On one level, this is not an unreasonable conclusion. A party that repeatedly sends candidates into public life only to discover, days later, that some of them have apparently spent years treating social media like a pub toilet wall is not a party that has fully mastered candidate vetting. A party that promises an assault on waste, bureaucracy and taxes, then finds itself raising council tax, cutting ordinary services or arguing about infrastructure finance, is not exactly staging a seminar in administrative seriousness. A party that takes control of local authorities and then starts losing councillors to resignations, suspensions, expulsions and internal warfare has, at the very least, a personnel problem.

But I think the incompetence explanation is incomplete.

The problem with Reform is not merely that it is bad at responsibility. The problem is that its politics depend on evading responsibility.

That distinction matters because it explains both Reform’s appeal and its vulnerability. Reform is very good at politics when politics means naming grievances, directing anger, finding enemies and promising the simple moral pleasure of finally saying what supposedly cannot be said. It is much worse at politics when politics means budgets, trade-offs, legal duties, procurement, adult social care, children’s services, scrutiny committees, risk registers and the melancholy discovery that a council cannot, in fact, pay for much-needed road repairs with the proceeds of cancelling a diversity training session.

This is not an argument that the grievances are invented. They are not. The country is full of places where people - rightly - feel that public services have deteriorated, that politics has become unresponsive, that immigration has changed communities faster than institutions have adapted, and that the old parties have offered managerial disappointment in slightly different colours. It is not hard to see why Reform’s story appeals. It tells people that the problem is not complexity, scarcity or institutional decay, but betrayal. Someone did this to you.

The trouble is that grievance is not a programme for government.

Reform’s national politics works because it converts dissatisfaction into momentum. Every failure can be folded back into the same story. If public services are struggling, it is because the old parties wasted the money. If councils are broke, it is because of bureaucrats, consultants, net zero, equality officers, asylum hotels, work-from-home culture, or whatever target happens to be nearest to hand. If a promise proves impossible, it was because the establishment hid the true scale of the problem. If a candidate collapses under scrutiny, the media are out to get them. The emotional structure is wonderfully convenient: success proves the people are rising; failure proves the establishment is afraid.

Local government ruins this game because it supplies receipts.

Reform’s breakthrough in the 2025 local elections was enormous. The party gained hundreds of council seats, won the Runcorn and Helsby by-election (albeit by six votes), took mayoralties, and won pluralities or majorities of councillors in Durham, Kent, Lincolnshire, Lancashire, Staffordshire, Nottinghamshire, Derbyshire, North Northamptonshire, Doncaster and West Northamptonshire. In 2026, the pressure continued in Labour’s northern heartlands, with Labour losing control of Barnsley for the first time in the council’s 52-year history.

For a national insurgent party, these are trophies. But for a party supposedly preparing for government, they are tests.

That is the difficulty. Reform has spent its entire existence making the claim that other people are useless. Then voters, reasonably enough, began giving them the opportunity to be useful. The party’s old posture was anti-establishment. Its new problem is responsibility.

Take council tax. This is where an anti-waste party should, in theory, have its easiest win. Reform’s pitch depends heavily on the idea that the state is bloated, captured and frivolous; therefore, a determined outsider should be able to find savings where the old parties found excuses. Yet Reform’s record is awkward for the easy version of this claim. The party made pledges at various levels to cut or freeze council tax, on this basis, with local election leaflets saying the party would “reduce waste and cut your taxes”.

Reform have not managed to do this anywhere. In many places, they’ve fallen into the kind of political doublespeak they rightly criticise the ‘uniparty’ for - my favourite example, from Kent, is the line “cutting taxes could mean not putting them up as much”. Kent then increased council tax by 3.99%. Elsewhere, the pattern has been even starker. In January 2026, the Guardian reported that Derbyshire, North Northamptonshire, West Northamptonshire and Leicestershire had proposed 5% rises, the maximum normally permitted. In Worcestershire, they received special dispensation to go above the usual cap on rises and upped it by nearly 9%.

Now, to be fair (a terrible habit, but sometimes necessary), this is not only a Reform problem. Local government is under extraordinary pressure. Councils with social care responsibilities face a 5% referendum threshold, including 2% for adult social care and 3% for general spending. The Local Government Association has warned that the finance settlement puts pressure on councils to raise council tax up to referendum thresholds to fund demand-led services such as adult social care. The Institute for Government has shown the deeper structural shift: between 2009/10 and 2024/25, per-person spending on adult and children’s social care rose in real terms, while spending on other services fell by more than a third; social care rose from just over half of local authority spending to more than two-thirds.

This is the boring-but-true answer, and it matters. Councils are not sitting on vast secret vaults of spare money waiting to be unlocked by someone sufficiently annoyed with HR departments. Everything that can be cut, has already been slashed during the years of Conservative austerity. And much of what is left is simply not optional. Adult social care, children’s services, homelessness, SEND transport and safeguarding obligations do not disappear because a councillor has found a pleasingly punchy phrase about waste.

But that’s precisely the point. Reform’s promise is not just that it would be a slightly better manager of scarcity. Its promise is that the old constraints are fake: that the system is rotten, the waste is obvious, the priorities are mad, and only cowardice or corruption prevents someone from sorting it out. Then it gets into office and starts sounding, very quickly, like every other council leadership that has ever had to produce a legal budget.

Worcestershire is the clearest example because it compresses the whole problem into one local authority. Reform took minority control of a council Farage later described as virtually bankrupt, adding that he wished the party “hadn’t bothered”. That is an extraordinary confession, not because Worcestershire’s finances were easy, but because the complaint amounts to this: governing would have been more convenient if the inherited problems had been less real. Reform’s tenure was then marked by emergency funding, a council tax rise of 8.98%, internal battles, the replacement and suspension of its former leader, and ultimately the loss of minority control after Conservatives, Greens, Liberal Democrats and independents formed a short-lived alliance.

What is revealing about that story is not simply chaos. It is the implied expectation that opposition politics could continue after power had been taken. But councils do not exist to supply press release material for Westminster. They exist to fix roads, fund care packages, maintain libraries, run children’s services, process planning applications, manage SEND transport, collect bins and keep the whole civic machine moving even when nobody is paying attention. Local government is unglamorous because it is real. It is where politics stops being a pose and becomes a spreadsheet.

The same thing appears in Doncaster, where Reform’s position on reopening the Doncaster-Sheffield Airport has moved through a strange cycle of support and opposition, confrontation and retreat. Reform councillors have certainly argued that they wanted scrutiny and better terms. That is not inherently unreasonable. A £57 million loan is not pocket change, and major regeneration schemes should be interrogated. But the politics around it has been deeply revealing. Reform councillors submitted a motion criticising the Labour mayor’s airport pledge and stressing their own democratic responsibility, after that mayor’s narrow victory in 2025, which was attributed in no small part to Labour being the only party with a plan to reopen the airport. Later, Reform members sought to revisit or rescind the previously approved loan, before an amended position allowed the loan to continue after negotiations.

Again, the point is not that every concern they raised was automatically illegitimate. The point is that attacking the symbol of another party’s campaign is easier than carrying responsibility for the mechanism that might deliver a major investment. Reopening an airport is an excellent local political goal because it condenses jobs, pride, regional neglect and economic hope into one image - and thus making sure it fails would show the public that their opposition can’t deliver. But the actual ‘getting stuff done’ work of leases, loans, commercial operators, regulatory hurdles, passenger forecasts, public risk and timelines is also extremely popular, and trying to shut it down showed Reform the limits of their own strategy; they caved and backed it once they realised they were hurting their own position.

Then there is candidate vetting. Here, Reform’s defenders sometimes say that rapid growth brings growing pains, and that every party has had bad candidates. This is true as far as it goes. The Greens, Labour, Conservatives and Liberal Democrats have all had candidates suspended, exposed or embarrassed. No party with hundreds or thousands of local candidates can guarantee moral perfection.

But Reform’s problem is not an isolated embarrassing Facebook post. It is a recurring inability to filter out candidates whose public records appear to contain racist, antisemitic, Islamophobic or otherwise extremist material. Mark Pack notes 73 Reform UK resignations or kickings-out between May 2025 and May 2026, with a further 17 (and counting) since May 2026. Add that to their two lost MPs since the General Election, and criticism of both Farage and Tice over their financial dealings.

That is not just a reputational issue. A political party asking to govern cannot treat candidate selection as a content acquisition strategy. Councillors are not disposable avatars for national messaging. They are their own people. They sit on planning committees. They handle casework. They make decisions affecting schools, safeguarding, highways, care, housing and community cohesion. In a political system already suffering from immense distrust, a party that floods public office with unfit or poorly-vetted candidates is not challenging the establishment. It is lowering the minimum acceptable standard for entering it.

The deeper problem is that Reform’s politics rewards irresponsibility long before it reaches office. It rewards the person who says the difficult thing is easy, the complex thing is a conspiracy, the constrained choice is a betrayal, and the boring institutional answer is cowardice. This is powerful because it feels like clarity, but clarity is not the same as truth. Sometimes the simple answer is simple because it has left out the parts of the problem that make the problem hard.

That is also why local government is so dangerous for them. It deprives them of the luxury of purity. It asks them not what they oppose, but what they would fund. It asks not what they would abolish, but what they would put in its place. It asks not whether they dislike waste, but whether the promised waste actually exists in sufficient quantity to pay for their promises. It asks not whether they can identify public anger, but whether they can turn that anger into functioning administration.

And when they cannot, they increasingly risk portraying themselves as what they most despise: another party promising change and delivering excuses.

There is a political lesson here for Reform’s opponents too. Simply calling Reform chaotic will not be enough, because some voters already believe the whole system is chaotic. Being ‘adults in the room’ counts for very little when a plurality of most voters in any given constituency want to burn the house down. Simply listing offensive candidates will not be enough, because some voters have become grimly accustomed to scandal as political weather. Simply saying their sums do not add up will not be enough, because many people already think expert arithmetic has been used for years to justify decline.

The stronger argument is about responsibility.

Reform should be judged not only by whether it reflects grievance, but by whether it can carry the burdens that come with power. Can it select candidates fit for public office? Can it make lawful budgets without pretending every rise is someone else’s fault? Can it scrutinise major projects without turning them into performative hostage situations? Can it govern places it wins, even when those places are financially strained, demographically complicated and institutionally fragile? Can it tell voters the truth before an election, rather than discovering the truth immediately afterwards?

If the answer is no, then Reform is not the antidote to broken politics. It is broken politics with a harder edge and a louder microphone.

The party’s great asset has always been its ability to stand outside responsibility and profit from everything that goes wrong inside it. Local government changes that. A pothole does not care whether your rhetoric is insurgent. A care package cannot be funded by vibes. A council tax bill is not impressed by the phrase ‘more of the same’. The moment Reform wins power, its central trick becomes harder to sustain, because every grievance it harvests becomes, in some measure, a problem it now owns.

That is why the title matters. Reform is not simply anti-establishment. Plenty of establishments deserve criticism, and many institutions have earned the public’s suspicion. The more precise charge is that Reform is anti-responsibility: fluent in anger, evasive in power, and increasingly exposed wherever voters have made the understandable mistake of asking it to govern.

There is a very odd little civil war happening in Labour circles at the moment.

Over the last few weeks, I’ve noticed more and more posts and comments in Labour forums and political discussion boards whose main purpose seems not to be defending the Labour Government, or arguing for a particular political direction, or even making the case for Keir Starmer. They exist, instead, to have a go at Andy Burnham.

Sometimes this is dressed up as strategic seriousness. Sometimes it is framed as irritation with the Westminster psychodrama. Often it is just a good news story about the Government posted alongside some little non sequitur comment about his opponents, like “Andy Burnham isn’t fit to lick Keir’s boots”, or a dark muttering about treachery, ambition, or betrayal - all the other words people reach for when politics becomes less about judgement than court etiquette.

I find this baffling.

Not because Starmer has no case to make. He absolutely does. There are serious Labour achievements to defend: the Employment Rights Act, the increase in the National Living Wage, Great British Energy, the Renters’ Rights Act, the scrapping of the Rwanda scheme, and the return of passenger rail services to public ownership. Those are not trivial things. They are the kind of things Labour members should be willing to defend, and the kind of things the Government should be much better at explaining.

The problem is not that Starmer has done nothing. The problem is that politics is not a university course in which policy delivery is marked against a neat little assessment rubric. Voters are allowed to dislike a Prime Minister who has passed good legislation. They are allowed to feel that the country still does not work. They are allowed to conclude, fairly or unfairly, that the Government does not speak to their lives. And at the moment, the available evidence is grim: Labour has taken a severe beating in every poll since 2024, Reform is miles ahead in the national polls, and Starmer’s personal ratings remain at Liz Truss levels of dire.

That is frustrating. It’s unfair. But it is not irrelevant.

This is the lesson Labour members should have learned from the Corbyn years, if nothing else. Personal loyalty to a leader is not the same thing as political seriousness. Nor is it evidence of moral fibre to keep insisting that the public must be wrong until the public, with weary impatience, removes the opportunity to insist on it any longer.

There is also a more immediate problem. Andy Burnham is not a hostile columnist, a freelance leadership plot, or a spectral threat haunting the fevered imaginations of Starmer’s supporters. He is Labour’s candidate in a high-profile by-election.

The party has already made its choice. It could have tried to block him again. It did not. The NEC had previously blocked Burnham from standing in Gorton and Denton; in Makerfield, it approved his route to selection, despite the same obvious issue that if he wins he is immediately disqualified from the Greater Manchester mayoralty, triggering a mayoral by-election.

That tells us something. Perhaps Starmer wants this contest? Perhaps he thinks it is better to have Burnham inside the tent than outside it. Perhaps he believes the party has no credible way of keeping him out without looking absurd. Perhaps he thinks a win in Makerfield shores up Labour, while a loss would wound Burnham. Whatever the calculation, it has been made.

Which makes the internal attacks stranger still. If Burnham loses, that is not just a defeat for Andy Burnham. It is another Labour defeat. It is another Reform advance. It is another headline about a Government unable to hold together its old heartlands. It is another piece of evidence for everyone - Left and Right alike - already saying that Labour is finished.

And for what? The emotional satisfaction of proving, in public, that some Labour members dislike the Labour candidate?

Have a look at the kind of print going out in Makerfield right now. I’ve attached two examples - one from Labour, one from Reform.

Reform’s letter is from Nigel Farage. It attacks the government, attacks Burnham, talks about boats, net zero, the ECHR, tax, crime and Westminster ambition. It claims that Makerfield needs a local person who will put the area first - but doesn’t even bother to name their candidate!

This is an extraordinary thing, really. A parliamentary by-election in which the would-be MP is treated almost as an administrative detail. The brand is Farage. The enemy is Labour. The target is Burnham. The candidate is somewhere behind the curtain, presumably waiting to be properly introduced once the vibes have done their work.

Labour’s letter, by contrast, is from the candidate. It talks about Makerfield. It talks about Burnham’s local connections. It talks about his record in Greater Manchester. It talks about his interests and plans; energy bills, rail fares, council housing, reindustrialising the North, and technical routes into good jobs.

It does not mention Reform. It does not mention Farage. It does not spend its limited space calling the other side names. This is a small thing, perhaps, but a revealing one. Reform accuses Labour of making the contest about attacks and Westminster ambition, while sending a leaflet from its party leader attacking Labour and Andy Burnham. Labour sends a leaflet from its candidate, simply making the case for what he has done and what he wants to do.

Set aside, just for a moment, the looming leadership question. Starmer clearly did when he chose not to push the NEC to repeat the Gorton and Denton decision. The immediate question is much simpler.

Labour members can have the argument about Starmer and Burnham after polling day. They almost certainly will. They may have it with spreadsheets, WhatsApp briefings, anonymous quotes, soft launches, hard launches, triangulated outriders and all the other grim ceremonial objects of internal Labour politics.

But right now, Andy Burnham is the Labour candidate in Makerfield. Reform is trying to beat him. And every Labour supporter who spends this by-election trying to undermine him is not defending Starmer. They are helping Nigel Farage make Labour look divided, bitter and unserious.

That is not loyalty. It is factional self-harm with a red rosette pinned to it.

It can be done in several registers.

• There is the moral register, which reaches almost automatically for Iraq.

• There is the generational register, in which Blair is treated as a man still trying to govern the country from the politics of 1997, like a ghost haunting the Downing Street sofa.

• There is the ideological register, in which every mention of the ‘radical centre’ is presumed to be a coded demand for privatisation, triangulation and a nice lunch with someone from a management consultancy.

• And there is the simpler, more tribal version: he is Tony, therefore he is wrong.

None of these responses is entirely unfair. That is what makes the problem awkward.

Blair’s recent essay on Labour’s direction contains plenty that any credible Labour person should reject. A Labour government should not abandon new net zero projects as though the climate crisis can be postponed until the bond markets are in a better mood. It should not treat laws for workers’ rights as a dispensable luxury. It should not ignore the need for a higher minimum wage, nor retreat from changes to non-dom tax status because the very wealthy have once again discovered their deep emotional vulnerability to being taxed. Nor should a Labour Prime Minister treat alliance management as a moral blank cheque for Donald Trump’s military choices in Iran.

If that were all Blair had said, the response could be short, satisfying and almost entirely useless.

The problem is that Blair has also said something true. Labour is in danger of confusing office with purpose. It won the 2024 election not as the author of a fully believed national project, but as the acceptable alternative to Conservative collapse. That was enough to win. It was even enough, for a time, to govern. But it was never going to be enough in the long run.

This is the uncomfortable part. Blair’s answers are often too thin, too technocratic, too impatient with the moral commitments that make Labour worth having at all. But his question is the right one. What is Labour for?

Not what is Labour against. We know that already. Labour is against Conservative Party’s austere selfishness, Reform’s authoritarian racist pantomime, the Green party’s magical thinking, The Liberal Democrats’ opportunism, and the general political habit of setting the house on fire then congratulating ourselves for having produced warmth.

The harder question is what Labour is trying to build.

The official answer still exists. Labour has its five missions. The government’s ‘Plan for Change’ still talks about milestones for mission-led government, built around economic growth, the NHS, safer streets, opportunity and clean energy. The language remains formally intact. There is still an architecture, a set of headings, a structure under which announcements can be filed.

But a filing system is not a governing story.

That distinction matters. A mission is only politically useful if it disciplines choices. It has to tell ministers what matters when two good things conflict. It has to tell the Treasury what cannot be sacrificed. It has to tell the public why pain is being asked of them, what trade-offs are being made, and what sort of country should exist at the other end of the work. Without that, a mission becomes a brand category. It gives announcements somewhere to live, but not necessarily a reason to exist.

This is where Ian Dunt’s essay, ‘How to build a good Prime Minister’, gets closer to the problem than Blair does. Dunt’s argument is not merely that Starmer needs better slogans, though God knows a few would not hurt. It is that successful government requires a relationship between vision, policy and delivery. You begin with a sense of direction. You translate it into practical measures. Then you build the machinery capable of making those measures real. Dunt’s criticism of recent prime ministers is savage, but the analytical point beneath the savagery is quite simple: Britain is not ungovernable, just badly governed.

This is, in one sense, an old Blairite lesson. The first New Labour government did not simply say ‘Education, education, education’ because somebody in the comms team had discovered repetition. It said it because the phrase did actual political work. It told voters what mattered. It told Whitehall what mattered. It gave ministers a test against which decisions could be judged. It allowed delivery mechanisms to form around something more substantial than the Prime Minister’s inbox.

This is also why the comparison is so uncomfortable for Labour now. Starmer’s government has done real things. Its defenders are not wrong to point to serious achievements. But the public impression remains one of drift, because the actions do not yet add up to a project. Too often the argument seems to shrink at the moment it should expand. We were promised national renewal. We are offered administrative reassurance. We were promised transformation. We are told that the grown-ups are in the room.

The grown-ups do need to be in the room. This is an underrated point in politics, mostly because people who sneer at competence have usually not had to deal with the consequences of incompetence. The country suffered badly under recent governments that seemed to treat seriousness as an optional extra. There is nothing trivial about restoring some basic capacity to the state.

But competence is not a political project. It is the precondition for one.

This, I think, is the trap Starmer’s Labour has fallen into. In opposition, competence was an electoral argument. It was necessary and, after Johnson, Truss and Sunak, rather more emotionally resonant than people sometimes admit. Labour did not have to persuade the country that it was thrilling. It had to persuade the country that it was sane. The campaign could contrast steadiness with chaos, adulthood with theatrical irresponsibility, the modest promise of repair with the exhausted absurdity of Conservative rule.

That was enough to win the election. It is not enough to govern through a period of national insecurity.

People do not merely want ministers who can operate the machine. They want to know what the machine is for. They want to know why their lives are still hard. They want to know why a Labour government, elected after fourteen years of Conservative decline, so often sounds as though the limits of politics have already been reached. They want to know whether the future being built is one they will have a place in.

Blair sees part of this. His essay is strongest when it insists that Labour cannot retreat into self-delusion, nostalgia or factional comfort. He is right that the country faces a world being remade by artificial intelligence, geopolitics, energy insecurity, demographic pressure and a changed global economy. He is right that government has to be serious about technology, growth, skills, state capacity and Britain’s place in the world. He is right, too, that politics cannot simply be the art of giving one’s own side comforting things to say to itself.

But Blair’s account is incomplete in precisely the way Blairism is often incomplete. It sees change more clearly than it sees security. It sees adaptation more clearly than it sees equality. It sees the need for Britain to face forwards, but not always the equally important need for people to believe that the future is being made with them and for them, rather than done to them by clever people who will not live with the consequences.

That is where Tom Watson’s response is useful. Watson’s argument, as I read it, is not that Blair is wrong to demand adaptation. It is that adaptation without fairness is not enough. The radical centre cannot be only a doctrine of reform, technology and growth. It has to be a doctrine of equality, security and democratic consent. Otherwise it becomes modernisation with the moral content stripped out.

This is the distinction the next Labour argument needs. The choice is not between Blairite modernisation and left-wing comfort politics. That is too neat, and much too convenient for everyone involved. The real question is whether Labour can produce a politics that is modernising and egalitarian at the same time.

New Labour, at its best, was able to do that because Blair and Brown, in their different and often mutually poisonous ways, held together two necessary truths. Blair understood change, national adaptability, the need to make Britain face forwards. Brown understood inequality, poverty, the role of the state in altering life chances, and the moral seriousness of redistribution. This is a simplification, obviously. All useful political memories are simplifications. But it captures something real.

The last successful Labour project was not simply a cult of managerial competence. Nor was it simply a programme of redistribution. It combined modernity with investment, reform with social mobility, constitutional change with public service improvement, and national confidence with a moral concern for those excluded from opportunity. It did not always do this well, and when it did, it often did it with an unbearable degree of self-congratulation. But it had a story.

The country Blair governed no longer exists. That is the point his admirers sometimes miss. The economy is weaker. Housing is more punishing. Work is less secure. Trust is lower. Public services are more strained. The promise that globalisation and technological change would lift all boats now sounds, to many people, less like optimism than a cruel joke told by someone who owns a boat.

So yes, Labour needs growth. Of course it does. Without growth, social democracy becomes an argument about how to distribute disappointment. But growth by itself is not a moral purpose. People have lived through too many versions of national success that did not quite reach them. They have heard too many clever people explain why the graph is improving while their rent rises, their bus disappears, their GP surgery becomes unreachable, and their children cannot imagine buying a home.

This is why AI cannot be treated only as a productivity question. Blair is right that it will reshape government and the economy. He is right that a state which cannot use modern technology will become slower, weaker and more expensive. But a Labour account of AI has to ask different questions too. Who owns the gains? Who bears the risks? What happens to young people if entry-level work is automated away before they have acquired the experience that makes them employable? What happens to training if every firm praises it in public and avoids funding it in private? What happens to dignity when work becomes more monitored, more precarious, more easily replaceable?

The same is true of welfare reform, public service reform, policing technology and the green transition. The question is not simply whether change is necessary. Often it is. The question is whether Labour can make change legitimate. That requires more than a plan. It requires a social contract.

This is also why the current debate cannot be reduced to Starmer’s personal limitations, though those limitations are real. Dunt’s sharper point is about the qualities required to govern: vision, delivery, courage, decision-making and curiosity. Those are not always the qualities British politics selects for. Parties often reward loyalty, message discipline, caution and the ability to survive internal systems without frightening anyone important. These are not useless qualities. But they are not the same as curiosity, strategic imagination or the capacity to make decisions under pressure.

Starmer’s particular tragedy may be that the traits which made him electorally plausible after 2019 are not the traits the moment now demands. His caution was once reassuring. His seriousness was once a relief. His refusal to emote theatrically was once a useful contrast with Johnsonian narcissism. But government changes the test. What looked like discipline in opposition can look like drift in office. What looked like restraint can become an inability to define the argument. What looked like seriousness can become a fear of politics itself.

There is a danger here, though. The Labour Party is very good at turning strategic problems into leadership psychodrama. It is perhaps the one national industry in which we remain world-leading. A party can always persuade itself that somewhere, just offstage, there is a leader who will resolve all tensions by force of personality. The right speech. The right accent. The right regional biography. The right soft-focus video. The right mixture of toughness and empathy, which usually means whatever the person speaking already wanted.

But the question is not simply whether Labour has the right Prime Minister. It is whether Labour has a governing project capable of surviving whoever the Prime Minister is.

That project cannot be a memory of New Labour. Nor can it be a retreat into a comforting left politics that treats public anger as proof of its own virtue. It has to begin from the world as it is: poorer than it should be, angrier than it used to be, more insecure than polite politics likes to admit, and increasingly vulnerable to parties which turn decline into resentment.

Labour’s answer cannot be to triangulate around Reform in slightly softer language. That is not realism. It is a category error. Reform’s strength does not come merely from its position on immigration, or from culture war theatre, or from Farage’s apparently inexhaustible ability to look like a man complaining about a hotel breakfast. It comes from the fact that too much of life in Britain feels broken, expensive, insecure and rigged. Mimicking Reform’s language does not answer that. It confirms their premise.

A serious Labour project would start somewhere else. It would say that Britain needs to adapt to the future, but that the future must be made fair. It would say that growth matters, but that growth must improve ordinary lives rather than merely decorate Treasury forecasts. It would say that technology matters, but that people are citizens, workers and neighbours before they are data points. It would say that public services need reform, but that reform must be done with the people who use and deliver them, not imposed from a sealed room in Whitehall. It would say that the state must become more capable, but also less remote. It would understand that equality is not only about income, but about power, security, place, dignity and childhood.

This would be neither nostalgic nor timid. In fact, it would require more courage than either Blairite adaptation without equality or left comfort politics without delivery. It would require Labour to say hard things to the public. Not everything can be fixed by taxing a handful of billionaires. Not every demand can be met at once. The state cannot repair fourteen years of damage instantly. Immigration is economically and socially necessary, but it has to be governed well and explained honestly. Net zero will require disruption, but the alternative is worse. Public services need more money, but money without reform will not be enough. Europe matters, and Britain will not prosper by pretending geography is optional.

That is what leadership is for. Not to recite public opinion back to itself, but to shape it. Not to avoid trade-offs, but to explain them. Not to make politics feel painless, but to make sacrifice feel purposeful.

The mistake Labour must avoid is assuming that because Blair’s prescription is flawed, his diagnosis can be ignored. This is a comforting error. It allows the party to preserve all its existing habits. Blair can be dismissed as out of touch. Starmer can be defended as merely unlucky. The missions can remain on the website. The machine can continue producing announcements. Everyone can agree that the alternative is worse, which it is, and then wonder why the public remains stubbornly ungrateful.

But the public is not obliged to be grateful for the absence of catastrophe. It is not enough to say that Labour is better than the Conservatives. It is not enough to say that Reform would be worse. Both things are true. Neither is a project.

This is why ‘competence’ becomes such a seductive but dangerous word. It sounds serious. It sounds adult. It sounds like the opposite of everything that went wrong before. But competence without direction becomes maintenance. And maintenance, however necessary, cannot meet a moment of national decline.

Labour needs something larger. It needs a story about work, power, technology, climate, security, care, place and democracy. It needs to connect the everyday frustrations of people’s lives to a plausible account of national repair. It needs delivery machinery, yes, but machinery pointed at something. It needs to stop treating values as decorative language attached to policy, and start treating them as the test by which policy is judged.

Blair is not the answer to Labour’s present crisis. The country has changed too much, and his own account leaves too much out. But he has put the right question back on the table. Dunt sharpens it into a question of governing capacity. Watson sharpens it into a question of equality and social contract. Taken together, they point towards the real problem.

Labour does not need to become Blairite again. It does need to remember that government requires purpose as well as process. It does not need to worship at the altar of the radical centre. It does need to become radical enough to face the future, and Labour enough to make that future fair.

The question is not whether Labour should listen to Tony Blair. It is whether Labour can answer him.

What is this government for?

Annotated bibliography

Tony Blair, ‘The Labour Party Is Playing With Fire Over Its Future and the Future of the Country’, Tony Blair Institute for Global Change. Used as the main irritant and provocation for the piece: Blair’s critique of Labour’s lack of coherent plan, his emphasis on the radical centre, technology, growth and state reform, and the points where his policy prescriptions become morally and politically contestable for Labour. (https://institute.global/insights/politics-and-governance/the-labour-party-is-playing-with-fire-over-its-future-and-the-future-of-the-country)

Labour Party, ‘Plan for Change’. Used to ground the claim that the five missions and formal language of mission-led government still exist; the argument is not that Labour has literally abandoned the missions, but that they are not presently functioning as a governing story. (https://labour.org.uk/plan-for-change/)

UK Government, ‘Plan for Change’. Used as the official government framing of the missions: growth, NHS, safer streets, opportunity and clean energy, set within a wider promise of national renewal. (https://www.gov.uk/missions)

Institute for Government, ‘Keir Starmer’s “Plan for Change” is necessary — and overdue’. Used to support the point that Labour’s mission language has shifted through several framings, and that consistency and staying power are essential if missions are to discipline government rather than merely organise announcements. (https://www.instituteforgovernment.org.uk/comment/keir-starmers-plan-change)

Ian Dunt, ‘How to build a good Prime Minister’. Used for the distinction between vision, policy and delivery, and for the wider argument that Britain is not ungovernable but badly governed.

Striking 13

How to build a good prime minister

Read more

17 days ago · 220 likes · 54 comments · Ian Dunt

Tom Watson, ‘Blair is right. But the radical centre needs equality‘. Used for the balancing argument: Blair is right about adaptation, but incomplete without equality, democratic consent and a new social contract.

Tom Watson's Newsletter

Blair is right. But the radical centre needs equality.

Tony Blair’s waspish note says something clear and uncomfortable. Labour is “playing with fire”. It won the 2024 election “not by acclaim”, but by being an “acceptable” default after Conservative failure. It has an “almost infinite capacity for self-delusion”. It lacks a “worked-out, coherent plan”. It has retreated into a “politics” bubble. It faces a …

Read more

12 days ago · 25 likes · 17 comments · Tom Watson

The Guardian, ‘Keir Starmer defends policy choices in rebuttal of Blair’s criticism’. Used for live political context around Starmer’s response to Blair, including the dispute over workers’ rights, net zero, and Blair’s criticism of Starmer’s position on Iran. (https://www.theguardian.com/politics/2026/may/28/keir-starmer-defends-policies-tony-blair-criticism-labour)

As Labour reels from the results of the 2026 local elections, Westminster and the national media have done what they do best: taken a complicated national picture and immediately turned it into a horse race between a handful of people who have already been profiled to death.

Angela Rayner. Wes Streeting. Ed Miliband. Andy Burnham, somehow, despite the awkward technicality of not being an MP. A familiar cast of plausible, semi-plausible and constitutionally inconvenient names has been arranged across the nation’s politics pages like the world’s least erotic swimwear calendar.

But this is old thinking. Tired thinking. Thinking of the sort that asks whether the next leader of the Labour Party might be someone with a national profile, an ideological project, a recognisable public image, a factional base, Cabinet experience, or any evident desire to do it.

That is precisely why the conversation must move on.

To be clear, this is not because the people listed below have done anything wrong. Quite the opposite. Their only offence is being sufficiently unknown to be available for projection. In the new politics, the decisive qualification for leadership is not experience, popularity or a coherent plan for change. It is not yet having annoyed enough people to be impossible.

By that measure, Labour’s true field of leadership contenders is not to be found among the so-called big beasts, but among the mysterious 2024 intake: that vast parliamentary hinterland of lanyards, maiden speeches, local newsletters, and the haunted eyes of 211 once-hopeful people who thought they had won a seat in a functioning government, not a timeshare in a bin fire.

Here, then, are the real runners and riders.

Chris Bloore: the Redditch Realist

In every leadership contest there comes a moment when the party yearns for someone who can cut through the noise. Someone grounded. Someone serious. Someone whose name has the lexicographical texture of a well-managed public consultation.

Step forward Chris Bloore, MP for Redditch.

Bloore’s great strength is that absolutely nobody in the Westminster bubble has yet formed a strong opinion about him, making him uniquely placed to unite the party. The left cannot denounce him, because it has not yet got around to maliciously editing his Wikipedia page. The right cannot brief against him, because the special advisers are still checking whether Redditch is a real place or a conceptual metaphor for Ordinary Working People. Soft-left MPs are said to admire his tone. The unions are understood to be monitoring developments, which in this context means someone from USDAW once nodded at him awkwardly while sharing a lift.

A Bloore leadership would offer Labour a clean break from the Starmer era while promising to preserve its most popular feature: an almost erotic commitment to saying very little, very firmly, in front of a plain background.

His supporters, believed to include a parliamentary researcher, two councillors, and a man who once replied great point Chris in a WhatsApp group, argue that he represents the future: sensible, understated, and extremely unlikely to be recognised in a Toby Carvery.

Maya Ellis: the Ribble Valley Reset

If Labour has learned anything from the local elections, it is that the party must reconnect with those places it has lost, those places it never quite understood, and those places it mainly imagines through stock photography.

This is where Maya Ellis comes in.

As MP for Ribble Valley, Ellis offers Labour something no other candidate can: the chance to say Ribble Valley in a serious setting.

The Ellis pitch is simple. Labour cannot win again by choosing between the cities and the towns, the suburbs and the shires, the Red Wall and the metropolitan base. It must go deeper.

It must go Ribble.

Early supporters, dubbed the Ribble Alliance, are already describing her as the candidate of renewal, by which they mean she has not yet been loudly blamed for anything. Her critics say she lacks national name recognition, but her allies counter that this is precisely the point. After the last few years, the British public may be ready for a Labour leader they have not already decided to put on mute.

In private, one supporter is understood to have said: she has a very strong backstory, once we’ve finished writing it.

Callum Anderson: the Bletchley Breakthrough

1900-1923. 1924-1929. 1931-1945. 1951-1964. 1970-1974. 1979-1997. 2010-2024. 2029?

For a party trapped in an endless cycle of interpreting and reinterpreting electoral defeats throughout its history as though it were an indecipherable code, the answer may have been hidden in plain sight all along: Bletchley.

Callum Anderson, MP for Buckingham and Bletchley, is already attracting interest from those who believe Labour must crack the code of modern politics. This phrase is expected to form the backbone of his leadership launch, his campaign merchandise, and at least one unbearable interview with The Rest Is Politics.

The symbolism is irresistible to Labour. Bletchley means codebreaking. Codebreaking means strategy. Strategy means meetings. Meetings mean PowerPoint. PowerPoint means government. It is the sort of intellectual chain senior party officials find reassuring.

Anderson’s supporters insist he offers the party a generational reset: young enough to sound fresh, serious enough to survive a select committee, and enigmatic enough to be projected onto by every faction simultaneously. The Blairites will see a moderniser. The soft left will see a listener. The hard left can see a firebrand. The unions will see someone who might answer an email. The public will see nothing at all - which may currently be Labour’s strongest available offer.

Alice Macdonald: the Norwich North Mutualist Moment

Every Labour leadership contest needs a candidate who can make the party feel briefly better about itself before everyone remembers it is not the 1945 landslide again.

Enter stage left: Alice Macdonald, Labour and Co-operative MP for Norwich North.

Macdonald’s appeal lies in her ability to combine the moral seriousness of the Co-op movement with the raw political glamour of Norwich. Hers would be a campaign of mutualism, community wealth-building, fair shares, ethical retail, and trying not to shout at one another on the NEC. In other words, completely doomed, but in a way that would produce a very beautiful launch video and several mentions of Alan Partridge.

After years of managerialism, triangulation, and ministers looking as though they are being held hostage by their own briefing notes, a Macdonald campaign would offer something genuinely different: the faint but moving possibility that politics does not have to be quite so spiritually degrading.

The problem, of course, is that Labour has historically responded to such possibilities by forming a working group, losing the minutes, and then inexplicably appointing Peter Mandelson to something again. I mean, seriously. Have we learned nothing?

Still, in a crowded field, Macdonald has one decisive advantage: she sounds like the sort of candidate most likely to know where the biscuits are kept at a community meeting room. After the 2026 results, luring exhausted activists in with the promise of snacks may be the closest thing the party has left to a mass movement.

Luke Charters: the York Outer Insurgency

Finally, no serious analysis of the Labour leadership can ignore Luke Charters, MP for York Outer.

Indeed, the case for Charters is almost embarrassingly obvious. Labour should not become inward looking. Labour must look outward. Charters represents York Outer. Sometimes politics really is that simple.

In an age when the party is divided between centrists, soft-lefts, Blue Labourites, democratic socialists, socialist democrats, former Corbynites, anti-former-Corbynites, never-Corbynites, people who claim not to care about Corbyn but still bring him up every eleven minutes, and the four remaining Labour councillors outside London, all of whom have developed a thousand-yard stare after months of gruelling trench warfare masquerading as local politics, Charters offers a much-needed new factional identity: Outerism.

Outerism rejects the tired categories of left and right. It asks instead: have we considered moving further out?

Out from Westminster. Out from the discourse. Out from the group chat. Out, ideally, beyond the point where anyone from GB News can find you.

Far out.

A Charters campaign would not seek to dominate the centre ground. It would orbit it, moving further and further until the Labour Party leaves the upper atmosphere, suspended like a planetary ring around the Earth. His pitch to the party would be calm, understated and geographically inevitable: if Labour cannot hold the centre, perhaps it can hold the outer.

Already, sources close to no one in particular suggest that the York Outer insurgency is gaining traction, which is code for I have decided to write about it.

The field is open, by which I mean nobody knows what they are talking about

Of course, critics will say that none of these MPs are likely to become Labour leader according to their sources. But these critics are trapped in the old politics, where likely means something, contender implies intent, and sources are expected to have names, knowledge or access.

The new leadership discourse is much more liberated than that.

A contender is anyone whose name can be placed in a sentence between some people believe and could emerge as a compromise candidate. A campaign is any conversation that has taken place within thirty miles of Portcullis House. A faction is three people with similar coffee orders. Momentum is when the second journalist rings the first journalist to ask whether there is any momentum.

So yes, Starmer may survive. Or he may not. Burnham might get back into Westminster. Or not. Rayner may run. Streeting may run. Miliband may be begged to return by people who have learned absolutely nothing except how to format nostalgia as analysis.

But somewhere, on the parliamentary estate, a new-intake MP is walking into a meeting unaware that they are, at this very moment, being positioned as the candidate who can heal Labour, reconnect Britain, redefine the centre-left, win back Runcorn and Gorton simultaneously, neutralise Reform, de-claw the Greens, reassure the markets, inspire young people, mollify old people, rebuild trust, and somehow explain why the party just got battered in places where the leaflets definitely all said standing up for local people.

I am being silly, of course.

But only slightly sillier than the serious coverage, which is the problem.

There is a familiar temptation, especially in religious contexts, to treat certainty as the purest form of faith. The more certain one is, the stronger one’s faith must be. The person who says I know appears to stand above the person who says I believe, and the person who says I believe, though I sometimes doubt seems weaker still.

This is understandable. Certainty is reassuring. It has the sound of courage. It tells others, and perhaps oneself, that the matter is settled.

But faith and certainty are not the same thing.

Survey data are useful here only as a starting irritant. Pew’s recent work has shown that remains unusually resilient in American life compared with much of Western Europe, and Pew’s Western European research similarly shows wide variation in belief, practice and religious identity across the region. But surveys can only tell us how people describe belief. They cannot settle what faith is.

A better starting point is conceptual. The Stanford Encyclopedia of Philosophy notes that some traditions have understood faith as a kind of knowledge attended by a certainty that excludes doubt. But it also describes other models in which faith is not simply possession of proof, but trust, venture, commitment and practical orientation. On these models, faith is not destroyed by doubt; indeed, it may be impossible without doubt of some kind.

That last point matters. Doubt by itself is not a virtue. It can be lazy, evasive, self-protective, or merely fashionable. There is nothing especially noble about standing permanently at a distance from commitment, congratulating oneself for never being naïve enough to believe anything too strongly. Doubt can keep faith honest, but it can also become a way of avoiding the risk of faith altogether.

The stronger claim is not that doubt is better than faith. It is that doubt can deepen faith when it is held inside commitment.

Kierkegaard is useful here, though he is often flattened into a slogan. His point is not simply that faith means believing absurd things for no reason. The Stanford account places him within a doxastic venture model: faith as full practical commitment to a truth that one recognises cannot be objectively secured by the evidence. His famous formulation of faith as objective uncertainty held fast with passionate inwardness is not an invitation to stop thinking. It is a description of what faith becomes when detached certainty is unavailable and yet one still has to live.

William James makes a related point from a different angle. Faith is not the arbitrary decision to believe whatever one likes. It concerns those live and momentous questions where refusing commitment is itself a form of commitment. One does not always get to wait outside life until proof arrives. Some truths, if they are to be lived at all, must be approached before they can be certified.

This is why doubt deserves a defence. Not as an end in itself, and certainly not as a pose of superior cleverness, but as one of the disciplines by which faith is kept from hardening into idolatry of itself.

The person who never doubts may be steadfast. But they may also be insulated, incurious, or merely fluent in the expectations of their community. The doubting believer, by contrast, knows that belief is vulnerable to question and continues to wrestle with it anyway. That struggle is not a failure of faith. It may be one of the things that makes faith morally serious.

Without doubt, faith can become brittle. Worse, it can become cruel. It can mistake inherited confidence for truth, certainty for righteousness, and discomfort with questioning for loyalty to God. Doubt does not solve these dangers, but it interrupts them. It leaves room for humility, repentance, self-criticism, and the possibility that one’s own understanding is partial.

If faith is to be more than social inheritance, tribal performance, or the refusal to revise oneself, then doubt is not its enemy. It is what keeps faith honest.

Annotated bibliography

Pew Research Center / AP reporting on religion in America.
Used lightly as the opening prompt rather than the foundation of the argument: survey data can show how people describe belief, but cannot settle what faith is. (https://apnews.com/article/1f1ac0da0577cfcb50f3c48e7014a070)

Pew Research Center, Being Christian in Western Europe.
Used for contrast with the American context and to frame the difference between reported belief, religious identity and lived religious practice. (https://www.pewresearch.org/religion/2018/05/29/being-christian-in-western-europe/)

Stanford Encyclopedia of Philosophy, ‘Faith’.
Used for the conceptual scaffolding of the piece: faith as certainty, faith as trust or venture, Kierkegaard’s objective uncertainty, and James’s account of faith as commitment beyond but not against the evidence. (https://plato.stanford.edu/entries/faith/)

The original outrage was broader than the endings themselves. Mass Effect had sold itself, more than almost any other series of its era, on the idea that choices mattered. Players had carried decisions across three games, saving or sacrificing characters, settling old conflicts, shaping alliances, curing plagues, preserving species, and occasionally punching journalists. Then, at the end, the whole thing seemed to collapse into three abstract options presented by a newly introduced god-child on the Citadel.

The debate has since become more specific and, somehow, even more tedious. Which ending is best? Destroy, Control, Synthesis, or Refuse?

There are many interesting takes on this.

I disagree with most of them.

Hence, regrettably, this article.

But What is the Best Ending?

There is a boring-but-true answer I could give here, which is that ‘best’ is a subjective category boundary and different people will come to different conclusions depending on what they value. That is true. It is also not very interesting. So the more useful answer is this: Synthesis is the best ending for the in-game universe, but it is not a good ending for Mass Effect as a story.

Yes, I know. No points for a noncommittal answer. But hear me out.

The distinction matters because there are really two different questions being argued over. The first is an in-universe question: which ending produces the best outcome for the galaxy? The second is a narrative question: which ending best completes the story Mass Effect has been telling?

Those are not the same question. In fact, the problem with Synthesis is that it answers the first question so neatly that it fails the second.

Within the frame of the game itself, Synthesis is quite plainly presented as the superior option. It is not just a synthesis of organic and synthetic life in the literal, in-story sense. It is also a narrative triangulation between the two binary choices the game has been placing in front of us all along: Destroy, associated with Anderson and the Alliance; and Control, associated with the Illusive Man and Cerberus.

Both of those endings have obvious benefits and obvious costs. Synthesis is written to retain the benefits while mitigating the costs.

Like Destroy, it removes the Reapers as an existential threat. But it does so without destroying the Geth, EDI, and synthetic life more broadly. Depending on your effective military strength, Destroy may also mean devastating collateral damage. Even at its best, it requires Shepard to save organic life by wiping out synthetic life.

Like Control, Synthesis preserves the Reapers’ vast knowledge and technological capacity for the use of the wider galaxy. But it does so without placing any one person, even Shepard, in a position of absolute authority over an armada of ancient Lovecraftian death machines. It avoids the obvious quis custodiet ipsos custodes problem, which Mass Effect has been worrying at since Saren and the Spectres in the first game.

On that basis, I think it is reasonable to say that the game codes Synthesis as the best ending within its own moral universe. Not necessarily the most emotionally satisfying. Not necessarily the most coherent. Not necessarily the one I would choose. But the one the game presents as the most complete resolution.

This is where some of the common objections come in. People often say that Synthesis is what Saren wanted, or that it is space communism, or that it turns the whole galaxy into a green Instagram filter with feelings. Some of these objections are more serious than others. Saren did not really want Synthesis; he wanted submission to the Reapers, and his vision of coexistence was already corrupted by indoctrination. The space communism thing is mostly internet noise. The green filter point, while aesthetically important to me on a spiritual level, is probably not the deepest moral critique available.

The consent objection is fairly credible, at least. Synthesis transforms every living and synthetic being in the galaxy without asking them. It alters bodies, minds, cultures, species, and perhaps consciousness itself. It is not just a military decision. It is a metaphysical intervention imposed universally. That is a fairly substantial ethical problem, and the game does not really know what to do with it.

This is not some minor nitpick invented by people trying to prove that Synthesis is secretly evil. The absence of consent is one of the reasons the ending feels so unsettling. Shepard does not merely choose a strategy for defeating the Reapers. Shepard chooses the future nature of life itself, on behalf of everyone, everywhere, forever. Even by the standards of space opera, that is quite a lot to authorise on a collapsing platform after a stressful conversation with a glowing child.

But I think this objection is better understood as a criticism of the writing than as evidence that Synthesis is, in-universe, supposed to be a bad ending. The game does not present Synthesis as violation, horror, or tyranny. It presents it as transcendence. The Extended Cut gives us a broadly utopian montage: conflict recedes, knowledge is shared, the galaxy rebuilds, and organic and synthetic life seem to enter a new era of mutual understanding. The consent problem is real, but it is real in a way the text largely fails to acknowledge.

That is the deeper issue. Synthesis is not bad because it’s secretly sinister. It’s bad because it’s too good.

Synthesis and Tonal Collapse

As an ending, Synthesis is poor writing. Bluntly, it reduces consequence at the very moment the trilogy should be most committed to it. It also swerves away from the series’ wider themes in order to offer a fairy-tale resolution that few people who had actually played Mass Effect were likely to find satisfying.

The earlier games are brilliant because, despite their emphasis on player choice and agency, they rarely pretend that choices are free of cost. You can shape events. You can make things better. You can prepare well, act wisely, and sometimes save people who would otherwise die. But the series usually resists the fantasy that every problem has a perfect outcome if only you completed enough side quests beforehand.

No matter how good you are, Jenkins still dies on Eden Prime. You may be able to save Kirrahe and Wrex, but someone still has to die on Virmire. There is no cost-free way to save the Council during the attack on the Citadel. Even Mass Effect 2, which softens this somewhat by allowing a perfect run through the Suicide Mission, at least makes you work for it. You have to build loyalty, commit Shepard firmly to a paragon/renegade leaning, resolve team conflicts, upgrade the Normandy, manage time properly, and choose the right people for the right tasks. The perfect outcome is possible, but it feels earned because failure is genuinely available.

Mass Effect 3 initially returns to the harsher logic of the first game. Its best moments are built around the impossibility of saving everyone. Mordin’s arc on Tuchanka works because redemption has a cost. The Geth-Quarian conflict works because peace is possible only if you’ve spent three games earning the trust and conditions that make it possible; otherwise, it can easily collapse into catastrophe. These are some of the strongest parts of the trilogy because they understand what Mass Effect is good at: making hope feel fragile, partial, and expensive.

Then Synthesis arrives and short-circuits that logic.

As presented, it is a utopian ending. The Reaper threat ends. The Geth survive. EDI survives. The Reapers help rebuild. Organic and synthetic life are reconciled. The central conflict of the trilogy is not merely resolved but apparently transcended. Everyone gets glowing green circuitry and, insofar as the ending shows us, a future of peace, understanding and shared existence.

That is not a conclusion. It is a cosmic group hug with particle effects.

The problem is not simply that the ending is happy. A happy ending can work. Nor is the problem that Mass Effect needed to end in misery. It did not. The problem is that Synthesis offers a form of happiness that feels disconnected from the series’ normal moral machinery. It gives us a choice without a meaningful cost, a victory without sufficient sacrifice, and a resolution so total that it stops feeling like Mass Effect.

That is especially jarring because the trilogy has never really been utopian. Even in the lighter first game, the galaxy is a bleak and difficult place underneath all the clean surfaces and dramatic uplighting. The series deals with racism, terrorism, slavery, genocide, political paralysis, militarism, state violence, corporate experimentation, refugee crises, and the long shadow of historical atrocity. The Reapers themselves are basically Lovecraftian horrors rendered in polished black metal. Your victories against them in the first two games are not true victories at all. You delay the crisis. You survive the encounter. You buy time.

That grimness is not incidental. It is part of the series’ texture. The Council is slow, cautious, compromised and politically evasive. The Alliance is brave but limited. Cerberus is effective but monstrous. The Spectres are a solution to bureaucracy that immediately reveals its own danger. Again and again, the series asks what you are willing to risk, sacrifice, overlook or destroy in order to achieve something good.

The final stretch of Mass Effect 3 understands this perfectly until, suddenly, it does not. Shepard reaches the Citadel bloodied, exhausted, and probably dying. Anderson dies in front of you. The Illusive Man’s ideal is revealed as another form of indoctrinated delusion. Everything points towards a final decision in which even victory must hurt.

And in every other ending, it does.

Destroy defeats the Reapers, but at the cost of synthetic life and perhaps the very relationships that proved organic-synthetic peace was possible. Control preserves the Reapers, but at the cost of Shepard becoming something distant, post-human, and frighteningly powerful. Refuse preserves moral purity, perhaps, but condemns the current cycle to destruction. These endings may be flawed, but they at least understand the series’ tragic grammar.

Synthesis does not. Or rather, it tries to rise above that grammar entirely. It offers not a hard-won political settlement, not an uneasy survival, not a victory shadowed by loss, but a metaphysical fix. The central conflict of galactic history is solved by an act of universal transformation. The ancient cycle ends. Organics and synthetics understand one another. The Reapers, who have spent hundreds of millions of years harvesting advanced civilisations, are folded into the reconstruction effort with surprisingly little awkwardness.

Again, this is why I understand the impulse behind theories that Synthesis must secretly be the bad ending. It feels too easy. It feels too smooth. It feels like the sort of offer that, in a better-written version of the story, would turn out to have a terrible hidden cost. Surely this much peace must be sinister. Surely the green glow is a warning. Surely the Reapers have won in some more subtle way.

But I do not think that is what the game is doing. I think the game means it. That is the problem.

Synthesis is not written as a trap. It is written as transcendence. The issue is that transcendence is a poor fit for Mass Effect. The series has always been at its best when dealing with compromise, consequence and partial victories. It is a story about building coalitions among people who mistrust each other for good reasons. It is about persuasion, sacrifice, history, loyalty, and the frustrating work of making common cause in an imperfect galaxy. The best moments do not erase conflict; they make resolution feel possible despite it.

Synthesis skips that work at the end. It solves the galaxy from above.

That is why the ending fails for me. Not because Synthesis is obviously the worst outcome for the people living in that universe. On the contrary, if you take the game’s presentation at face value, it may well be the best. It saves the geth. It saves EDI. It stops the Reapers. It preserves knowledge. It ends the cycle. It offers the possibility of peace between forms of life that the series has repeatedly placed in conflict.

But that is precisely why it fails as an ending.

The best ending for the Mass Effect universe is not necessarily the best ending for Mass Effect as a story. In fact, in this case, it is the opposite. Synthesis is too clean, too complete, too totalising, too uninterested in the costs that would make it dramatically and morally legible. It gives the galaxy everything, and in doing so it gives the story too little.

That, finally, is why the debate will probably never end. The Synthesis ending is infinitely defensible in-universe and infinitely unsatisfying out of universe. It is morally attractive and narratively weak. It is the best answer to the Reaper problem, and a bad answer to the story problem.

What makes Synthesis a bad ending is not that it is secretly evil. It is that the game genuinely believes it is good.

Addendum: My Secret Fifth Ending

There is, however, one final complication.

While I absolutely love the series, I hate the fact that the lasting legacy of Mass Effect has been reduced to this endless back-and-forth over which of the endings we think should be the right one. Destroy, Control, Synthesis, Refuse. Pick your colour, choose your metaphysics, justify your war crime, and prepare to argue about it forever.

After a certain point, the endings do not need to be theorised, interpreted, explained or analysed any more. They need to be ignored. Which is, if we take Warlord Okeer seriously in Mass Effect 2, the greatest insult one can deliver to an enemy.

Because the endings are, hands down, the worst part of the trilogy. They are a shocking dismissal of the tone and style established by the writers over five years, two preceding titles, and a small mountain of tie-in books and comics. The series spent hundreds of hours building a universe of compromise, consequence, tactical uncertainty, historical grievance and hard-won coalition-building, only to end by asking us to choose between three abstract space levers and a sulking refusal option. There comes a point where analysis starts to feel like collaboration.

So, me? I take the WarGames route. I choose not to play.

The Reapers must lose, obviously. Let us not be silly about this. But Destroy is doing it wrong. Synthesis (as we’ve just discussed at length) is contrived and unsatisfying. Control is irresponsible. Refuse is sociopathic. So I leave the galaxy as it is: Reapers, Cerberus and the allied forces locked forever in unresolved tension, suspended at the last possible moment before the assault on Cronos Station locks you out of the free-roaming Galaxy Map.

This is not cowardice. It is curation.

In my unofficial version of events, Shepard retires to Anderson’s extremely tasteful apartment on the Citadel, plays Armax Arena, visits the arcade, throws a party, has several emotionally necessary conversations, and allows the final mission to remain indefinitely unlaunched. The galaxy is still on fire, yes, but it is at least on fire somewhere else. Commander Shepard discovers that the most powerful war asset is shore leave.

And oddly, this works better than it has any right to.

The Citadel DLC is one of the most perfectly self-aware pieces of storytelling in gaming. It is not merely fan service, though it is absolutely that. It is a love letter to the characters, the tone, the friendships, the jokes, the accumulated emotional architecture of the trilogy. It understands something the formal endings somehow forgot: that Mass Effect was never only about the Reapers.

So my secret fifth ending is not canonical. It is not even really an ending. It is a refusal of the ending as offered. Through the crucial interaction between the ludic rules of the game and the narrative satisfaction of the setting, the final attack never happens. Shepard and co. remain on break. The Reapers wait patiently, twiddling their tentacles. The galaxy hangs in the balance. Garrus is somewhere calibrating something. The arcade high scores become, in practice, the last meaningful objective.

It was not a formal ending.

But it should have been.

Sykes-Picot.

Sometimes this is done with a map. Sometimes with a thread. Sometimes with the weary confidence of someone who has just remembered that imperialism existed and would now like the rest of us to know about it. Iraq, Syria, Lebanon, Palestine, Israel, Jordan, the Gulf, Iran, sectarian conflict, jihadism, borders, mandates, oil, coups, state failure, the lot: all of it, apparently, can be traced back to two men with a ruler, a secret agreement, and an unusually high tolerance for straight lines.

It is not hard to see why this story appeals. It has villains. It has a document. It has a map. It has a date. It has that special explanatory tidiness that history almost never provides, but which public argument is always desperate to find. It also has the great advantage of being partly true.

The Sykes-Picot Agreement really was a secret imperial bargain. In 1916, Britain and France, with Russian assent, discussed how Ottoman Arab territories might be divided into zones of control and influence if the Entente powers won the war. The agreement’s own language is full of the mechanics of imperial privilege: priority of enterprise, local loans, advisers, foreign functionaries, ports, spheres, access, control. It is not a pretty document. It does not need to be rescued from its reputation.

So the problem with the Sykes-Picot myth is not that Mark Sykes, François Picot, or their less-mentioned Russian counterpart Sergei Sazonov were innocent. They were not. The problem is that the agreement has become too useful and too lazy an explanation. It is asked to explain more than it can bear.

That distinction matters. A historical episode can be symbolically powerful while still being explanatorily inadequate. Sykes-Picot is an excellent symbol of imperial arrogance. It captures, in a particularly neat form, the habit of European powers treating foreign lands as pieces on a board; assets to be allocated, routes to be secured, influence to be apportioned. If you want a shorthand for the moral atmosphere of wartime imperial diplomacy, Sykes-Picot will do nicely.

But shorthand is not explanation. And the more we treat Sykes-Picot as the master key to the modern Middle East, the more the actual history disappears.

This mattered in 2014, when Islamic State made the symbolism literal. Its propaganda campaign around the End of Sykes-Picot presented the demolition of the Iraq-Syria frontier as an assault on the old colonial order. Daesh released English- and Arabic-language material showing fighters destroying parts of the border infrastructure and symbolically creating a single territory across Iraq and Syria. The image was irresistible: jihadists with bulldozers erasing the line drawn by European imperialists.

It still matters now. In 2026, with Gaza devastated by prolonged war, Israel expanding territorial control inside the strip, and a wider regional conflict involving Iran disrupting politics and shipping across the Gulf, the temptation to reach for a grand origin story remains powerful. Reuters reported today that Netanyahu had directed Israeli forces to expand control in Gaza to 70% of the territory; separate reporting has described the 2026 Iran war as having widened the conflict across the region, with consequences for Gulf shipping and neighbouring states.

That does not make the colonial age irrelevant. It does mean we should be wary of letting it become a historical vending machine: insert crisis, receive colonial explanation.

The older I get, the less patience I have with explanations that are satisfying in exactly the way they are inadequate. The Sykes-Picot story is one of these. It offers an answer which is not quite wrong, but which becomes wrong through overuse. It gives us a morally legible past, at the cost of flattening the very thing we claim to be trying to understand.

The real history is worse, and more interesting.

The collapse of the Ottoman Empire was not settled by Mark Sykes and François Georges-Picot in 1916. It unfolded through a long, violent and contingent crisis stretching from before the First World War to several years after it. To make sense of the modern Middle East, one has to widen the frame: the Italian invasion of Libya in 1911, the Balkan Wars of 1912–13, Ottoman defeat and radicalisation, the rise of the Young Turks, German-Ottoman strategy, Russian ambitions in the Black Sea, British anxieties over India and the Suez Canal, French claims in Syria and Lebanon, Arab dynastic politics, Zionism, Armenian genocide, Greek expansionism, the Mandate system, the Turkish War of Independence, and the final settlement at Lausanne in 1923.

That is not as elegant as two men and a ruler. History is often inconsiderate like that.

Sean McMeekin is useful here because he forces the argument out of its narrow Anglo-French groove. The Berlin-Baghdad Express follows Germany’s attempt to use its Ottoman alliance, the Berlin-Baghdad railway, and pan-Islamic politics as instruments of world power during the First World War. The Ottoman Endgame widens the story further, framing the making of the modern Middle East through war, revolution and imperial collapse across the first three decades of the 20th century. One need not accept every emphasis in McMeekin’s account to see the corrective value. The Ottoman Empire was not simply a passive body on an operating table while Britain and France sharpened the knives. Germany was not incidental. Russia was not marginal. Ottoman leaders were not decorative, and local power-brokers and strongmen within the empire were not waiting politely for Europeans to pick them.

This is the first thing the Sykes-Picot myth obscures: the post-Ottoman Middle East was not made by one agreement. It was made by a series of interconnected wars spanning decades, born of European imperial ambitions but also of local political, ethnic, cultural and religious divides.

The second thing it obscures is that the war did not end neatly in 1918.

If the modern settlement had simply been imposed by Allied fiat, the Treaty of Sèvres in 1920 would have mattered more than Lausanne. Sèvres was the punitive, maximalist postwar settlement for the Ottoman Empire. It envisaged enormous losses of sovereignty and territory, including Allied zones, Greek expansion, and provisions that Turkish nationalists considered intolerable. But Sèvres was never fully ratified or implemented. It was overtaken by force. The Turkish national movement under Mustafa Kemal fought, won, and compelled the Allies back to the negotiating table.

The Treaty of Lausanne in 1923 then recognised the boundaries of the modern Turkish state. Turkey made no claim to its former Arab provinces; the Allies abandoned several claims they had previously pursued, including autonomy for Turkish Kurdistan and territorial cession to Armenia. Whatever else Lausanne was, it was not Sykes-Picot being rubber-stamped seven years late. It was the result of military defeat, nationalist victory, diplomatic retreat and hard bargaining.

This is where the familiar story begins to wobble. If Sykes-Picot was the blueprint, why did so much of the final settlement emerge through later war and revision? Why did Sèvres fail? Why did Lausanne replace it? Why did Mosul become a subsequent dispute? Why did the borders of Turkey come to be settled by Kemalist success rather than by the wartime dreams of British and French officials? Why did then-Persia remain intact? Why did some projected zones of influence never take the form imagined for them?

The answer is not that Sykes-Picot did not matter. It did. The answer is that it mattered as one episode in a wider imperial scramble, not as the master document from which everything else followed.

This is especially important in Palestine, because Sykes-Picot is sometimes also made to stand in for the whole tangled set of British wartime commitments: promises to Arab leaders, understandings with France, Zionist diplomacy, strategic anxieties, and later Mandate administration.

The significance of this is not that Sykes-Picot is unrelated to Palestine. The agreement placed a rump state in Palestine under an international administration to be agreed with Russia and other allies, which is itself revealing. The point is that Palestine cannot be explained by Sykes-Picot alone. The conflict’s later development involved the Balfour Declaration, British Mandate policy, Zionist immigration and institution-building, Palestinian Arab opposition, imperial retreat, UN partition, war, displacement, state formation, occupation, nationalism, religion, regional rivalry, settlement, American power projection, and repeated failures of diplomacy. To say Sykes-Picot over this history is not to explain it. It is to put a label over it.

The same is true of Iraq and Syria. The border between them - one of the few lines from the original Sykes-Picot map which did basically survive the agreement - became the poster child for the Sykes-Picot myth, but the actual states themselves were formed through mandate rule, local revolt, military occupation, dynastic bargaining, administrative improvisation and later authoritarian consolidation. Their problems cannot be reduced to the artificiality of a single line. The more useful question is not whether a state’s borders are artificial, but whether the institutions inside them gained legitimacy, capacity and durability.

This is where the Sykes-Picot explanation becomes actively misleading. It encourages us to imagine that the Middle East’s central problem is bad cartography. As if the region would have been spared its tragedies had only the lines been more aesthetically sensitive, perhaps with a little more attention to ethnographic shading. This is comforting, in its way. It locates violence in the geometry of the map rather than in the harder history of power: armies, parties, monarchies, oil companies, foreign interventions, patronage networks, sectarian mobilisation, secret police, coups, sanctions, invasions, occupations and civil wars.

Bad borders can matter. Of course they can. Borders can trap minorities, divide communities, reward clients, punish enemies, and convert administrative convenience into permanent grievance. But the great error of the Sykes-Picot story is to confuse map logic with political history.

This is also why the anti-imperial critique loses force when it becomes too neat. British and French (and Russian and German and Italian) imperialism did enormous damage in the region. It treated local agency and sovereignty as something to be managed (at best), ignored, or destroyed (at worst). There is no need to soften that. But blaming Sykes-Picot for everything risks letting imperialism off the hook ,by making it seem like a single decision made by bad actors rather than a system of practices.

The British did not need Sykes-Picot to be duplicitous in Palestine. The French did not need Sykes-Picot to pursue their ambitions in Syria and Lebanon. Russia did not need Sykes-Picot to covet Armenia or Constantinople and the Straits. Germany did not need Sykes-Picot to cultivate pan-Islamic agitation against its enemies. Ottoman leaders definitely did not need Sykes-Picot to centralise, gamble with, repress, deport, or massacre their own citizens. Later dictators did not need Sykes-Picot to build prisons, secret police states, and cults of leadership. The United States did not need Sykes-Picot to invade Iraq. Iran and Saudi Arabia did not need Sykes-Picot to pursue regional competition. Israel did not need Sykes-Picot to occupy Palestinian territory. Hamas did not need Sykes-Picot to massacre civilians. Hezbollah, Assad, the Ba’athists, the Hashemites, the PLO, the Gulf monarchies, the Iraqi army, the CIA, the IDF, the IRGC, and UN diplomats all belong somewhere in this history too.

That is not a pithy slogan. It is not meant to be. The untidiness is the whole point.

One reason the Sykes-Picot myth survives is that it performs a moral simplification. It gives progressive Western audiences a way of admitting imperial guilt without having to understand very much about what they’re admitting to. It says: yes, we made a mess; two of our men drew the wrong line; the consequences followed. There is a kind of shallow contrition in this. It looks like historical seriousness, but in the worst case it’s just another form of self-centring. Even in guilt, Europe still gets to be the centre of attention.

The actual history is less flattering and less convenient. The injustices of European empires mattered profoundly, but they were not omnipotent. Local actors mattered enormously too, though not always nobly. Anti-colonial movements could be both liberatory and authoritarian. National self-determination could coexist with ethnic violence. Imperial retreat could produce freedom in one place and abandonment in another. The Ottoman Empire was not a paradise of pluralism before Europeans descended with their pencils. Nor was it simply a doomed relic awaiting partition. It was a real imperial state, modernising and decaying, coercive and adaptive, capable of reform and catastrophe.

This is why Lausanne matters so much as a corrective. It reminds us that the post-Ottoman order was not merely imposed; it was contested. The Turkish national movement did not accept the settlement prepared for it. It fought, and it changed the result. That does not make the final order just. Lausanne came with its own human costs, including the compulsory population exchange between Greece and Turkey. It does, however, make the history harder to narrate as a simple morality play.

A better account would treat Sykes-Picot as one layer in a palimpsest. Beneath it lie Ottoman structures, provincial identities, imperial rivalries, religious communities, trade routes, dynastic ambitions and local politics. Around it sit the Hussein-McMahon correspondence, the Balfour Declaration, Russian claims, French and British military occupations, and wartime bargaining. Above it come Sèvres, the mandates, Lausanne, state-building, anti-colonial revolts, coups, oil politics, Cold War alignments, American intervention, Islamist movements, authoritarian collapse and the accumulated failures of governing elites.

That is harder to fit on a graphic. It is also much closer to reality.

None of this means Sykes-Picot should be discarded. Symbols matter in politics because people act through symbols. Daesh understood that perfectly well. It chose the Iraq-Syria border not because it had conducted a dispassionate seminar on the finer points of wartime diplomacy, but because Sykes-Picot had become a name for humiliation, division and imposed order. In that sense, the myth became politically real. A bad explanation can still motivate real action.

But historians and commentators owe us more than the reproduction of a myth because the myth is potent. The fact that Sykes-Picot matters symbolically is precisely why it needs to be handled carefully. Otherwise, we end up confusing the history people invoke with the history that happened.

The question, then, is not whether Sykes-Picot was bad. It was.

It is not whether imperialism matters. It does.

It is not whether borders can produce grievances. They can.

The question is whether one secret agreement in 1916 can bear the explanatory weight placed upon it. It cannot.

Sykes and Picot belong in the story, but not in the starring roles to which posterity has oddly promoted them. They were participants in a wider process: one crowded with soldiers, ministers, revolutionaries, dynasts, nationalists, colonial officials, railway men, generals, propagandists, oil interests, religious leaders and insurgents. The story’s dramatis personae come from pretty much every major European state involved in the First World War, but also draws its key figures from North Africa, the Middle East, and as far afield as Afghanistan. It was written in chancelleries, certainly; but also on battlefields, in collapsing provinces, in refugee columns, in cabinet rooms, in nationalist assemblies, in imperial offices and in the ruins of failed settlements.

A famous name is not the same thing as an adequate explanation.

That, finally, is the trouble with Sykes-Picot. Not that it tells us nothing, but that it lets us stop too soon. It allows us to mistake recognition for understanding. It gives us the satisfying click of a pattern falling into place, when what we need is the patience to see how many patterns were layered over one another.

The Middle East was not broken by a line. It was remade through the collapse of empires, the violence of war, the improvisations of victors, the resistance of the defeated, the ambitions of local rulers, the failures of new states, and the repeated interference of outside powers who were almost always less clever than they imagined and more destructive than they admitted.

Sykes-Picot is part of that history. It is not the history.

And if we want honest answers about the modern Middle East, we need to stop mistaking a useful symbol for the truth.

Annotated bibliography

The Sykes-Picot Agreement, 1916, Avalon Project.
Used for the primary text of the agreement itself, particularly its language around British and French zones of control and influence, local loans, advisers, and proposed arrangements for Palestine. (https://avalon.law.yale.edu/20th_century/sykes.asp)

Meghan Tinsley, ‘Whose colonialism? The contested memory of the Sykes-Picot Agreement’.
Used for the discussion of Daesh/Islamic State propaganda in 2014, especially the End of Sykes-Picot campaign and the symbolic destruction of the Iraq-Syria border. (https://pomeps.org/whose-colonialism-the-contested-memory-of-the-sykes-picot-agreement)

Reuters, ‘Netanyahu directs Israeli forces to expand Gaza control to 70 percent’.
Used to frame the 2026 relevance of the article: the Middle East remains in violent flux, and current crises again tempt commentators towards sweeping historical explanations. (https://www.reuters.com/world/middle-east/netanyahu-directs-israeli-forces-expand-gaza-control-70-percent-2026-05-28/)

Reuters, ‘How has the Iran war affected Middle East states?’
Used for the current 2026 regional context, including the wider effects of the Iran war on neighbouring states, Gulf shipping and regional politics. (https://www.reuters.com/world/middle-east/how-has-iran-war-affected-middle-east-states-2026-05-13/)

Sean McMeekin, The Berlin-Baghdad Express.
Used as a major interpretive source for widening the story beyond Britain and France, especially Germany’s Ottoman strategy, railway geopolitics and wartime use of pan-Islamic politics. (https://www.hup.harvard.edu/books/9780674064324)

Sean McMeekin, The Ottoman Endgame.
Used for the broader frame of Ottoman collapse between 1908 and 1923, helping to move the essay from a narrow 1916 focus to the longer sequence of revolution, war, partition, resistance and settlement. (https://www.penguin.co.uk/books/187942/the-ottoman-endgame-by-mcmeekin-sean/9780718199715)

Balfour Declaration, 1917, Avalon Project.
Used to distinguish Sykes-Picot from Britain’s later and separate commitment to support a Jewish national home in Palestine, while supposedly protecting the civil and religious rights of existing non-Jewish communities. (https://avalon.law.yale.edu/20th_century/balfour.asp)

Treaty of Sèvres summary.
Used for the discussion of the failed 1920 postwar settlement, which Turkish nationalists rejected and which was superseded after the Turkish War of Independence. (https://en.wikipedia.org/wiki/Treaty_of_S%C3%A8vres)

Britannica, ‘Treaty of Lausanne’.
Used for the 1923 settlement: recognition of the modern Turkish state’s boundaries, Turkey’s abandonment of claims to its former Arab provinces, and the Allies’ retreat from several earlier demands. (https://www.britannica.com/event/Treaty-of-Lausanne-1923)

Qatar Digital Library, ‘Lausanne’s Legacy’.
Used for the point that Lausanne delimited boundaries and was tied to the compulsory population exchange between Greece and Turkey, helping avoid an over-neat or celebratory account of the Turkish nationalist victory. (https://www.qdl.qa/en/lausanne%E2%80%99s-legacy-peace-treaty-led-century-conflict)

In April 2025, the UK Supreme Court gave gender-critical campaigners much of what they had spent years demanding. In For Women Scotland Ltd v The Scottish Ministers, the Court held unanimously that the words man, woman and sex in the Equality Act 2010 refer to biological sex. The Court’s reasoning was not that trans people have no rights under the Act. Quite the opposite: it explicitly affirmed that trans people remain protected under the characteristic of gender reassignment, and may also be protected in other ways by sex discrimination and harassment provisions. But on the central interpretive question, the Court was clear. For the purposes of the Equality Act, sex means biological sex.

That clarification has now moved from judgment into institutional guidance. On 21 May 2026, the Minister for Women and Equalities laid the EHRC’s updated Code of Practice for services, public functions and associations before Parliament. Unless Parliament disapproves it within the review period, the code will come into force as statutory guidance. The EHRC’s own summary says that, for Equality Act purposes, a person’s legal sex is the sex recorded at birth, and that obtaining a Gender Recognition Certificate does not change that position under the Act.

So here we are. The law has, in important respects, caught up with the gender-critical argument. Biological sex is now treated as the operative category for the Equality Act. Single-sex provision can be organised on that basis. Organisations have been told to take a practical and proportionate approach to toilets, changing rooms, services, associations and public functions.

One might have expected that to settle something.

It has not.

Instead, it has exposed the argument beneath the argument.

What are we Talking About?

This is not a piece about whether sex ever matters. Of course sex matters. It matters in medicine, some sports, pregnancy and maternity, sexual violence, intimate care, and some forms of single-sex provision. There are contexts involving trauma, coercion, undress, vulnerability, privacy or bodily risk where it is not merely permissible to notice sex, but negligent not to. A liberal and tolerant society does not have to pretend that these things are irrelevant. In fact, it would probably become less liberal and less tolerant if it did.

Nor is single-sex provision inherently irrational. There are women who have survived male violence and who have good reasons to experience some spaces differently from others. There are service providers who have to make difficult decisions under competing legal, ethical, and practical obligations. There are institutions that must weigh inclusion against privacy, dignity, risk, feasibility, evidence and the needs of particular users. These are not imaginary problems. They cannot be resolved by slogans, however inclusive the slogan sounds.

But that is precisely the point. There is a difference between sex-conscious policy and disgust-conscious politics.

Sex-conscious policy asks a set of disciplined questions, critically and on a case-by-case basis. What is this service for? Who uses it? What risks are actually present? What does the law require? What does the evidence suggest? What alternatives are available? What would exclusion achieve? Who would be harmed by inclusion? Who would be harmed by exclusion? Is the rule proportionate to the aim?

Disgust-conscious politics does something else. It begins with a bodily flinch, then goes looking for a principle grand enough to dignify it. It starts with discomfort around a person’s voice, face, clothes, body, name, mannerisms or self-description, and then retrofits that discomfort into arguments about safeguarding, feminism, children, free speech, science or ‘reality’. The policy language may be serious, but the emotional engine is often something older, uglier, and harder to admit.

That distinction matters because the trans debate is now full of people who speak as though they are doing one thing while plainly doing another. They say they are defending women’s rights, but keep returning to the physical revulsion they feel towards trans women. They say they are protecting children, but talk about gender-questioning young people as though they are symptoms of a social ill. They say they are defending reality, but use reality as a polite synonym for disbelief in another person’s social existence.

This is not true of everyone who holds a gender-critical position. That needs saying clearly. Some people have serious, bounded, good-faith concerns about the interaction between sex and gender identity in law. Those concerns should be argued with, not psychoanalysed away. But it is also dishonest to pretend that all anti-trans rhetoric sits in that register. A great deal of it does not. A great deal of it is not policy reasoning at all. It is the politics of recoil.

The old analogy here is with the anti-gay rhetoric I grew up listening to in the years either side of the millennium, when the gay age of consent and later the issue of marriage equality were the ‘culture war’ of the age. The argumentative machinery is often strikingly familiar; one vulnerable out-group is presented as the thin end of the wedge for some larger horror.

• ‘If gay people are allowed to marry, the sanctity of marriage will be undermined and the family unit will collapse.’ (What does this mean?)

• ‘If gay men are allowed to adopt, what will it do to the children?’ (Implying that either gay parents would indoctrinate ‘their’ children to be gay too, or - more sinisterly - that this would erode the protection of children and enable paedophiles.)

• ‘What, will people marry their dogs next?’ (No, because there’s a world of moral difference between consenting adults and schnauzers.)

The claim is rarely just that this particular group wants something. It is that granting rights to one group is somehow transactional with another, and will open the door to predators, corruption, contagion or civilisational decay in some form or another.

This does not mean every contemporary concern about trans rights is equivalent to old-fashioned homophobia. But it does mean we should recognise the pattern. Moral panics often work by smuggling an older disgust response into a newer public vocabulary. Nobody says, or at least nobody respectable says, I want the law to enforce my revulsion. They say they are thinking of the children. They say they are protecting the vulnerable. They say common sense has been outlawed by fanatics. They say there is a silent majority. They say they are only asking questions, usually while standing next to a very large pile of answers.

The point is not that predators do not exist. They do. The point is that predators become rhetorically useful. The imagined faux-transgender male predator in a wig, lurking in a women’s toilet, is asked to do a suspicious amount of political work. They stop being merely a risk to be assessed. They become a symbolic solvent, dissolving distinctions between trans women, abusive men, cross-dressers, sex offenders, drag queens, identity-testing adolescents, Stonewall, queer theory, gender clinics, university HR departments and whatever someone on Twitter said yesterday. Once the panic is fully assembled, all of these become parts of one shadowy machine.

That is not safeguarding. That is folk horror.

Spaces and Gender.

The issue with toilet panic is not that safety is irrelevant. Women’s safety is obviously relevant. So are privacy and dignity. The problem is that the public version of the argument often bears only a loose relationship to practical safeguarding. A serious policy discussion would ask what kinds of facilities are being discussed, who uses them, how they are supervised, what evidence exists about risk, whether single-user provision is available, and how staff should respond to harassment or assault. It would distinguish between a school changing room, a nightclub toilet, a hospital ward, a domestic abuse refuge, a leisure centre, and a small workplace with two lockable cubicles in a corridor.

Instead, we often get a morality play about toilets. On one side, helpless women and girls. On the other, trans women treated less as people than as an aperture through which an unrestrained male violence will pour.

The evidential basis for that panic is thin. One US study found no evidence that gender identity non-discrimination laws increased criminal incidents in restrooms, locker rooms and changing rooms, though one should be cautious about importing American evidence wholesale into a British legal context. More importantly, even if risk varies by setting, the answer cannot simply be to treat all trans presence as predatory presence. That is exactly the slippage that good policy exists to prevent.

The serious question is not whether every trans woman must be admitted to every women’s space in every circumstance. That absolutism is politically fragile and practically unhelpful. The serious question is what justifies exclusion in a particular setting. A rape crisis service working with traumatised women may reach one conclusion. A public library toilet may reach another. A changing room in a small leisure centre may require a third solution altogether.

The government’s own framing of the draft code emphasises practical, proportionate decision-making, safety, dignity and freedom from harassment. That is what the debate should be about. Not whether trans women make some people uncomfortable. Not whether someone can imagine a lurid scenario. Not whether a boundary feels emotionally satisfying. What rule is justified here, for this purpose, with these effects?

Think of the Children!

The same distinction applies, even more urgently, to children.

Here, the argument is genuinely difficult. Anyone who pretends otherwise is selling something. The Cass Review has changed the terrain. NHS England’s implementation plan states that children and young people’s gender services are being redesigned around a more holistic model of care, and that there is insufficient knowledge about who may benefit from medical interventions. It also says that such treatments, where considered, should be available through carefully constructed research arrangements to build a better evidence base. In December 2024, the government announced that restrictions on puberty blockers for under-18s would be made indefinite, citing advice from the Commission on Human Medicines and the Cass Review’s concerns about the evidence base.

None of this should be brushed aside. Clinical caution is not moral panic. Evidence matters. Children’s healthcare should be careful, especially where interventions may have long-term consequences and where the evidence base is contested. There are legitimate questions about social transition, medicalisation, waiting lists, comorbid distress, family dynamics, safeguarding, consent, clinical governance, detransition, and how to support young people without either rushing or abandoning them.

But clinical caution is not the same thing as cultural panic.

Clinical caution says: the evidence is uncertain, the stakes are high, and care pathways need reform.

Cultural panic says: trans children are proof that society has gone mad.

Clinical caution asks how to support distressed young people safely.

Cultural panic asks who infected them.

Clinical caution is interested in service design, data, follow-up, risk, consent and therapeutic support.

Cultural panic is interested in villains.

This is where the rhetoric becomes dangerous. The figure of the child is almost always the most powerful weapon in a moral panic. Children are vulnerable. Children are suggestible. Children can be harmed. All of that is true. But precisely because it is true, think of the children arguments require more discipline, not less. Otherwise, children become rhetorical hostages. Their vulnerability is used to end the argument rather than improve it.

There is a good version of the concern about children. It worries that some young people may be medicalised too quickly. It worries that gender distress may sometimes be entangled with trauma, autism, sexuality, sexism, homophobia, body image, depression or social contagion. It worries that institutions may adopt affirming language without a sufficiently robust clinical framework behind it. These are serious concerns.

There is also a bad version. It imagines trans identity as a contagion spreading through schools, TikTok, activist teachers, weak parents and fashionable ideology. It treats young people not as complex human beings in distress, but as evidence of a decadent culture. It is less interested in helping children than in prosecuting the adults it already disliked.

Again, the distinction matters. A society can be cautious about puberty blockers without becoming cruel to transgender and gender-questioning children. It can insist on evidence without turning every trans adult into a groomer, every supportive parent into a fanatic, and every gender-questioning teenager into a dupe. The tragedy of the present debate is that too many people seem to regard these distinctions as evasions. They are not evasions. They are the minimum conditions of thought.

What is Really Real?

Then there is reality.

Kathleen Stock’s Material Girls is subtitled Why Reality Matters for Feminism. Stock’s central claim, broadly put, is that biological sex has material significance and that gender identity theory should not override it in contexts such as law, data, healthcare and women-only provision. There is a serious argument there. Bodies matter in a whole host of contexts. Reproductive sex matters. Law cannot function if its categories become completely detached from the purposes for which they exist. Medicine cannot be run on vibes. Data collection cannot be subordinated entirely to politeness. Feminism, if it is to mean anything material, must be able to talk about female embodiment, male violence, pregnancy, menstruation, sexual dimorphism and the social meaning attached to sexed bodies.

The problem is not the word reality in itself. The problem is what it can be made to imply.

In a careful argument, reality means constraint. It means that not every subjective claim can settle every institutional question. It means that law and medicine need categories stable enough to do their work.

In a careless argument, reality means dismissal. It means: you are not what you say you are. It means: your identity is make-believe. It means: I will tolerate your costume, perhaps, but not your claim on the shared world.

The difference is not small. One is a claim about institutional limits. The other is a claim about social illegitimacy.

This is where high-profile progressive women in the gender-critical debate have often been most revealing. Germaine Greer has not merely argued that womanhood should be defined by sex. She has done so with a relish for bodily disparagement, famously describing trans women in terms that linger on voice, hair, hands, physical incongruity, and a dehumanising “it”. The argument is not simply I disagree with this legal category. It is look at this body. The disgust is not incidental. It is part of the performance.

"On the day that The Female Eunuch was issued in America, a person in flapping draperies rushed up to me and grabbed my hand. ‘Thank you so much for all you’ve done for us girls!’ I smirked and nodded and stepped backwards, trying to extricate my hand from the enormous, knuckly, hairy be-ringed paw that clutched it. The face staring into mine was thickly-coated with pancake makeup through which the stubble was already burgeoning, in futile competition with a Dynel wig of immense luxuriance and two pairs of false eyelashes. Against the bony ribs that could be counted through its flimsy scarf dress swung a polished steel women’s liberation emblem."

J. K. Rowling’s rhetoric is more complicated, and it is worth being fair about that. In her 2020 essay, she wrote that trans people deserve protection and acknowledged the vulnerability of trans women, particularly those exposed to male violence. She has also argued, repeatedly, that sex matters, that women’s spaces matter, and that women should not be punished for saying so. I don’t believe those claims can simply be waved away as transgender hatred.

But Rowling’s public argument also repeatedly returns to the idea that recognising trans women as women opens the door to any man who wishes to enter women’s spaces. Her 2019 support for Maya Forstater framed the issue around whether women were being forced out of jobs for saying that “sex is real”. The trouble is not that sex is real is false. It is that the phrase functions as though it settles questions it has barely begun to ask. Sex is real. But so is gender dysphoria. So is gender-affirming care. So is social transition. So is violence against women. So is violence against trans people. So are privacy, dignity, law, fear, trauma, recognition and disgust. Reality is not a single trump card. It is the whole difficult mess.

Greer, Stock and Rowling are not the same thinker, and it would be lazy to pretend otherwise. Greer often sounds openly contemptuous. Stock is more philosophical and institutional. Rowling writes as a survivor of male violence and as a public campaigner who sees herself as defending women from a misogynistic ideology. Their arguments differ in tone, seriousness and object.

But across much of this terrain, the same slide recurs. A legitimate point about the role of sex as a category boundary becomes a general suspicion of trans identity. A concern about safeguarding becomes a lurid fantasy of invasion. A defence of reality becomes a licence to treat other people’s lives as delusion. The argument begins with policy and ends with metaphysical exasperation at the existence of trans people at all.

The Argument Beneath the Argument.

It is not enough, in a liberal society, to say that someone makes you uncomfortable. Many people make many other people uncomfortable. Gay couples make some people uncomfortable. Public displays of affection make some people uncomfortable. Masculine women make some people uncomfortable. Feminine men make some people uncomfortable. Disabled bodies make some people uncomfortable. Religious clothing makes some people uncomfortable. Neurodivergent behaviour makes some people uncomfortable. The sight of poverty makes some people uncomfortable. Ageing makes some people uncomfortable. Public grief makes some people uncomfortable. The human body, in most of its actual biological and social variety, is a near-endless affront to those who feel they were promised a tidy world where everyone would be basically just like them.

Some of this discomfort may be understandable. That does not make it authoritative. Disgust is a real human response, but it is a terrible political philosopher. It is quick, certain, embodied, and almost always convinced of its own innocence. It tells us that what repels us must be wrong, and that what unsettles us must be dangerous. It gives us moral clarity at the exact moment we should be most suspicious of ourselves.

This is why the progressive self-image matters. There is something especially jarring about people who have spent their intellectual lives criticising oppression, essentialism, moral panic and patriarchal control, only to become so incurious about their own recoil when the subject is trans people. Feminism at its best is not merely the defence of one’s own category. It is a discipline of critical attention to power.

The law has now clarified some things. It has not clarified everything, it has muddied some things (particularly the legal status of gender, which now sits in a sort of limbo) and it certainly has not absolved us of the need to think. There will still be hard cases. There will still be bad policies, bad-faith activists, frightened institutions, confused guidance, opportunistic politicians, and people who would rather shout adult human female or trans women are women until the room empties, than do the miserable work of institutional design.

But if we are going to have the argument, we should at least have the real one.

The real argument is not whether sex ever matters. It does.

The real argument is not whether women’s safety matters. It does.

The real argument is not whether children’s healthcare should be evidence-led. It must be.

The real argument is whether those truths justify the wider cultural-political project now being built around them: a project in which trans people are treated not merely as participants in difficult policy questions, but as symbols of suspicion, fraud, danger, unreality and social decay.

That is where the line should be drawn.

A liberal society cannot be built on the rule that nobody must ever encounter a person whose body, identity, name, pronouns, or existence makes them uncomfortable. Discomfort is real. It may even be understandable. We are all, in one way or another, carrying the prejudices of the world that made us. But discomfort is not, by itself, an argument.

And it is certainly not enough to carry the moral weight that anti-trans rhetoric keeps asking it to bear.

Sources

UK Supreme Court, For Women Scotland Ltd v The Scottish Ministers, press summary.
Used for the opening legal context: the Court’s 2025 ruling that ‘man’, ‘woman’ and ‘sex’ in the Equality Act 2010 refer to biological sex, while also noting that trans people retain protection under the characteristic of gender reassignment. (https://supremecourt.uk/cases/press-summary/uksc-2024-0042)

House of Commons Library, ‘Supreme Court judgment on the meaning of “sex” in the Equality Act 2010: For Women Scotland’.
Useful as a clearer parliamentary briefing on what the judgment did and did not do; especially the point that the ruling gave an authoritative interpretation of the existing law rather than formally changing the statute. (https://commonslibrary.parliament.uk/research-briefings/cbp-10259/)

EHRC, ‘UK Supreme Court ruling on the meaning of sex in the Equality Act: our work’.
Used for the 2026 update: the EHRC’s account of how it has responded to the Supreme Court ruling, including the statement that legal sex for Equality Act purposes is sex recorded at birth and that a Gender Recognition Certificate does not change this for Equality Act purposes. (https://www.equalityhumanrights.com/our-work/uk-supreme-court-ruling-meaning-sex-equality-act-our-work)

EHRC, ‘Code of Practice for services, public functions and associations’.
Used to establish the procedural status of the updated code: laid before Parliament on 21 May 2026, subject to the 40-day review period, and intended to become statutory guidance once commenced. (https://www.equalityhumanrights.com/equality/equality-act-2010/codes-practice/code-practice-services-public-functions-and-0)

GOV.UK, ‘Equality Act 2010: Draft Code of Practice for services, public functions and associations 2026’.
Used for the discussion of practical, proportionate decision-making by service providers, and for the distinction between serious sex-conscious policy and symbolic boundary-making. (https://www.gov.uk/government/publications/equality-act-2010-draft-code-of-practice-for-services-public-functions-and-associations-2026/equality-act-2010-draft-code-of-practice-for-services-public-functions-and-associations-2026)

NHS England, ‘Children and young people’s gender services: implementing the Cass Review recommendations’.
Used for the section on children and clinical caution, especially the shift towards redesigned, holistic gender services and the post-Cass emphasis on evidence, safety and service reform. (https://www.england.nhs.uk/long-read/children-and-young-peoples-gender-services-implementing-the-cass-review-recommendations/)

GOV.UK, ‘Ban on puberty blockers to be made indefinite on experts’ advice’.
Used in the discussion of puberty blockers and youth transition pathways, following the Government’s December 2024 decision to make restrictions indefinite. (https://www.gov.uk/government/news/ban-on-puberty-blockers-to-be-made-indefinite-on-experts-advice)

House of Commons Library, ‘Hormone treatments for children and young people: Clinical guidance, policies and regulation’.
Useful for the most compact current-policy summary: restrictions on puberty blockers, review planned for 2027, and NHS England’s 2026 consultation on stopping routine prescription of gender-affirming hormones to children and young people. (https://commonslibrary.parliament.uk/research-briefings/cbp-10596/)

Hasenbush, Flores and Herman, ‘Gender Identity Nondiscrimination Laws in Public Accommodations’.
Used cautiously in the spaces/toilets section as evidence from the US context that gender identity non-discrimination laws were not associated with increased criminal incidents in restrooms, locker rooms and changing rooms. (https://escholarship.org/uc/item/4rs4n6h0)

Williams Institute summary of the same public accommodations study.
A more accessible source for the same point: it summarises the finding that inclusion of gender identity in non-discrimination laws did not affect the number or frequency of criminal incidents in the relevant facilities. (https://williamsinstitute.law.ucla.edu/publications/ma-public-accommodations/)

Kathleen Stock, Material Girls: Why Reality Matters for Feminism.
Used as a primary example of the ‘reality’ argument: I try to treat Stock as making a serious philosophical and institutional case about sex, gender identity theory, and feminism, while criticising the way ‘reality’ can slide from institutional constraint into social dismissal. (https://www.hachette.co.uk/titles/kathleen-stock/material-girls/9780349726595/)

J. K. Rowling, ‘J.K. Rowling Writes about Her Reasons for Speaking out on Sex and Gender Issues’.
Used as a primary example of Rowling’s own framing of her position, including her arguments about sex, women’s spaces, vulnerability, safeguarding and her personal experience of male violence. (https://www.jkrowling.com/opinions/j-k-rowling-writes-about-her-reasons-for-speaking-out-on-sex-and-gender-issues/)

J. K. Rowling, 2019 Maya Forstater tweet.
Used as a compact example of the slogan-like function of ‘sex is real’ in the public debate: a phrase that is not false, but which often operates as though it settles much more than it actually does.

Germaine Greer, ‘On Why Sex Change is a Lie’ / quoted extracts.
Used as the clearest example of rhetoric that moves beyond policy disagreement into bodily disgust, particularly Greer’s appalling description of a trans woman in terms of voice, clothing, hands and physical revulsion. (https://www.artandpopularculture.com/On_Why_Sex_Change_is_a_Lie)