The Machinery Does Not Care: Why Great Powers Slide Into Wars Nobody Chose

What the road to 1914 reveals about US–China rivalry, multipolarity, and the wars that structure produces without anyone choosing them.

In the summer of 1914, the three men who could have stopped the war were cousins.

George V of Britain, Wilhelm II of Germany, Nicholas II of Russia — bound together in one family. George and Wilhelm were both grandsons of Queen Victoria; Nicholas was married to her granddaughter, and cousin to the other two through the same dynastic web. They holidayed together. They wrote to each other in English, signed the letters with childhood nicknames. Nicky. Willy. In the last days of July they exchanged telegram after telegram, frantic, personal, begging each other to hold back the machinery.

The machinery did not care that they were family.

Within weeks their three empires were feeding a generation into the mud. Two of the three cousins would lose their thrones. One would lose his life, and his children’s.

The historian Odd Arne Westad has spent a career studying how great powers slide toward catastrophe, and the warning is precise: the people at the top in 1914 were not stupid, and they were not powerless. They were trapped inside a structure that made their personal relationships irrelevant.

That structure is assembling again. This is about how.

This is not an essay about whether war is coming. It is about how a structure can make catastrophe likely — with no one at the top having chosen it.

The Argument Before This One

For more than a century, one question has divided everyone who studies the catastrophe of 1914: was the First World War an accident or a verdict? One tradition treats it as a tragic stumble — sleepwalkers, miscalculation, a crisis that ran ahead of the men managing it. Another treats it as something closer to a sentence passed: a rising Germany, a declining Britain, a structural collision that the diplomacy of July merely dated rather than caused. The argument has never fully settled, and it matters now because we have started, quietly, to apply it to ourselves.

The Coming Storm refuses the choice. The late nineteenth century, in its account, was not a peace that failed but a particular kind of order — a long great-power calm resting on conditions that happened to hold together, and then, within a single generation, happened to come apart. A British hegemon that carried its role lightly. Old empires content, mostly, to expand outward rather than against each other. Elites in Vienna, St. Petersburg, and London more frightened of their own workers than of one another. None of this was permanent. None of it was designed to be. And when the supports were removed — Britain’s relative decline, Germany’s rise, military power dispersing faster than anyone could track — the same continent that had avoided major war for ninety-nine years produced, in its hundredth, a war that killed more men in four months on the Somme than all the great-power battles of the previous century combined. The point is not that the calm was fake. The point is that calm is not the same thing as stability, and a system can pass from one to the other without anyone deciding that it should.

The Machine, Described

Start, then, with the thing that is easy to get wrong. The First World War is often blamed on its alliances — the two armed camps that turned a murder in Sarajevo into a continental catastrophe. That is the comfortable version, because it locates the fault in a structure that can be named and, presumably, avoided. But the alliances did not cause the war. They failed to prevent it. The distinction is the whole argument.

What actually did the work was two mechanisms running at once, and neither of them was a treaty.

The first was the cult of the offensive. By 1914, military power across Europe had become a matter of railway timetables. Every general staff believed that victory belonged to whoever struck first and struck fast — Germany’s plan to defeat France in six weeks before Russia could mobilise, France’s faith in the headlong infantry assault, Russia’s bet on sheer numbers arriving in time. Each plan was a clock. And once one clock started, every other clock had to start, because to wait was to lose. The effect was to compress the time available for diplomacy until there was, in any real crisis, almost none. A system can be full of cautious men and still move toward war if the structure they have built rewards speed and punishes hesitation.

The second mechanism was quieter and more dangerous, because it lived inside people’s heads. Westad calls it the conflation of grievances. By 1914 the great powers had stopped treating their disagreements as separate problems. Economic rivalry, naval competition, colonial friction, a diplomatic slight in the Balkans — these were no longer distinct files to be managed one at a time. They had merged into a single, total perception: the others are out to destroy us. Once a leader believes that, every individual act by a rival — a ship launched, a railway financed, a treaty signed — stops being an event to be assessed and becomes one more piece of evidence for a conclusion already reached. Nothing can be reassuring anymore, because the framework no longer admits reassurance. This is the structure. Not the alliances. The habit of mind that reads the whole world as proof of a threat.

Here is why this is not history. The claim — and it is a measured one, not a prophecy — is that both mechanisms are assembling again, and the second one is further along than most people think. In Washington and Beijing today, almost no one is doing the patient work of disaggregation. China’s naval build-up, its industrial subsidies, its technology transfers, its position in global supply chains, its conduct toward Taiwan — in the American framework these have fused into a single intention. In the Chinese framework, American tariffs, export controls, alliance-building, and support for Taiwan have fused into a single intention of their own: containment, the deliberate strangling of a rising power. Each capital now reads every act by the other as confirmation. Neither is disaggregating. That is the 1914 habit of mind, and it is not a metaphor — it is the same cognitive structure, observable in the same kind of language.

And the cult of the offensive has a modern descendant too. It is not railway timetables now. It is the compression that comes from missiles with short flight times, from cyber-weapons that must be used before they are patched, from a Taiwan contingency that every war game treats as a race to be won in its opening hours. The technology is unrecognisable. The logic is identical: whoever waits, loses. Which means the structure once again rewards the leader who moves first and punishes the one who pauses to think.

The uncomfortable part is what this does to the comforting belief that war requires a villain. It does not. It requires only that enough capable, frightened people, each behaving rationally inside a structure that compresses their time and merges their fears, make a sequence of locally reasonable decisions. The catastrophe is the sum. No one at the top has to choose it. They only have to fail, separately, to stop it.

How the Machine Runs

So how does a structure actually produce a war? It is worth being precise, because the mechanism has parts, and the parts run in sequence.

The first part is the loaded condition. Over a generation, the fears accumulate and merge — economic anxiety, nationalism, strategic suspicion — until the major powers no longer see a set of manageable problems but a single adversary bent on their ruin. This is the conflation. At the same time, the military planning compresses: every general staff builds its strategy around moving first, so the window for diplomacy in a crisis shrinks toward zero. Neither of these, by itself, is a war. They are the equivalent of a room slowly filling with gas. The conditions under which a small event becomes a large one are now in place.

The second part is the trigger. The loaded condition does not discharge on its own — it waits for an acute crisis. In 1914 the crisis was a murder in Sarajevo: a single act of terror, in a contested province, that under other conditions might have been absorbed. But it did not land under other conditions. It landed in a room full of gas. And here the structure does its quiet work: the conflation of grievances means the crisis is not read as an isolated event but as confirmation. To Austria, Sarajevo proved Serbia was an existential threat. To Germany, backing Austria proved its alliance was worth something. To Russia, Austria’s response proved a Germanic war of conquest was beginning. Each capital fitted the same event into a story it had already written.

The third part is the compression. Once the crisis is read as existential, the military timetables take over from the diplomats. Russia cannot mobilise slowly, because slow mobilisation means defeat. Germany cannot let Russia mobilise first, because the entire Schlieffen Plan depends on hitting France before Russia is ready. Each clock, once started, forces the next. The book’s account of late July 1914 is almost unbearable to read precisely because you can watch the time for decision evaporating — the Kaiser, too late, trying to walk it back; the Tsar begging his cousin by telegram; the generals informing their monarchs that events are now beyond their control.

And this is the fourth part, the one that matters most and is most often left out. The structure does not pull the trigger. People do. The account is exact about this: the war came from concrete decisions — Austria’s choice to attack, Russia’s choice to fully mobilise, Britain’s choice to stay ambiguous a few days too long, and, in the last instance, Germany’s willingness to risk the war and then to declare it. The structure did not abolish the choices. It did something more insidious. It arranged the situation so that at each decision point, the locally rational choice — protect the ally, don’t fall behind in mobilising, preserve flexibility — pushed toward catastrophe, and the choice that would have stopped it required a leader to accept what looked, in the moment, like humiliation or defeat. The gas and the compressed timetable did not light the match. They ensured that an ordinary person, holding the match, reaching for the most reasonable-seeming option in front of him, would light it without meaning to.

That is the full mechanism. Loaded condition, trigger, compression, decision. Remove any one part and 1914 does not happen. Which is also the only good news the structure offers: because the war still runs through human decisions, it can still, in principle, be stopped by them.

The Same Machine, Now

Bring the mechanism forward, then, and lay it against the present. Not to predict — the book does not predict, and neither should this essay — but to see which parts are already in place.

The loaded condition is in place. The conflation of grievances, the merging of every separate problem into one total perception of hostile intent, is not a future risk between the United States and China. It is the current operating condition of the relationship. In Washington, China’s industrial subsidies, its naval expansion, its technology acquisition, its supply-chain position, and its conduct toward Taiwan are no longer assessed as distinct policy questions. They are read as a single project. In Beijing, American tariffs, export controls, alliance-building in the Pacific, and arms sales to Taiwan are likewise read as a single project — containment, the deliberate prevention of China’s rise. Each capital has written its story. Each now reads the other’s every act as another line in it. This is the 1914 habit of mind, and it does not need to be manufactured. It already exists.

The compression is in place too. The cult of the offensive has a twenty-first-century form: missiles whose flight times are measured in minutes, cyber-capabilities that lose their value the moment they are disclosed, and a Taiwan contingency that, by the book’s account, the People’s Liberation Army is built to resolve in its opening two to three weeks before outside help can arrive. The premium on striking first, on not being the side still deciding when the other side moves, is structurally identical to the railway timetables of 1914. The technology would be unrecognisable to the men who drew up the Schlieffen Plan. The logic would not.

What is missing — and this is the essay’s one piece of genuine reassurance — is the trigger and the decisions. No crisis has fired. And the human choices that 1914 shows to be decisive have not been made. That is the whole space in which catastrophe can still be averted. But it is worth being honest about how that space is shaped. The book’s account of the Taiwan question is deliberately unsettling: it is, in one phrase, Alsace, Bosnia, and Belgium rolled into one — contested territory, a question of national identity, and a great-power tripwire, all in a single place. And the structure around it carries the same dangerous feature the book identifies in 1914: the belief, in Beijing, that time may be running out, that the moment of maximum relative advantage is now and may not return. A power that believes its position can only worsen is a power with a structural reason to act sooner rather than later. That belief was in the room in Berlin in 1914. A version of it is in the room now.

Here the distinction has to be stated plainly, because it is the moral centre of the whole argument. That a war over Taiwan is foreseeable does not make it acceptable. That the structure makes it more probable does not make it desirable, or excusable, or in any sense permitted. The book is a warning, and a warning is the opposite of a prophecy: it is offered precisely so that the thing described does not happen. The point of seeing the machine clearly is not to conclude that it must run. It is to locate the places where a hand can still be put on it — and there are such places, because the mechanism still passes, as it did in 1914, through human decisions that are not yet made.

There is one difference between the two eras that cuts both ways. After 1914, the world built institutions — designed, explicitly, to give great powers somewhere to take a crisis other than the battlefield. That apparatus still exists. But the book records something corrosive happening to it: the most powerful state in the system has grown unpredictable about its own commitments, scaling back its participation, treating its alliances as transactional, leaving both rivals and friends unsure what it will actually do in a crisis. In 1914, Britain’s fatal contribution was ambiguity — no one knew, until too late, what London would do. A century later, the structure has rebuilt that same ambiguity, and placed it where the old certainty used to be.

The Objection Worth Taking Seriously

An argument this clean invites a serious objection, and it should be met head-on rather than waved past.

The most powerful version comes from the historian Christopher Clark, whose account of 1914 in The Sleepwalkers has shaped how a generation reads the July crisis. Clark’s emphasis falls precisely where this essay’s does not. To him, the road to 1914 was a chain of particular, contingent, avoidable moments — a sequence so specific to its personalities, its misread telegrams, its accidents of timing, that the more honestly you reconstruct it, the less it looks like a structure and the more it looks like a run of bad luck that could easily have broken the other way. From that vantage, the structural reading carries a real danger. If you tell people 1914 was a machine, you teach them to look at their own moment and recognise a machine — and a machine is a thing that runs whether or not anyone acts. The analogy itself can become fatalism. It can persuade leaders that the catastrophe is already decided, which is the one belief most likely to make them stop trying to prevent it.

This objection has force, and it should not be dismissed — though it is worth noting that Clark’s own contingency-first reading is itself contested, with critics arguing it relaxes the question of responsibility too far. But notice what the objection actually disputes. It does not dispute that the loaded conditions exist; Clark’s own narrative is full of them. It disputes the weight the analogy should bear — whether pattern or contingency deserves the emphasis. And on that narrower question, the structural argument is closer to Clark than a quick reading suggests. Westad does not claim the war was inevitable. The book insists, repeatedly, that it ran through concrete decisions, that better choices were available, that deterrence and patience and an answered telephone could have changed the outcome. The structure, in this account, never abolishes contingency — it loads it. It tilts the table so that the contingent moments are more likely to fall one way than the other. The disagreement with Clark, then, is not structure versus accident. It is whether, knowing the table is tilted, you are better served by studying the tilt or by studying the dice. The wager of this essay is that the tilt is the part you can still do something about.

What This Means Now

Step back from Taiwan, from China, from the particular crisis, and look at what the argument actually claims. It is not a forecast. It is a description of a recurring shape.

The shape is this. A long period of relative great-power calm rests on a particular distribution of power. That distribution begins to shift — not because anyone wills it, but because economies grow at different rates, because technology disperses, because the slow tectonics of population and production move underneath the politics. The shift is real and it is not reversible by anyone’s preference. And then the danger arrives, not from the shift itself, but from how the powers respond to it: whether they treat the adjustment as something to be managed over time, or as an emergency to be settled at once.

This is the pattern worth holding onto, because it reframes what the present moment is. The rise of China, the relative decline of the United States, the return of a world with several centres rather than one — these are not the catastrophe. They are the weather. The book makes the point with a single, devastating observation: Germany, had it never started the First World War, would very likely be the leading power in Europe today regardless — which, within the European Union, many would say it now is. It arrived at that position anyway. It simply took two world wars and a century of ruin to get there, when the underlying economics would have delivered something close to the same outcome without a shot fired. The structure was going to move. The only open question was whether it would be allowed to move slowly, through a hundred small adjustments, or whether someone would try to resolve it in a single afternoon.

That is the real lesson, and it is not a comfortable one, because it places the danger not in the rivalry but in the impatience. The conflation of grievances is, at bottom, a failure of patience — a refusal to keep separate problems separate, to let an adjustment take the time an adjustment needs. The cult of the offensive is impatience hardened into doctrine — the conviction that waiting is the same as losing. A structure produces catastrophe when the people inside it become convinced that time is against them and that the moment to act is now.

And this is where the argument reaches the reader, because the habit it describes is not only a property of states. The instinct to fuse every piece of bad news into a single proof of hostile intent, to read the worst into each separate act, to feel that the window for acting is always closing — that is a recognisably human pattern, visible well below the level of foreign ministries. The value of seeing it operate in 1914, at full and terrible scale, is that it becomes easier to recognise at smaller scales, including the ones a reader will meet this week. The discipline the essay is really asking for is the discipline of disaggregation: the willingness to look at a frightening situation and insist on separating what is actually happening from the total story the fear wants to tell.

The Thread This Continues

The argument here does not stand alone. It connects, directly, to a question this publication has been circling for some time.

In The Avoidable War, the essay traced how a US–China collision could be produced by structure rather than by malice — how rivalry has a logic that runs ahead of the intentions of the people inside it. The Coming Storm supplies the missing mechanism. It explains how that structural collision actually converts into a war: through the conflation of grievances, the compression of decision time, and the trigger that finds a room already loaded. Read together, the two essays make a single claim from two directions. The Avoidable War showed that the war is not willed. This essay shows the machine that could produce it unwilled.

There is a second thread. The argument about sanctions and economic coercion traced in the Mulder essay, The Weapon That Builds Its Own Targets, described a tool that works by friction — by sustained pressure rather than decisive blows. The analysis of the cult of the offensive in The Coming Storm is the dark complement to that. Where economic coercion has no natural endpoint and so freezes a conflict in place, military doctrine built on striking first has the opposite pathology: it manufactures an artificial deadline, a now-or-never. A system caught between the two — frozen by sanctions, hurried by the offensive — has no comfortable speed at all.

Ending

The powers did not choose 1914. They chose, separately, a hundred smaller things — and the war was the sum.

The machine is not destiny. It has no will of its own.

It only needs people who are frightened, and certain, and out of time.

Two of those, we already are.

The third is still a choice.

---

If you are trying to understand the structure beneath the noise, rather than the noise itself — subscribe.

---

Further reading, for those following this thread:

Christopher Clark, The Sleepwalkers — the contingency reading, against which this essay’s structural argument should be tested.

Graham Allison, Destined for War — the rising-power and ruling-power frame, in its bluntest form.

Odd Arne Westad, The Coming Storm — the book this essay reads.