There is a joke in Finding Nemo that works because it is both terrible and perfect. Marlin, the anxious clownfish, is trying to be funny. He reaches for the old line: “With friends like these, who needs enemies?” But because this is a fish joke, it becomes: “With fronds like these, who needs anemones?” It is a groaner. But the original joke survives underneath the pun because it contains a sharp little truth: sometimes the people on your own side can do as much damage as your opponents. That, unfortunately, is where we are with international development assistance.

The anti-aid right had a simple story: aid was wasteful, naïve, corrupt, indulgent and unaffordable. It was money sent “over there” while people struggled “over here.” It was an easy target: foreign, complex, morally earnest and poorly understood. That argument was often crude. It was often cynical. But it was clear. It is the same argument the right-wing press has been feeding the British public for decades: money wasted frivolously while hardworking families struggled back in the UK.

The more interesting failure was not that aid’s opponents attacked it. That was predictable. It has been consistent for more than a decade. The failure was that so many of aid’s natural defenders responded by reaching for their own guns. This was friendly fire. At precisely the moment when a broad coalition was needed to defend the basic principle of international development assistance, the coalition fragmented. Instead of saying, clearly and confidently, that aid has saved lives, supported reform, helped countries through crisis, expanded opportunity and represented something serious countries should do, many inside and around the sector chose to say something narrower.

Instead, too often, the message was this: Aid has failed, unless it is my kind of aid. That became the unspoken refrain. Aid had failed because it was not political enough. Aid had failed because it was not local enough. Aid had failed because it was not sufficiently feminist, catalytic, adaptive, decolonised, climate-aligned, private-sector-led, politically smart, systems-oriented, investment-focused, evidence-based or transformative. Each critique contained some truth. That was precisely the problem. Truths, badly timed and politically misused, can become ammunition. There is a world of difference between saying “aid must keep improving” and saying “aid has failed.” Too many people who knew better blurred that distinction. At the very moment when aid needed defending, the sector was off debating what aid 2.0 should look like.

Some of the friendly fire came from parts of the left. For them, the attack on aid was not simply a threat to public spending, global solidarity or international responsibility. It was an opportunity to confirm a prior theory: that international development was a colonial construct, a mechanism of dependency, a continuation of empire by technocratic means. There are serious critiques here. Development assistance has too often reflected unequal power. It has too often allowed donors to define problems, set priorities, control money and measure success on their own terms. The language of partnership has not always matched the reality of power. But critique is not the same as demolition.

When the anti-aid right says “aid has failed,” and parts of the left reply “yes, but for different reasons,” the practical effect is not liberation. It is permission. Permission to cut. Permission to retreat. Permission to tell the public that nothing of value is being lost. The sharper problem, though, was not the ideological left. It was the international development intelligentsia itself.

The sector’s technocrats, thinkers, advisers, consultants, academics and senior practitioners often knew that aid had achieved a great deal. Many had seen it first-hand. They knew that vaccines had been delivered, children educated, health systems strengthened, famines mitigated, refugees supported, reformers backed, institutions built, diseases treated and crises contained. But many still accepted the premise that aid was fundamentally failing, because doing so gave force to their preferred reform agenda.

For some, the answer was political economy. Aid had supposedly failed because it had been too technocratic and insufficiently attentive to power, incentives, institutions and elite bargains. There is a lot of truth in that. Political economy matters. It matters enormously. But it was sometimes presented as though the insight had just been discovered, as though everyone before had been wandering around with logframes and blindfolds. That was not my experience. When I was at DFID from 2009 onwards, political economy analysis was already part of the conversation. We talked about it constantly. And I’m pretty sure practitioners had been talking about it long before then. Was it always done well? No. Was it always acted upon? No. Did bureaucratic incentives sometimes turn political insight into a box-ticking exercise? Of course. But that is a different argument from saying aid had somehow been innocent of politics until the clever people arrived.

Others argued that aid had failed because it was not sufficiently focused on capital, investment and financial mobilisation. Development assistance, they said, should be used to unlock private finance, de-risk investment and leverage much larger flows. Again, there is truth here. Development finance can be valuable. Blended finance can work well where the conditions are right: where capabilities are in place, investable opportunities exist, institutions can manage risk, and the missing ingredient is affordable capital. But this was not a new revelation either. Blended finance has been around for a long time. Nor is it a universal solution. Where capabilities are weak, where institutions are fragile, where projects are not yet investable, where markets are thin, or where the binding constraint is not simply the cost of capital, financial engineering will not magically create development.

Used well, development finance is an important tool. Used badly, it becomes a theology. That was the deeper problem. Too much of the aid debate became theological.

The humanitarians thought the long-term development people were too slow and abstract. The governance people thought service delivery was superficial unless it shifted power or hit a standard of institutionalised sustainability that the NHS would struggle with. The economists worried that social programmes were sentimental unless they generated growth. The private finance advocates thought grants were old-fashioned. The localisation advocates treated anything insufficiently locally led as suspect. The systems thinkers distrusted anything too measurable. The results specialists distrusted anything that could not be measured. Everyone had a critique. Everyone had a better model. Everyone could explain why someone else’s aid was not the real thing. And so, at the very moment when aid needed defenders, it got reviewers. We needed to stand together. Instead, we debated aid 2.0.

This matters because politics is not a seminar. In a seminar, you can distinguish between aid modalities, instruments, theories of change, delivery channels, fiduciary risk, absorptive capacity and institutional incentives. In politics, messages collapse. The public did not hear: “International development assistance has achieved a great deal, but must evolve.” It heard: “Even the experts admit aid does not work.” That was a gift to those who wanted to cut it.

None of this means aid should have been defended uncritically. It should not. Not all international development expenditure is well used. Some of it is badly designed. Some of it is too slow. Some of it is too donor-driven. Some of it is captured by fashion. Some of it is more concerned with announcing ambition than delivering results. Some of it fails. But that is true of all public spending. Defence procurement fails. Infrastructure projects fail. Industrial strategies fail. Domestic welfare programmes fail. Tax incentives fail. No one concludes from this that the state should abandon defence, roads, welfare, investment or economic policy altogether.

Aid, by contrast, is often judged by an impossible standard. Any failure becomes evidence against the category. Any waste becomes proof of systemic futility. Any scandal becomes a morality play. Those of us who have worked with aid budgets should be honest about failure. But honesty should not become self-harm. In my own time overseeing aid budgets, I saw things that did not work. I saw programmes that were too optimistic, projects that moved too slowly, volatile budgets that made planning implementation a struggle, reporting burdens that distorted incentives, and designs that underestimated politics, capacity or context.

But I saw far more examples of international development assistance being used well. I saw spending that I was proud of. Not perfect spending. Not spending immune from criticism. But serious, purposeful, defensible spending that improved lives, supported reform, responded to crisis, helped countries navigate difficult transitions and made the world a safer place.

That experience matters because it sits so uneasily with the fashionable despair of the aid debate. The truth is not that aid has failed. The truth is that aid is difficult. Development is difficult. Public action across borders is difficult. Working in fragile states is difficult. Supporting reform without controlling it is difficult. Balancing humility with ambition is difficult. Spending money quickly, ethically and effectively in complex environments is difficult. But difficulty is not failure.

The tragedy of the friendly fire was that too many people inside the sector mistook sophistication for scepticism. They thought being clever meant being disenchanted. They thought seriousness required distancing themselves from the basic case for aid. They were so eager to say “not that kind of aid” that they forgot to defend aid itself. And the anti-aid right understood the opportunity perfectly. It did not need to win the argument on development effectiveness. It only needed to create enough doubt, enough embarrassment, enough fragmentation, enough elite self-criticism, that the defenders of aid would hesitate.

That hesitation enabled the sector to be hit. A better response would have sounded different. It would have said: Aid works often enough to be defended, and fails often enough to be improved. It would have said that political economy matters, but it is not a magic key. That investment matters, but it is not a substitute for public goods. That localisation matters, but it is not an excuse for donors to abandon responsibility. That grants, loans, guarantees, technical assistance, humanitarian relief, budget support, research, diplomacy and development finance all have their place. It would have said that no single tribe owns the future of development.

Most of all, it would have refused the premise that aid’s imperfections invalidate its purpose. The lesson is not that the aid sector should stop arguing with itself. Internal criticism is necessary. Reform is necessary. Some orthodoxies deserve to be challenged. Some institutions deserve to be uncomfortable. Some old habits deserve to die. But there is a time to improve the ship and a time to stop it sinking.

When aid came under attack, too many of its defenders behaved as though the real danger was that the wrong kind of aid might survive. So they sharpened their critiques, refined their frameworks, attacked rival models and congratulated themselves on their sophistication. Meanwhile, the budget was cut, the public case weakened, and the people who had never believed in international development assistance at all watched the friendly fire do its work.

As I wrote in my previous article, I do not believe aid is dead. There is a case (like the inverted dead parrot sketch) that the sector has been well and truly stunned. Knocked over. It’s been hit by bullets from its enemies and perhaps most surprisingly from friendly fire. Next time, the coalition needs more discipline. Not silence. Not groupthink. Not nostalgia. Discipline. The discipline to distinguish reform from capitulation. The discipline to criticise aid without validating bad-faith attacks on it. The discipline to say that some aid is flawed without implying that aid itself is folly. The discipline to remember that winning an internal argument is not the same as winning a political one.

Because the most dangerous wound is not always the one inflicted by your enemies. Sometimes it is the one delivered by people standing on your own side, insisting, as they reload, that they are only trying to help.