The first group says: name the evil. Hold the words intact. Refuse the sanitised language. Keep the moral siren ringing until the building empties. The work of describing what is happening, in the register the institutions are working hard to suppress, is a load-bearing form of public action and a precondition for everything else.

The second group says: the moral case is already won, or already lost, depending on how you look at it. Polls say what the polls say. The institutions are not moving anyway. What actually moves Western governments off positions for which the moral defence has already collapsed is patient, unromantic, mechanism work. Divestment campaigns over years. Voting blocs built over decades. Internal institutional defection cultivated over generations. The lever work, slow and frustrating and heroic only in retrospect.

Both groups are correct about their own subject. The moral siren is necessary and the institutions designed to amplify it have largely failed to do so. The patient lever work is the only mechanism that has historically moved Western states off positions where the moral case had already been lost. A serious response to the present moment requires both kinds of work, and most of the people doing one of them are dismissive of the people doing the other, which is its own kind of failure.

This piece is not about the contradiction between the siren and the lever. The contradiction between those two is real but it is the smaller one. This piece is about a deeper contradiction that becomes visible only when you accept both of the above arguments and then ask the question they both decline to ask, which is: how much time do the historical mechanisms require, and how much time do we actually have?

The Tools We Have, and What They Were Built For

It is worth being precise about both of the inherited tools, because the case I want to make rests on the fact that they are good tools designed for a situation that no longer obtains.

The moral siren is the older of the two. It is the prophetic register, the public denunciation, the writer or preacher or reporter who refuses to look away and refuses to soften the language. It runs from the Hebrew prophets through Las Casas through Frederick Douglass through Émile Zola through Hannah Arendt through James Baldwin through, in our own moment, the small number of journalists and public intellectuals willing to use the word genocide about Gaza without flinching. The function of the moral siren is to maintain the field of what is publicly sayable against constant institutional pressure to corrupt it. It does not, on its own, change policy. It changes the cultural conditions under which other forms of action become possible.

The lever is the second tool, and it is younger. The patient mechanism work I described in another piece, the kind that ended Vietnam and dismantled apartheid and won civil rights, is largely a product of the post-1945 period and depends on a specific set of preconditions that have to be true at once for it to function. When those preconditions hold, it works. When they don't, it doesn't.

What were the preconditions?

The first was a basically growing economy. The lever work assumes that the political class can absorb compromise, that material concessions to one constituency do not have to come out of the hide of another, and that the time horizon over which costs can be socialised is long enough that today's losers can be promised they will be tomorrow's winners. None of the great post-1945 reform projects in the West were paid for out of contraction. They were paid for out of growth. The Civil Rights Act and the Voting Rights Act passed during the longest sustained expansion of US median wages in American history. The British post-war welfare state was built on the expectation of a rising tide. The German federal model was constructed with American capital and rebuilt onto an industrial base that the war had not completely destroyed. The lever worked when there was slack in the system to absorb the cost of moving it.

The second precondition was demographic momentum. Long-horizon political projects require a population structure in which young cohorts are larger than old cohorts and in which the time investment of organising voters, training cadres, cultivating elites, and building institutions can be expected to pay off because the people doing the organising will still be there to harvest. This held for most of the twentieth century in the West. It does not hold any more. The median age in every Western country is now older than the people who built the post-war institutions were when they built them, and the demographic profile is now compressing rather than expanding.

The third precondition was a stable ecological backdrop. The post-war reform projects assumed that the physical conditions of human life would remain roughly constant over the timescales of political organising, that the harvests would come in, that the rivers would flow, that the climate envelope inside which agricultural civilisation developed would still be there in 2030. None of those assumptions are still safe.

The fourth precondition was institutional legitimacy. The lever work required that the institutions being moved retained enough residual authority that defection from them was meaningful. A judge had to be a judge whose ruling other people would respect. A civil servant had to be a civil servant whose resignation would carry weight. A senator had to be a senator whose committee chairmanship could compel testimony. When the institutions hollow out, the levers that operated through them stop being connected to anything.

The fifth precondition was time. The lever work always took longer than the people doing it expected, and longer than the people watching it wanted. Vietnam ran for seven years after the moral case had collapsed. The anti-apartheid struggle took thirty years from Sharpeville to majority rule. The American civil rights project took a hundred years from emancipation to the Civil Rights Act, and another generation after that to deliver even the partial implementation we have today. The lever requires time the way a slow fire requires fuel, and the people who succeeded with it succeeded because they were willing to spend their entire working lives on a project they would not see completed.

All five of those preconditions held, in some combination, in the period when the great post-war reform projects were possible. In the present moment, none of them clearly hold, and the most important of them, the time precondition, is the one that the people who care most about the lever work are most reluctant to look at directly.

What the Time Horizon Actually Looks Like

The argument that has run through several of our previous pieces is that the present moment is not a moment within history of the kind that historical mechanisms were built for. It is something different. It is the period in which the conditions under which those mechanisms worked are themselves coming apart, simultaneously, in a synchronisation that previous crises did not produce.

The data on this is not contested by serious people. The IPCC's Sixth Assessment Report, the UN Environment Programme's Global Resources Outlook, the demographic projections from the UN Population Division, the Earth systems work coming out of the Stockholm Resilience Centre, and the energy transition arithmetic from sources as conservative as the International Energy Agency all converge on the same picture. The biosphere is moving outside the envelope inside which agricultural civilisation developed. The fossil energy regime that built modern life is in its late phase and has no successor at equivalent scale and density. The demographic transition is well advanced almost everywhere, which means the labour and consumption base that growth economics requires is contracting rather than expanding. The political institutions designed to coordinate collective responses are losing legitimacy faster than they can be reformed, and the information environment that would be needed to rebuild that legitimacy is itself fragmenting into mutually unintelligible pieces.

The crucial feature of these crises is not their individual severity. It is their synchronisation. Each of them alone would be the defining problem of a generation. Together, they are something else, because each one consumes the resources and attention and institutional capacity that would be needed to address the others. The lever work that solved Vietnam happened against a stable ecological backdrop. The lever work that ended apartheid happened in a global economy that was still expanding. The lever work that won civil rights happened inside institutions that, whatever their faults, still functioned well enough that a Supreme Court ruling meant something and a federal civil rights commission could compel state action. None of those background conditions can be assumed any more, and the timescales over which the deteriorations are running are measured in years and single-decade periods, not in the multi-generational horizons that the lever required.

If the climate, the energy regime, the demographic profile, and the institutional legitimacy of Western states are all going to look substantially different in 2035 than they do in 2026, then any political project whose payoff requires ten or twenty years of patient construction is being launched into conditions that may not exist when its payoff is supposed to arrive. The civil rights movement could plant in 1955 because there was a 1965 to harvest in. The anti-apartheid movement could plant in 1960 because there was a 1990 to harvest in. The question we have to ask honestly is whether there is a 2045 that the seeds being planted now will still recognise, and the honest answer is that we do not know, and that the central case is probably not.

This is the precise point at which the contradiction becomes unavoidable. The patient lever work is the only thing that has historically delivered the kind of change the moral siren is calling for. And the conditions that made the lever work possible are themselves vanishing on a timescale faster than the lever work can run. We have inherited two good tools from the previous era, both of which were designed for a situation that is ending, and neither of which has a good answer to the question of what to do when the situation has changed underneath them.

The Position We Cannot Avoid

The position this puts us in is not comfortable, and the discourse on every side has good reasons for not naming it.

The moral siren tradition does not want to name it because once you accept that the institutional response is unlikely to arrive in time, the prophetic register starts to look like a form of moral self-stewardship rather than a political programme. The siren keeps the language intact for the people who will need it later, but if there is no later in which the institutions are still functioning enough to be moved, the function of the siren reduces to a kind of historical witnessing for an audience that may not exist.

The lever tradition does not want to name it because the entire emotional and intellectual investment of patient mechanism work depends on a faith that the harvest will eventually come in. The civil rights organisers of the 1950s could spend their lives on the project because they believed, and were largely correct to believe, that there was a longer arc in which their work would be vindicated. The lever work in the present moment requires the same faith, and that faith is much harder to sustain when the climate envelope is closing and the demographic base is contracting and the institutions through which the lever operates are themselves losing legitimacy. It is much easier to keep doing the work if you do not look too directly at the question of whether the conditions for the work to pay off will still exist.

The activist response of the last decade has split the difference by adopting forms of action that look like both at once: direct action that performs the moral siren function while also gesturing at lever work, mass mobilisation that produces the affective experience of organising without committing to the multi-decade discipline that the lever historically required. This is not a criticism of the people doing it. It is a recognition that the form of action available to a generation that knows it does not have multi-decade horizons is going to look different from the form of action available to a generation that thought it did. The split-the-difference approach is the honest expression of working with two inherited toolkits in a moment that neither was built for.

But the split-the-difference approach does not, by itself, generate a third tool. It is what people do when they cannot quite admit that the first two are insufficient. And the central question of the present moment, the question that almost no one is willing to ask in plain language, is what the third tool would look like.

What the Third Tool Might Be

The honest answer is that nobody knows yet, because nobody has ever needed it before. The historical record of human responses to civilisational-scale stress is thin and mostly grim: the responses we know about are the ones in which the political class hardened its grip, the elites cannibalised whatever was left, and the broader population was sorted into those who had access to the surviving pockets of order and those who did not.

These three movements are not a hazard to avoid. They are the present operational state of most Western democracies, and any honest description of the third tool has to take that as the starting condition rather than as a future risk. The hardening is visible in the executive order regimes, the deportations of legal residents for protected speech, the threats to broadcast licences, the prosecutions of academics, the dissolution of protest groups, the criminalisation of forms of speech that were lawful five years ago. The cannibalisation is visible in the corporation tax concentration, the data centres consuming sovereign electricity, the savings schemes that subsidise existing wealth-holders, the institutional landlord model in housing, the way every recent crisis has produced a transfer from the diffuse public to a concentrated private interest. The sorting is visible in the housing crisis that locks out an entire generation, the bifurcated healthcare systems, the visa apartheid, the gated communities, the literal walling-off of populations rendered surplus to the new economy. None of this is anticipated. It is current. It has been current for some time. Anyone designing a third tool inside it has to start from that fact, not from the more comforting assumption that the work is happening in the period before the failure mode arrives.

That changes what the third tool has to do, and it changes the question of who the third tool is for. We will return to that second question, because it is the one most of the existing literature on local resilience has refused to engage with, and it is the one that determines whether the work is part of the solution or part of the sorting.

But there are some features of the third tool that can be guessed at, by the simple method of asking what kind of work is robust to the failure modes of the first two and to the active sorting that is already underway. The first two tools were both designed to operate at the scale of the nation-state and to work through formal institutions over multi-generational timescales. The third tool, if it exists, almost certainly works at smaller scales, through less formal channels, on shorter timelines, and with the explicit assumption that the larger systems will not deliver.

What does that look like in practice?

It looks like local food systems built by people who do not assume that the global agricultural supply chains will keep functioning at current capacity through the 2030s, and who therefore plant orchards and rebuild soil and re-establish seed networks not because they expect to produce a counter-economy at scale but because they want their region to have something to fall back on when the larger system stutters.

It looks like local energy infrastructure built by people who do not assume that the national grid will be reliable, and who therefore install distributed generation, microgrids, community storage, and the technical knowledge to maintain them, on the principle that the redundancy itself is the value, regardless of whether the larger grid actually fails.

It looks like local water security, in the literal sense of knowing where the rivers and aquifers are, who has rights to them, what condition they are in, and how much of the local population could actually be supplied if the long-distance pipelines became unreliable. This is mostly the work of small public bodies and community trusts that nobody pays attention to, in places that nobody writes articles about, but it is the kind of work that becomes load-bearing when other things stop bearing.

It looks like the construction of mutual-aid networks at the scale of streets and parishes and small towns, on the assumption that the formal welfare state of the post-war period will be increasingly unable to deliver the protection it was designed to deliver, and that the gap will have to be filled by people who already know each other's names. This is unromantic work and it does not produce the affective experience of political action, but it is the layer that holds when other layers fail.

It looks like the deliberate transmission of difficult skills, low-input agriculture, basic medicine, repair and maintenance of old machinery, traditional construction methods, languages and knowledge systems that the global standardisation has been eroding, to people young enough to use them. The point is not nostalgia. The point is that the skills will be needed and the people who hold them are dying.

It looks like the formation of small political units willing to make decisions on the assumption that the larger political unit is not going to. A county council that decides its job is now managed contraction. A small national government that openly stops pretending the growth model is coming back. A city that builds for the population it can actually sustain rather than the population the developers wish it had. These are the political analogues of the local resilience work, and the same logic applies: the value is in the redundancy, not in the hope of replacing the larger system.

It looks like institutions of cultural transmission, libraries, schools, community presses, that take seriously the possibility that the longer arc of recorded human knowledge may need to be carried through a period of disruption by people who deliberately set themselves the task. This is an old idea. It worked once before, in the Irish monasteries and elsewhere, and the conditions under which it would need to work again are no longer unimaginable.

Who the Third Tool Is For

The risk inside everything written above is the risk that this whole approach becomes another mechanism of the sorting. If the third tool is just local resilience-building done by the people who can already afford to do it, in the regions that already have the social capital to organise it, then it is not a third tool at all. It is one more advantage flowing to people who already have advantages, dressed up in the language of preparedness and community. The parish that already has a working community trust adds a seed library and a microgrid. The neighbourhood that already has a residents' association adds a mutual-aid network. The small town with a functioning town council declares itself in managed contraction. Each of these is real and useful and survivable for the people inside it, and each of them, on the present trajectory, makes the people outside slightly worse off, because every functioning local system pulls professional capacity, social trust, and political attention out of the pool that the people on the wrong side of the sort would otherwise have access to.

This is not a hypothetical objection. The early signs of it are everywhere in the existing localist literature. The eco-villages that turn out to be priced for the upper middle class. The transition towns that produce wonderful conferences and almost no transitional infrastructure for the working-class neighbourhoods two miles away. The off-grid homesteading scene that is structurally a hobby for people with capital. The intentional communities that become extensions of class self-selection. None of these are bad in themselves. They become a problem when they are mistaken for the third tool, because the people doing them are mostly the people who would have been on the right side of the sort anyway, and the work they are doing, however genuine, does not travel.

The question of who the third tool is for is therefore not a side question. It is the central question, and it is the one that determines whether the work is doing anything that the sorting itself is not already doing. A third tool that only protects the comfortable is a fourth wall in the gated community. A third tool that builds redundancy in the neighbourhoods that the formal systems have already abandoned, on terms decided by the people in those neighbourhoods, is a different kind of thing.

This is harder than it sounds. The neighbourhoods that need the work most are also the neighbourhoods with the least capacity to organise it, because the same processes that have made them vulnerable have also stripped out the social infrastructure the work depends on. The patient lever tradition had an answer to this problem, which was the union, the church, the working-class political party. Those institutions are mostly gone or hollowed out. Rebuilding the equivalent in conditions where the formal organising bodies have been removed is itself a major piece of work, and there is no shortcut for it. The third tool, done honestly, includes the rebuilding of the social infrastructure that would be needed to do the third tool in the places where it would actually matter. That is a recursive problem, and it is the one most local-resilience writing avoids.

Why This Is Not a Counsel of Withdrawal

The objection that always arrives at this point is that the picture I am describing is a counsel of localism, of withdrawal from the larger struggle, and of giving up on the political institutions that are still in place. It is not.

The lever work and the moral siren work both still need to be done, because they are the only tools we have for the question of what happens to the larger institutions in the time we have left, and because in some places, against the central case, they may still deliver. The lever work in particular has the property that even when it does not produce the headline outcome, it produces second-order effects that matter: the trained organisers, the cultivated networks, the legal precedents, the inhabited language, all of which become useful in a different way when the conditions change. The moral siren work creates the cultural conditions in which the local resilience work becomes possible, by maintaining the public language in which the threat can be discussed and the alternative can be imagined.

The point is not to abandon either tool. The point is to stop pretending that either tool alone is going to deliver, and to add to the mix a third kind of work that is not waiting on the institutional response and is not betting on the multi-generational harvest. The third kind of work is what you do when you take seriously, in your own life and your own region, that the central case is not the one in which the institutions hold. It does not require giving up on the institutional case. It requires not betting your survival on it, and not pretending that the survival of your particular pocket of it is the same as the survival of the broader social fabric in which it sits.

The medical metaphor is helpful only up to a point. A doctor treating a chronic illness does not stop the slow treatments while also managing the immediate attacks, and that is the right model for combining the lever, the siren, and the third tool. But the medical metaphor breaks down on the question of the patient. A doctor has one patient. The third tool has to choose, every day, which patient it is for, because the resources and the attention and the trained capacity are finite, and because the sorting is already running in the background and absorbing whatever is not deliberately directed elsewhere. The lever and the siren are the slow treatment. The local resilience work is the management of the next attack. The question that is unique to the third tool, and that the inherited tools did not have to answer, is whose attack it is managing, and whose it is not.

The Honest Closing

We are in a moment for which we have no playbook. The two inherited tools, the moral siren and the patient lever, are both real and both necessary, and neither is sufficient on its own, and neither was designed for a situation in which the conditions for the lever to operate are themselves coming apart on a timescale faster than the lever can run. The third tool, the one that would be sufficient for our actual situation, has not yet been invented, because no one has ever needed it before.

It is also being attempted inside a process that is already running. The hardening, the cannibalisation, and the sorting are not future failure modes that the third tool might prevent. They are the present operational state of the systems the third tool has to work within, and any honest account of the work has to start there. The question is not whether to build the third tool in time to avert the failure mode. The failure mode is here. The question is whether the building of the third tool, done at the necessary scale and in the necessary places, can produce something that is not just one more layer of the sorting it claims to be answering.

That is harder than the previous version of the question, and it is the version that the localist and resilience traditions of the last twenty years have mostly avoided, because their constituency has been the people on the right side of the sort and their language has been the language of preparedness for individuals and communities that already had the means to prepare. A serious third tool requires the rebuilding of the social infrastructure that the lever tradition used to rely on, in places where that infrastructure has been deliberately stripped, by people who do not look like the typical resilience-conference attendee. Whether that rebuilding is happening at any meaningful scale right now is an open question, and the honest answer is mostly no.

This is uncomfortable, but it is not a reason to despair, and it is certainly not a reason to retreat into the moral self-stewardship that mistakes its own purity for a programme. The situation is unprecedented in the specific sense that no historical model fully applies, and that means the work of the present moment is partly the work of inventing the missing tool by attempting it and seeing what holds, and partly the work of refusing the version of the tool that would just become a fourth wall in the gated community. The people who succeed in that work will mostly not know they have succeeded, because the verification will come later, in conditions they will not be present to witness. The people who fail will fail in two distinct ways: some will fail by building resilience that turns out not to hold when it is needed, and some will fail by building resilience that holds only for themselves and contributes nothing to anyone outside the gate. The second kind of failure is worse, because it is invisible to the people committing it, and because it is structurally indistinguishable from success on its own terms.

The work is therefore both more humble and more urgent than the inherited tools suggest, and more politically demanding than the existing local-resilience literature has been willing to acknowledge. More humble, because the people doing it cannot know whether what they are doing is enough. More urgent, because the timescales the inherited tools assumed are no longer available. More politically demanding, because the question of who the work is for is no longer one the work can avoid by gesturing at community in the abstract.

The lever and the siren both still matter. So does the third thing, which we are inventing as we go, inside conditions that are already producing winners and losers on a daily basis, with the third tool itself in danger of becoming one of the mechanisms by which the losers are being made. Most of us will be wrong about most of it. That is what working without a playbook means. It is not a reason to stop working. It is a reason to be honest about what the work actually is, including the part of it that the comfortable version of localism has been pretending is somebody else's problem.

The siren rings. The lever turns slowly, in a system whose own gears are wearing out. The sorting continues in the background, faster than either of the inherited tools can move. And in the space that remains, a third kind of work is waiting to be done, by the people who can hold all three of those things in their head at once, who can pick up a shovel without mistaking their own garden for the world, and who can keep building in the places where the social fabric is thinnest, knowing that those are the places where the work matters most and where it is least likely to be done.

This article is a synthesis of arguments made in three previous pieces: "The Acceptance," "The Island That Went First," and "What the Surplus Hides." It draws on the IPCC Sixth Assessment Report, the UN Environment Programme's Global Resources Outlook, the demographic projections of the UN Population Division, the Earth systems research of the Stockholm Resilience Centre, the historical literature on the Vietnam anti-war movement and the international anti-apartheid movement, the deep adaptation work of Jem Bendell, and the long tradition of writing on civilisational transitions from Tainter through Wright through Catton through Donella Meadows.

Overwatch Report is an independent publication. We have no financial positions in any entity mentioned.