The trap of breaking the mould

Posted on: 16 May 2026

When Marcel Duchamp bought a porcelain urinal from the J. L. Mott Iron Works showroom in New York in April 1917, rotated it ninety degrees on its side, signed it R. Mutt and submitted it to the inaugural exhibition of the Society of Independent Artists, he was not making art. He was doing something far more consequential, demonstrating that the category we call art was a convention defensible only through acts of exclusion. The Society had publicly declared that there would be no jury, that any work submitted with the six-dollar fee would be exhibited. Duchamp, who sat on the board of the organisation himself, knew precisely what he was doing. He wanted to force the system to reveal itself.

It worked. The board, after lengthy debate, hid the urinal behind a partition for the duration of the show, Duchamp resigned from the board in protest, and Alfred Stieglitz photographed the piece for the magazine The Blind Man. The original was lost, most likely thrown out as rubbish. A century later, in 2004, a survey of five hundred British artists, curators, critics and dealers would name it the most influential work of art of the twentieth century.

The interesting observation is not that Duchamp broke the category of art. It is that he involuntarily created another category, one within which every subsequent artist has had to position themselves. Warhol and his Campbell's soup cans, Beuys and his animal fat, Cattelan and his duct-taped banana, Banksy and his self-shredding canvas, Hirst and his formaldehyde shark. None of these artists operates by ignoring Duchamp. Each operates inside the Duchamp-category, or against it, but in both cases they presuppose it. Even those who returned to figurative painting did so as a choice made despite Duchamp, never as the natural continuation of a tradition Duchamp had simply leapt over.

This is the mechanism that interests me. Breaking a category does not free the space behind it. It captures that space. It creates a new category, involuntarily, and that category binds all successors. The only difference between the old constraint and the new one is that the old constraint was visible as tradition, whereas the new one disguises itself as liberation.

There is a certain cruel elegance in this mechanism. The mould-breaker believes he has opened a field. In fact he has only moved the boundaries, and those new boundaries are often narrower than the previous ones, because they operate invisibly, because they present themselves as natural acquisition rather than as convention. A category that does not know it is a category is more powerful than one that does.

Consider Margaret Thatcher, the clearest political case of this mechanism in twentieth-century Europe. Her masterpiece was not the privatisation of British Telecom, nor the confrontation with the miners, nor her transformative relationship with Reagan. Her masterpiece was Tony Blair.

When asked what her most important legacy had been, Thatcher famously replied: "New Labour". This was not a quip, it was the most precise diagnosis available of what had happened to the British Labour Party between 1979 and 1997. Blair had won the election on the slogan "Time for a Change", and that change consisted, in substance, of accepting the Thatcherite TINA, "There Is No Alternative", as the shared substrate of British political debate. The theatrical abandonment of Clause IV of the Labour constitution, the party's formal commitment to socialism, was the public signal that the class war was over, and over because one side had won.

Stuart Hall, who had analysed Thatcherism in real time in his 1979 essay "The Great Moving Right Show" before Thatcher even won her first election, saw the trap clearly. In a late interview he observed that what the cultural left had recommended was a project on the left of the same breadth and depth as Thatcherism. New Labour, he said with the dry exasperation of someone who had watched a misunderstanding play out for two decades, understood it as meaning that the same project was needed. Hall's diagnosis is decisive precisely because it comes from the left, from a Jamaican-British Marxist who had no Tory sympathies and no interest in apologising for Thatcher. He simply observed that the category had captured those who thought they were opposing it.

Thatcher had not merely defeated the British left. She had created the category within which the British left would have to reconstruct itself. Blair was not Thatcherite in the ideological sense, he negotiated peace in Northern Ireland in ways Thatcher would never have countenanced, he equalised the age of consent that Thatcher had criminalised through Section 28, he massively expanded the welfare state through investment in health and education. Yet he operated entirely inside the Thatcherite economic framework: financial deregulation, labour market flexibilisation, the primacy of market mechanisms, privatisation of public services as method, to name only a few. Pat McFadden's synthesis is precise, Blair did not press the rewind button on the previous eighteen years, but neither was he a continuation of Thatcher. He was something else. He was what had become possible after Thatcher.

The mechanism then ran forward through the Conservatives as well, in case anyone is tempted to read this analysis as one-sided. Cameron and May governed inside the Blairite operational consensus that had itself emerged from inside the Thatcherite economic consensus. Nested categories, each successor convinced of having broken with the previous one, each in fact occupying a position made available by their predecessor. The mechanism does not care which party benefits, it only cares that the perimeter holds.

And here the present week renders the point dramatically visible. As I write, Keir Starmer is facing the worst crisis of his premiership. Ninety-two Labour MPs are calling for his resignation, four ministers have resigned from his government in the last forty-eight hours. Wes Streeting, his Health Secretary, is preparing to challenge him for the leadership. The most unpopular Prime Minister in British polling history, with a net favourability rating of minus fifty-seven, matched only by Liz Truss.

The conventional diagnosis is that Starmer has failed tactically, has mishandled communications, has lost touch with the traditional Labour electorate, all true and all marginal. The structural problem is different. Starmer is losing because he operates inside the Thatcher-category without the awareness Blair had of being inside it. Blair knew he stood within the Thatcherite perimeter and made strategic virtue of that knowledge, he inhabited it like a master. Starmer inhabits it through inertia, without the same lucidity about the perimeter he occupies, and when that perimeter starts to crack under pressure from Reform and from an electorate that wants something different, he finds himself without the instruments to renegotiate it. It is the price of not having understood which category contains you. Even Brexit, which presented itself as rebellion against the Thatcher-Blair establishment, ended up reproducing the fundamental economic structures in a different form. The mould-breaker is still there, beneath the surface, binding the successors, even when the successors no longer know where they stand.

What I would call involuntary social design is precisely this: every act of exit from a category, if sufficiently powerful to produce rupture, leaves no neutral space behind. It leaves a new category, immediately populated by those who come after. The mould-breaker believes he has liberated successors, when in fact he has bound them to a new perimeter, often without either party realising it. The freedom of the mould-breaker is always relative to the system he has broken. The servitude of successors is absolute relative to the system the mould-breaker has created.

Is there a way out of this mechanism? I would say no, and this is the most interesting part. Trying to exit the Duchamp-category by making traditional art means positioning oneself in relation to Duchamp. Trying to exit the Thatcher-category by practising radical social democratic politics means positioning oneself in relation to Thatcher. The mould-breaker captures even those who reject him, because he imposes the terms of the rejection. This is a theorem, not a contingency.

What remains, then, is a certain epistemic modesty. When someone today presents themselves as the figure who breaks free of stereotypes, who refuses to be pigeonholed, who thinks beyond affiliations, it is worth asking which new category they are involuntarily founding. It is worth asking who, ten or twenty years from now, will have to operate inside the perimeter that is presented today as liberation. Because that category will form, whether one wills it or not. It is in the very structure of the breaking gesture.

The only way not to create categories is to break nothing. But then one is not the mould-breaker, one is simply among those who operate inside the prevailing perimeter, and all things considered that is a more honest position. Knowing oneself to be inside a category, whatever that category may be, is the first step in not deluding oneself into thinking one is outside it. Duchamp knew, I believe. Thatcher, I suspect, also knew. Starmer does not, and that is precisely his problem.

The most interesting mould-breakers have always been those aware that their gesture was founding a new perimeter. The others, those who genuinely believed they had opened a neutral space, produced the fiercest constraints, precisely because they remained invisible.