Chesterton's Fence

Chesterton’s Fence is a mental model that says you should never remove something until you understand why it was put there in the first place. If you come across a fence stretched across a road and can’t see any reason for it, your first instinct might be to tear it down. But the principle insists that the fence was erected by someone with a reason, and until you understand that reason, you can’t know whether removing it will improve things or make them worse.

The idea comes from G.K. Chesterton’s 1929 book The Thing, in which he argued against the reformer who says “I don’t see the use of this; let us clear it away.” Chesterton’s response was that if you don’t see the use of it, you are precisely the wrong person to clear it away. The person qualified to remove the fence is the person who can explain why it exists - and who can therefore evaluate whether the original reason still holds.

What Chesterton’s Fence means in practice

The model isn’t an argument against change. It’s an argument against uninformed change. It draws a clear line between two kinds of reformer: the one who says “I’ve examined why this exists, the original reason no longer applies, so let’s remove it” and the one who says “I can’t see why this exists, so it must be useless.” The first is making a judgement. The second is making an assumption.

This distinction matters because complex systems - organisations, laws, institutions, codebases, social norms - are full of things that look unnecessary from the outside but serve important functions. A seemingly redundant approval process might exist because of a fraud incident ten years ago. A clunky piece of legislation might be worded that way to close a loophole that was being exploited. A cultural tradition might serve a social bonding function that isn’t obvious to someone encountering it for the first time.

The fence might be outdated. The original reason might no longer apply. But you can only make that judgement responsibly if you first do the work of understanding why it’s there. This connects to the broader principle of feedback loops - the fence may be part of a system where removing one element disrupts the balance of the whole.

Chesterton’s Fence in organisations

The new leader problem

One of the most common violations of Chesterton’s Fence happens when a new leader joins an organisation and begins restructuring before understanding the existing structure. The new leader sees processes that seem inefficient, reporting lines that seem illogical, and policies that seem unnecessarily complex. They conclude that the organisation needs modernising, and they begin dismantling.

Sometimes they are right. But often, the things they dismantle were solving problems they haven’t yet encountered. The weekly cross-departmental meeting that seemed like a waste of time was preventing the silo effect that now returns. The approval chain that seemed bureaucratic was catching errors that now slip through. The old system was imperfect, but it was imperfect in known ways. The new system introduces unknown imperfections.

This connects to second-order thinking - the practice of asking not just “what happens if I make this change?” but “what happens after that?” Removing the fence has a first-order effect (the road is now clear) and a second-order effect (whatever the fence was keeping out can now get in). Chesterton’s Fence insists that you consider the second-order effect before celebrating the first.

In software and technology

Programmers encounter Chesterton’s Fence constantly. A piece of code that looks redundant or poorly written might be handling an edge case that isn’t obvious from reading the surrounding logic. Developers who delete “unnecessary” code without understanding what it does are a reliable source of bugs that are difficult to diagnose because the failure only appears under conditions the developer didn’t anticipate.

The principle is so widely recognised in software engineering that it has become standard practice to comment code that handles non-obvious cases and to resist the urge to refactor code you don’t fully understand. “This looks wrong” is not the same as “I understand what this does and I’ve confirmed it’s wrong.”

Chesterton’s Fence in policy and politics

Reform versus demolition

Political reform regularly violates Chesterton’s Fence. When institutions are perceived as failing, the temptation is to abolish them rather than understand what they do. But institutions, like fences, usually exist for reasons. They may not be functioning well, and they may need radical reform, but “abolish it” and “reform it” are different proposals with different consequences.

The distinction between understanding and agreeing is critical here. Chesterton’s Fence doesn’t require you to endorse the original reason for the fence. It requires you to know the reason. You might discover that the fence was built for a purpose you disagree with. You might discover it was built for a purpose that no longer exists. In either case, you are now making an informed decision rather than an ignorant one.

This connects to unintended consequences - the principle that interventions in complex systems produce effects that the intervener didn’t anticipate. The more complex the system, the more likely it is that removing a component will have ripple effects that weren’t part of the plan. Understanding why the component exists is the minimum homework required before making changes.

Tradition and reform

Chesterton’s Fence is sometimes misused as an argument for pure conservatism - the position that existing arrangements should be preserved because they exist. But that’s a misreading. Chesterton wasn’t arguing that fences should never be removed. He was arguing that they shouldn’t be removed by people who don’t understand them.

This is an important distinction because it separates Chesterton’s Fence from status quo bias - the irrational preference for the current state of affairs simply because it is the current state. Status quo bias says “don’t change things because change is risky.” Chesterton’s Fence says “understand things before you change them.” One is a bias toward inaction. The other is a call for informed action.

The model is equally useful for progressives and conservatives. A progressive who understands why an institution exists and can articulate why that purpose is no longer served has a stronger case for reform than one who simply declares the institution obsolete. A conservative who understands the original purpose can make a more compelling case for preservation than one who simply appeals to tradition.

The limits of Chesterton’s Fence

The model has genuine limitations. In some cases, the original reason for the fence is deliberately hidden or has been lost entirely. Historical injustices are sometimes embedded in laws and institutions whose discriminatory origins have been forgotten or obscured. In these cases, the fact that you can’t find the reason isn’t necessarily a sign that you haven’t looked hard enough - it might be a sign that the reason was never legitimate in the first place.

There is also a practical question of how much investigation is required before you’ve “understood” the fence. In fast-moving environments, the cost of delay can be significant. A startup that applies Chesterton’s Fence too rigidly might never move fast enough to survive. The model is most valuable in contexts where the cost of getting it wrong is high and the time to investigate is available.

The first principles thinking model offers a useful complement. Where Chesterton’s Fence says “understand the existing structure before you change it,” first principles says “build your understanding from the ground up rather than inheriting assumptions.” Used together, they produce a powerful approach: first understand why things are the way they are, then evaluate whether the underlying reasoning still holds, then redesign from first principles if it doesn’t.

How to apply Chesterton’s Fence

When you encounter something that seems pointless, unnecessary, or inefficient, resist the immediate urge to remove it. Instead, ask three questions:

Why was this created? What problem was it solving, what risk was it mitigating, or what behaviour was it encouraging or preventing? If you can’t answer this question, you need to investigate before proceeding.

Does that reason still apply? Circumstances change. The problem the fence was built to solve might no longer exist. The risk it was mitigating might have been addressed by other means. If the original reason no longer applies, removal may be justified.

What will happen when it’s gone? Even if the original reason no longer applies, the fence might have acquired secondary functions. People might rely on it in ways the original builder didn’t intend. Removing it might create a gap that needs filling. Think through the consequences - including the second-order effects - before acting.

The fence might need to come down. But the person who takes it down should be the one who can explain exactly why it was there and exactly why it no longer needs to be.