Chesterton's Fence Is the Best Argument Against Rushed Reform

Picture a country road, a gray morning and a gate. It is a plain wooden gate, hung across the road in the middle of nowhere, with a fence running off into the fields on either side. Nothing on this side seems to need it. Nothing on the far side seems to need it. A man in a hurry comes along, looks at the gate, sees no point in it, and says what any sensible modern fellow would say: "Let's clear this away." That is the opening of an argument that a British writer set down nearly a century ago, and I think it is the most useful single idea in all of conservative thought. Let me tell you what he said, and what I think it is worth.

G. K. Chesterton published a book of essays called The Thing in 1929. In it he laid out what he called a plain and simple principle about reforming institutions, as distinct from deforming them. There is a fence or gate across a road, he said. The modern type of reformer goes up to it and says, "I don't see the use of this; let us clear it away." To which the more intelligent type of reformer should reply, in effect: if you do not see the use of it, I certainly will not let you clear it away. Go away and think. Then, when you can come back and tell me that you do see the use of it, I may allow you to destroy it.

That is the whole idea. It has become known as Chesterton's Fence, and you can use it tomorrow morning.

What the principle says, and what it does not

Let me define it with care, because it is often misread.

It says: The burden of understanding comes before the burden of change. A rule, an institution or a custom that has lasted a long time probably does something, even if you cannot see what. Before you remove it, find out what it does.

It does not say: that the fence must stay forever. Chesterton was clear that after the thinking, the reformer may be allowed to destroy it. It does not say that old things are good because they are old. And it does not say that the person defending the fence must produce a perfect justification on the spot. It says the person who wants to remove it should do the work first.

This puts the principle in a different category from reflexive opposition to change. A reflex says "no" to everything new. Chesterton's rule says "not yet, and here is the homework." A reflex is lazy. The rule is demanding, and it demands the most of the reformer, who must show that he has learned something before he acts.

There is a deeper reason this makes sense, and it is the one the economist Friedrich Hayek put in a different way. Rules and customs that have survived for generations often embody knowledge that nobody wrote down. A custom that worked, and that was tested by the people who lived under it, may contain lessons about human nature and local circumstances that no committee could list. See Hayek Explained Why the Smartest Planner in the Room Always Loses for the economics, and Edmund Burke and the Case for Changing Things Slowly and Carefully for the older version of the same insight from Edmund Burke.

Example one: the sparrows of 1958

Start with a case where the fence was taken down at full speed and with enormous energy. In 1958, China's government launched what became known as the Four Pests Campaign, part of the Great Leap Forward. The pests were rats, flies, mosquitoes and sparrows. The sparrows were targeted because they ate grain seeds. Millions of people were mobilized to beat pots and drums to keep the birds in the air until they dropped from exhaustion, and to destroy nests and eggs.

The plan assumed that the sparrow's role in the field was simple: it ate grain, so a world without sparrows would have more grain. But sparrows also eat insects. With the birds gone, insect populations, including locusts, were reported to have boomed, and the Four Pests Campaign was later revised, with bedbugs replacing sparrows on the list. Most historians and scientists who have studied the period say the ecological disruption contributed to the famine that followed, though the famine had several causes, and the biggest were the policies of the Great Leap Forward itself. I would not want to overstate the sparrow's share. The lesson is narrower than that, and sturdier: someone looked at a fence, a bird in a field, and saw no use for it.

The rule of thumb is that when you remove something from a system, you remove the thing you can see and the dozen things you cannot. The man who does not know what the sparrow does is the last man who should be given the net.

Example two: the cash drawer rule

A hypothetical, to bring it home. Picture a small shop in your town, with a rule that has been on the wall for thirty years. Whenever the register is emptied at night, two employees must be present, and both must initial the slip. A new manager arrives. He does the math and notices that the rule costs about twenty minutes of paid time each evening. Nobody can remember anyone being cheated. It looks like a fence with nothing on either side.

If he removes it without asking, he may be right. The rule may be a relic of a long-forgotten problem. But suppose he does what Chesterton recommends and asks the oldest employee why it is there. The answer might be: three owners ago, a cashier was falsely accused of taking money, and the two-initial slip was the only thing that cleared her name. Or it might be: an insurance company required it. Or it might be nothing at all, in which case he removes it with a clear conscience, and he is better off for having asked.

Notice what the principle did in that case. It did not forbid the change. It cost a conversation, and the conversation produced either a reason to keep the rule or a confirmation that it could go. The cost was small, and the potential saving, avoiding a future loss that the old rule was quietly preventing, could be large. This is a rule about managing risk when you cannot see the whole system.

Example three: two chambers

Now take a fence from our own Constitution. A first-time reader sometimes thinks the design of Congress is absurd. Why two chambers? They do the same job. A bill must pass both, in identical form, which slows everything down. Surely one chamber, elected by the whole people, would be quicker and more democratic.

Here is the answer the framers gave. In Federalist No. 62, which is usually credited to James Madison, the argument for a Senate included the observation that a single numerous assembly is prone to the impulse of sudden and violent passions and to being seduced by factional leaders into intemperate and pernicious resolutions. A second chamber with a different term of office, a different size and a different method of election gives a law a second look, and gives a hasty majority time to cool. The Constitution also gives the states a seat at the table. You can read the larger design in Madison's Plan for a Government That Fights Itself on Purpose.

You are of course free to think the framers were wrong. People have argued that point since 1787. But notice that the framers had a reason, and the fence has a purpose. A modern reformer who proposes to remove the Senate should be able to explain the purpose before he proposes to destroy it. That is the standard Chesterton asks for, and it is a modest one.

The family version: the neighbor's gate

You can apply this at the family level too. Suppose you move into a house and find a latch on the inside of a closet door that serves no purpose you can see. Don't tear it out until you have asked the previous owner or noticed that the door swings into the hall and the latch stops it from hitting a baby gate. Suppose your town has an ordinance that seems pointless, say a setback rule keeping buildings a certain distance from the property line. Before you campaign against it, ask the planning office what it was meant to do. Many such rules are about fire spread, drainage or access for emergency vehicles. Some are about nothing, and some have outlived their purpose. Find out which.

Five questions to ask of any fence

Here is a short checklist, adapted for ordinary citizens facing a proposed change.

1. Who built it, and when? A rule that comes out of a specific crisis tells you what the crisis was.
2. What problem was it meant to solve? Ask for the original reason. If nobody knows, that is a finding in itself, and a reason to look in the records.
3. What did things look like before? If the rule is gone and the old problem returns, who pays?
4. What does it cost to keep, and who bears the cost? The reformer is right to ask this. Costs count.
5. Who gains if it is removed, and who loses? A proposal that is easy to write and hard to reverse deserves the most scrutiny.

The same method applies to sweeping bills. Congress sometimes passes thousands of pages in a single package, and nobody, including the members, has read all of it. Our piece on How a Bill Really Becomes Law, Including the Steps Schoolhouse Rock Skipped shows how that happens. And Ten Questions to Ask Before You Support Any New Government Program gives you a similar set of questions for new programs.

The best objection

Now the objection, stated fairly. Critics say that Chesterton's Fence is a trick for protecting the status quo. Every injustice in history has been defended by someone who said, "Let's not be hasty." The people who benefit from a rule are always glad to ask for an investigation. And someone who demands that reformers explain the purpose of every fence can postpone any reform indefinitely, since the explanation can always be said to be incomplete.

That is a real danger, and conservatives should not wave it away. Slavery was a fence in the sense that it was an old institution defended by those who profited from it, and the reply to "go away and think" was that the thinking had been done and the institution was a monstrous wrong. The same holds for the legal segregation that survived long after the end of slavery.

When the fence rightly comes down

So the question is when to take the fence down. I'd say four conditions together make the case.

1. You can state the purpose, and the purpose fails. The Eighteenth Amendment, which began national Prohibition in 1920, was a reform meant to reduce the harm from alcohol. The effort produced widespread evasion, organized crime and disrespect for the law, and it was repealed by the Twenty-first Amendment in 1933. People had studied the fence, understood it and judged it a failure.
2. The principle behind it is wrong. Slavery was ended by the Thirteenth Amendment in 1865. Plessy v. Ferguson (1896), which approved "separate but equal," was overturned in Brown v. Board of Education in 1954. These were not fences with a mysterious purpose. They were fences with a known purpose that was unjust.
3. The cost of keeping it is clear and the risks of removing it are small or reversible. A pilot program in one state is better than a national rule. A sunset clause is better than a permanent one.
4. The people who live under the rule have a say. The Constitution built the amendment process, Article V, for this reason: so that the people may change what must be changed, with deliberation.

Burke put the balance well in 1790 when he wrote that a state without the means of some change is without the means of its conservation. The conservative is not the man who never changes anything. He is the man who asks, before changing it, what it was for.

A word to our own side

The principle applies to conservatives too, and it should humble us. When a Republican majority wants to repeal a rule, restructure an agency or rewrite a program, the same question applies. What was the fence for? Who will be hurt if we are wrong? A party that has spent decades warning against the rushed schemes of the other side must hold itself to the same test. Thomas Sowell's habit of asking what happens after the policy passes is the working form of this idea; see Thomas Sowell Asks What Happens After the Policy Passes.

The bottom line

• Chesterton's Fence, from The Thing (1929): do not remove a fence until you can say why it was built. It is a rule about order of operations, not a ban on change.
• It is not reflex. The person who wants to change something has to do the homework, and the person defending it should help him find the answer.
• Use the five questions on any proposal in your town, your church, your school board or Congress: who built it, what problem, what before, what cost, who gains.
• Some fences should come down, and some of the best moments in American history were fences coming down. The test is whether the purpose is understood and has failed or was unjust.
• At your next public meeting, when someone says "this rule makes no sense," ask politely whether anyone in the room knows why it was adopted.