Sunday night, after dinner, I stumbled across a phrase I'd heard a hundred times without ever knowing where it came from: Hanlon's razor.
One lookup turned into three, because the answer kept name-dropping its cousins. By the end of the evening I had a razor, a principle, and a fence in my pocket — and the realization that they're basically the same tool wearing three different hats. If you work anywhere near systems, people, or legacy anything, you already use all of them. You just might not know their names.
Hanlon's razor: the paranoia brake
Never attribute to malice that which is adequately explained by stupidity.
The origin story is almost disappointingly casual: Robert J. Hanlon submitted the line to Arthur Bloch's Murphy's Law Book Two in 1980. A joke book. The idea itself is much older — Heinlein wrote a version in 1941, and similar lines get misattributed to Napoleon and Goethe all the time. Good ideas don't need noble births.
Why it's useful: it's a paranoia brake. Most "they did this to me" moments are really "they weren't thinking about me at all." Malice is rare and cognitively expensive. Carelessness is free and everywhere. So when you're deciding between the two, the base rates are not close.
In IT this is practically a survival skill. The outage usually isn't sabotage — it's someone who didn't read the runbook. The config wasn't cursed; it was Friday. Hanlon's razor pairs nicely with its more famous cousin, Occam's razor: Occam picks the simplest explanation for the world, Hanlon picks the simplest explanation for people's intentions.
One honest caveat: it's a default, not a verdict. A one-off screw-up gets the benefit of the doubt. A pattern of repeated harm earns the malice hypothesis eventually.
The Peter Principle: promoted to your level of incompetence
In a hierarchy, every employee tends to rise to their level of incompetence. — Laurence J. Peter, 1969.
The mechanics, in four steps:
- You're good at your job, so you get promoted
- You're good at that job, so you get promoted again
- Repeat until you land in a role you're not good at
- The promotions stop — and you stay there, underperforming, indefinitely
The punchline is structural: promotions reward skill at your current job, but the next job usually needs a different skill entirely. Given enough time, every position fills with someone who's bad at it. Peter's own summary was brutal: "In time, every post tends to be occupied by an employee who is incompetent to carry out its duties."
The classic IT version: brilliant sysadmin gets promoted, becomes a mediocre manager who misses the server room. Writing great automation and running great one-on-ones are different sports — but the org chart treats one as the prize for winning the other. It's why "Lead" in a job title is worth reading twice: sometimes it means deepest technical authority, sometimes it means meetings all day. Neither is wrong. Only one might be wrong for you.
Chesterton's fence: earn the demolition
This one comes from G.K. Chesterton's The Thing (1929). A reformer spots a fence across a road that looks pointless: "I don't see the use of this — let's clear it away." Chesterton's answer, paraphrased: if you don't see the use of it, I certainly won't let you clear it away. Go and think. When you come back and can tell me why it's there, then we can talk about removing it.
The fence isn't there by accident. Someone built it for a reason, and that reason exists whether or not you can see it. The burden of proof sits on removal, not on preservation.
Every legacy system is a field full of unexplained fences:
- The cron job that "doesn't seem to do anything" is quietly feeding someone's monthly report
- The commented-out firewall rule is commented because someone deleted the active one years ago and took the phones down for a day
- The ugly workaround deep in the codebase is load-bearing until proven otherwise
The junior asks "why is this here, it's stupid, delete it." The senior asks "why is this here?" — then decides. Same fence. Different posture.
And the caveat matters just as much: Chesterton doesn't say never remove fences. "We've always done it this way" is not a reason, either. You earn the demolition by understanding the purpose first — and if the purpose died in 2019, out comes the fence.
Why these three live in the same drawer
Laid side by side, they're a little decision kit:
- Someone did something that hurt you → reach for the razor first
- You're wondering why the org keeps promoting good people into bad fits → that's the principle
- You're about to delete something old and unexplained → check the fence
Each also fails in its own characteristic way. The razor can excuse real harm if you apply it to patterns. The principle can become an excuse to never promote anyone. The fence can paralyze reform entirely. A mental model isn't an answer — it's a pointer to the question you should ask first, and every pointer can be followed off a cliff.
That's the part I like most about the set: none of them tells you what to think. They tell you where to look. Malice or mistake? Skill mismatch or bad hire? Load-bearing or dead weight? Three questions, and the honest answer to each one starts with "let me check before I assume."
So — what's the fence in your system you've been meaning to tear down, and do you actually know why it's there?
