it that way.