Trade Was Supposed to End War. Nobody Told History.

Here is a question worth asking while there is still time to do something about the answer: what would have to be true for us to arrive in 2030 with no war between China and Japan, with Iran quietly simmering rather than boiling, and with Ukraine no worse than it is today? It is the sort of question that sounds simple, in the way that "where did the universe come from?" sounds simple, and it has roughly the same number of satisfying answers.

There is a persuasive answer doing the rounds among geopolitical analysts, and it is a historical one. To understand why the world looks the way it does, go back to 1995 and 1996, when two rival theories of how to make peace collided in the Middle East. One of them lost. We are now living with the consequences, the argument goes, and it is too late to put them back in the box.

I think that argument is right about more than is comfortable, wrong about some important details, and most interesting of all on the point that gets the least attention. Let us take it in order.

 

Theory One: Sell Them an Air Conditioner

In the mid-1990s, Washington ran on a simple and rather lovely idea. If people trade with each other, and if consumers are busy buying things, nobody has any reason to go to war. Prosperity is the cure for hatred. The usual exhibits are Germany and Japan, two of the most militarised states in history, which were allowed to trade, given access to air conditioning and large cars, and turned into some of the most peaceful nations on Earth. If it worked for them, why not for Israel and the Palestinians?

This is often called neoliberalism, or the Washington Consensus. Strictly speaking, the Washington Consensus was something narrower: a list of ten economic policy prescriptions, written down by the economist John Williamson in 1989, covering things like fiscal discipline, trade liberalisation, and privatisation, mostly aimed at Latin America. It was a recipe for balance sheets, not for peace treaties. But the label points at something real, which is the mood of the era. The Cold War had ended. Francis Fukuyama had announced the end of history. Markets were winning, democracy was spreading, and a great many clever people believed that the rest was just a matter of logistics.

The idea itself is much older than the 1990s, and it has an awkward track record. Montesquieu wrote about the gentle effect of commerce in the eighteenth century. In 1910 Norman Angell published The Great Illusion, arguing that the economies of the great European powers were now so intertwined that war between them had become irrational and therefore, in practice, obsolete. The book was a bestseller. Four years later, the great European powers went to war with each other for four years. History does not, as a rule, read the bestseller lists.

In 1996 the journalist Thomas Friedman offered a cheerful modern variant: no two countries with McDonald's restaurants had ever fought a war against each other. The theory survived until roughly the moment it was tested. Russia and Georgia both had McDonald's in 2008. Russia and Ukraine both had them in 2014, and again in 2022. Germany built an entire foreign policy around a similar idea, Wandel durch Handel, change through trade, and expressed it most permanently in the form of two very large gas pipelines under the Baltic Sea. We all know how that ended, and so do the pipelines.

The Germany and Japan example is also more complicated than it first appears. Both countries became peaceful after total military defeat, years of occupation, new constitutions written under supervision, and the permanent presence of American troops and an American security guarantee. The Marshall Plan and open markets mattered enormously. But trade was the reward for a settled peace, not the thing that created it. Selling someone an air conditioner works best after the shooting has stopped and someone else is guaranteeing it will not start again.

 

Oslo: The Theory Meets the Real World

Applied to the Middle East, the theory became the Oslo process. The 1993 handshake between Yitzhak Rabin and Yasser Arafat on the White House lawn, with Bill Clinton beaming between them, was supposed to begin a sequence: a two-state solution, then trade, then integration into the regional economy, then prosperity, and finally the slow evaporation of hatred as everyone became too busy and too comfortable to bother with it.

It was an engineering approach to peace, and like a lot of engineering projects it had a critical path with no slack in it. In November 1995 Rabin was assassinated by an Israeli extremist who opposed the process. In early 1996 a wave of Hamas suicide bombings killed dozens of Israelis. In May 1996 Benjamin Netanyahu won the election by less than one percentage point, running against the man who had inherited Rabin's project. The critical path did not merely slip. It was struck by lightning, twice, from opposite directions.

 

Theory Two: A Clean Break

This is where the story turns. In 1996 a study group convened by the Institute for Advanced Strategic and Political Studies, led by the American strategist Richard Perle and including Douglas Feith and David Wurmser, produced a paper for the incoming prime minister. It was titled A Clean Break: A New Strategy for Securing the Realm, and the title was not modest.

Its core recommendations were roughly these:

• Abandon "land for peace." Replace the Oslo bargain, in which Israel traded territory for security, with "peace for peace" and peace through strength.
• Reshape the region. Work with Turkey and Jordan to contain, weaken, and roll back Syria, and treat the removal of Saddam Hussein from power in Iraq as a strategic objective in its own right.
• Prefer pre-emption to retaliation. Stop absorbing blows and responding to them, and instead act first against threats before they mature.

 

The paper is often summarised in far more theological and much harsher language than it actually uses. Read in the original, it is written mostly in the dry dialect of balance-of-power strategy, the native tongue of the think tank, in which nothing is ever destroyed but merely "rolled back," and the word "realm" is doing a great deal of quiet work. The three pillars, though, are clear enough, and the claim now made about them is striking: every one of them has now been achieved.

 

The Scorecard

On the face of it, the claim holds up. The Oslo process is not so much stalled as embalmed. Saddam Hussein was removed in 2003, by an American invasion that several of the paper's authors helped shape from senior positions: Feith became Under Secretary of Defense for Policy, and Perle chaired the Defense Policy Board. Syria spent more than a decade in civil war before the Assad regime finally collapsed in December 2024. Hezbollah was badly degraded in 2024. And pre-emption, as a doctrine, has been applied repeatedly, up to and including direct strikes on Iran.

So: three pillars, three ticks. A memo that reads, thirty years later, like a project plan with every milestone marked green.

Before we hand out the prize for prophecy, though, two cautions.

The first is that think tanks produce policy papers the way fig trees produce figs: abundantly, seasonally, and mostly unnoticed. Somewhere in the archives are hundreds of confident 1996 strategy documents recommending things that never happened, and nobody quotes those. When one paper appears to have predicted the future, we should ask how many others predicted different futures and were quietly composted. I have written before about thinking in probabilities, and this is exactly the trap: an impressive hit means rather less when you cannot see the misses.

The second is that the outcomes on the scorecard had a great many other causes. The Iraq war was impossible without September 11th. Syria's collapse ran through the Arab Spring, Russian intervention, Iranian overreach, and a regime that was rotten on its own account. Hamas's choices, from the bombings of 1996 to the attack of October 7th 2023, shaped Israeli politics at least as much as any American strategist did. A memo is better read as a window onto a worldview that went on to win a lot of arguments than as a blueprint that history dutifully followed. Ideas matter enormously, especially when their authors end up in the rooms where ideas become policy. But history is not a waterfall project, however much it sometimes resembles one in hindsight.

And there is one pillar of the paper that almost nobody mentions, because it did not happen. A Clean Break also argued that Israel should become more self-reliant, reform its economy, and reduce its dependence on American aid. That is precisely the part of the plan that failed, and it matters, because it is the hinge of the counterfactual.

 

The Lever Nobody Pulled

What would have had to happen for the world to look different? The most direct answer is that the United States should have pressured Israel into making peace with its neighbours. Israel depends on America for technology, finance, and arms, and is, cumulatively, the largest recipient of American foreign aid since the Second World War. That is leverage, and leverage unused is just a very expensive paperweight.

This is not a fantasy. It has been done. In 1991 President George H. W. Bush and his Secretary of State James Baker withheld ten billion dollars in loan guarantees from Israel over settlement building. It was a bruising, very public fight. The following year the Israeli government changed, and the Oslo channel opened soon after. Whatever one thinks of the policy, it demonstrated that the lever exists and that it moves things.

The reason it was not pulled again is mostly domestic. American support for Israel is woven deep into Congressional politics, and no president since has judged the fight worth having at that scale. It is also worth saying, because it is often left out, that peace has two sides to the table, and the Palestinian leadership's own failures, from the collapse of Camp David in 2000 to the second intifada, did their part to convince Israeli voters that strength was safer than compromise. A lever can move a negotiation. It cannot manufacture a counterpart.

 

"You Can't Put the Genie Back in the Bottle"

The most sobering claim is that it is now too late. The structural forces have been unleashed. The global trade system that underwrote the whole peaceful-commerce project is breaking down, and once it breaks, it is every nation for itself. The conditions already exist, the argument goes, for Japan and China to go to war, and for Germany and Russia to go to war.

The direction of travel is hard to argue with. Tariffs are back in fashion, after eighty years of being regarded as something your grandparents did. Export controls on chips, sanctions on energy, and the polite euphemism of "friend-shoring" have all turned trade from a peace mechanism into a weapons system. Germany has torn up decades of post-war restraint, created a hundred-billion-euro special fund for its armed forces, and exempted defence spending from its own constitutional debt brake. Japan is doubling its defence budget and acquiring long-range strike capability it spent most of the post-war era carefully avoiding.

Where I would push back is on the word "collapse." Global trade volumes remain close to record levels. What is happening is a fracturing, a re-routing of supply chains along political lines, which is dangerous and expensive but not the same thing as the lights going out. And "too late" is a claim about the future made with the confidence usually reserved for the past. Genies, in my experience, are notoriously poor at staying in bottles, but they are also not famous for following predictions.

 

A Systems View: Single Points of Failure

I have spent my career designing systems that are expected to stay up when things go wrong, and the lesson of this story, seen through that lens, is almost embarrassingly familiar.

In security architecture, nobody sensible relies on a single control. You do not protect a network with just a firewall, or just passwords, or just a very stern policy document. You build layers, so that when one fails, and one always fails, the others hold. We call it defence in depth.

Both of the theories in this story were single-control architectures. The 1990s consensus treated trade as the entire security stack: get the commerce flowing and everything else follows. The Clean Break treated military strength as the entire stack: dominate the region and the rest becomes irrelevant. Each ignored what the other got right, and each ignored the layers neither of them cared much about: legitimate institutions, a political settlement people can live with, justice that is felt rather than declared, and the dull, unglamorous, essential work of building trust between people who have every historical reason not to extend it.

 

Back to 2030

So what would have to be true for us to reach 2030 without the catastrophes on that list? Not, I think, a return to 1995. That world believed too much in one mechanism, and its optimism was part of what made it fragile. Nor a doubling down on 1996, which believed too much in another.

What would have to be true is that the people with leverage use it, on every side, and that someone, somewhere, starts building the layers that neither theory bothered with. It is slow work. It does not fit in a strategy paper with a grand title. It produces no satisfying moment on a White House lawn. It is, in short, exactly the kind of work that history keeps being too busy to do, right up until the moment it discovers it should have done it thirty years ago.

The pessimists may be right that the genie is out. But it is worth remembering that the bottle was never really the problem. The problem was believing that one wish would be enough.

All Posts