You already know why first-principles thinking is worth doing: reasoning from bedrock truths instead of borrowed conclusions. This lesson is the how. We’re going to take a vague, intimidating idea — “decompose the problem to its fundamental parts” — and turn it into a concrete, repeatable move you can run on any stuck plan, any inflated budget, any “that’s just how it works.” The move has a name we’ll borrow from a factory floor, an enemy we’ll learn to spot in disguise, and a stopping point so you don’t dig forever. Bring a shovel.
Before we hand you the tool, a quick gut check.
Before you read — take a guess
You're told 'we can't ship the feature without a 6-week QA cycle — that's our process.' What's the most first-principles next move?
The “why” ladder: dig until you hit rock
Here’s the whole technique in one sentence: take a claim, ask “why is that true?”, and keep asking it of each answer until you hit something you genuinely cannot reduce further — a bedrock fact. A bedrock fact is something the universe enforces, not something a human decided: a law of physics, a verified number, a real human need. That’s where you stop digging, because there’s nothing underneath.
The analogy is peeling an onion — except instead of layers of vegetable, you’re peeling off layers of because-someone-said-so. Each “why” strips one layer. Or, if you prefer a louder metaphor: you’re digging a hole, and most of what you dig through is loose soil (assumptions). You keep digging until the shovel hits rock. The rock is bedrock. You can’t dig the rock.
This has a famous cousin in the manufacturing world: the “5 Whys.” Invented at Toyota, the 5 Whys says that to find the root cause of a problem (not just the symptom you can see), you ask “why?” about five times — each answer becomes the thing you interrogate next. Five isn’t magic; it’s a rule of thumb. Sometimes you hit bedrock in three whys, sometimes seven. The point is to keep going past the first satisfying-sounding answer.
Symptom vs. root cause
The 5 Whys was built to fix machines, not minds — but the logic is identical. “The robot stopped” (symptom) → why? → “a fuse blew” → why? → “the bearing seized” → why? → “no lubrication” → why? → “the oil pump filter clogged” → why? → “no filter on the intake.” That is the root cause. Replace the fuse and you’re back here next week. Fix the intake and the problem’s gone for good.
A quick ladder, rung by rung
| Rung | Claim | ”Why is that true?“ |
|---|---|---|
| 0 | ”We have to print these reports in color.” | Why? |
| 1 | ”Because the charts need to be readable.” | Why does color make them readable? |
| 2 | ”Because we use red and green to show pass/fail.” | Why red/green specifically? |
| 3 | ”Because that’s the template we inherited.” | ← inherited, not bedrock |
| Bedrock | The reader needs to tell pass from fail at a glance. | (a real human need — stop) |
Notice that the real requirement — “distinguish pass from fail” — can be met with shapes, labels, or bold text. Color was an assumption riding on a genuine need. The ladder separated the two.
Assumptions vs. facts: the whole game
This is the heart of the method, so it gets its own section and your full attention. Every why-ladder is really a sorting machine that splits the world into two piles:
- A fact is something true about the world that you can verify independently of who told you. Water boils at 100°C at sea level. This alloy melts at 1085°C. Our top customer churned last month. You could go check.
- An assumption is something you’re treating as true — usually inherited, usually unexamined, often correct by accident but never verified. “A startup needs an office.” “Reports are 40 pages.” “Customers won’t buy this online.” Nobody handed you a proof; you absorbed it.
The entire skill of first-principles thinking comes down to one thing: catching assumptions that are wearing a fact costume. Assumptions are sneaky precisely because they feel exactly as solid as facts. They’ve been sitting in your head so long they’ve grown a beard of legitimacy. Your job is to frisk every “fact” at the door.
The two-question frisk
Hold any claim up to the light and ask:
- How do I actually know this is true? (Can I point to evidence, or just to a source?)
- Could it be otherwise? (Is there a world where this is false and nothing breaks?)
If the honest answer is “because that’s how it’s always been done” or “everyone knows that” — congratulations, you’ve found an assumption wearing a fact costume. A real fact survives both questions.
Let’s put it to work.
Frisk each statement. Is it a verifiable fact, or an inherited assumption?
Place each item in the right group.
- Customers in our market won't buy online
- A startup needs an office to be taken seriously
- A 'real' report has to be at least 40 pages
- Water boils at 100°C at sea level
- A round trip from A to B and back covers twice the one-way distance
- This metal melts at 1085°C
Worked example: “I can’t start a business”
Time to run a full ladder on a real, knotted-up belief. Picture a friend who says:
“I can’t start a business. I’d need an office, staff, and a big loan. I just don’t have that.”
That sentence sounds like a sober list of facts. It’s actually three assumptions in a trench coat. Let’s frisk each one with the why-ladder.
Claim: I need an office. Why? → “To meet clients and look legitimate.” Why does looking legitimate require a leased office? → “…because that’s what real businesses have.” Could it be otherwise? → You could meet clients at their site, over video, or at a café. “Looks legitimate” is delivered by a clean website and showing up prepared, not by square footage.
Bedrock fact left standing: You need a way to meet clients. (Cost: roughly $0.)
Claim: I need staff. Why? → “There’s too much work for one person.” Which task, right now, actually needs a second human? → For an idea you haven’t validated yet… none. Selling, building a first version, and talking to customers can be one person’s job at the start. Could it be otherwise? → Hire when a specific task is both essential and provably too big for you — not before.
Bedrock fact left standing: Some tasks will eventually exceed one person. (Cost: $0 today; a hire later, on evidence.)
Claim: I need a big loan. Why? → “To pay for the office, the staff, and the inventory.” But we just dissolved the office and the staff — so what’s the loan actually for? → “To… test whether anyone wants this?” What’s the true minimum to test the idea? → A landing page, a few conversations, maybe a small batch of the product. Hundreds, not hundreds of thousands.
Bedrock fact left standing: You need enough money to test demand. (Cost: small, possibly your own.)
Look at what happened. We started with a wall — office + staff + big loan — that made “start a business” feel impossible. After the ladder, the wall collapsed into three modest, real constraints:
| Original “fact” (assumption) | What the ladder left (bedrock need) | Cost |
|---|---|---|
| Need an office | Need a way to meet clients | ~$0 |
| Need staff | Some tasks exceed one person eventually | $0 now |
| Need a big loan | Need enough to test demand | Small |
The pile of “I need X” items didn’t shrink because we got braver or more reckless. It shrank because most of the pile was never load-bearing. It was inherited furniture, and the ladder carried it out.
Mid-ladder, your friend says: 'Okay, fine, but I'd still need inventory to sell anything.' What's the best next why-question?
A second example: the 40-page report
The business example was a big knot. Here’s a smaller, everyday one to show the move is the same at any scale.
“This report has to be 40 pages. Our reports are always 40 pages.”
Why 40 pages? → “Because that’s the length of a proper report.” Why does ‘proper’ mean long? → “…because a thin report looks lazy.” Who is this report for, and what do they do with it? → “The director reads it to decide whether to fund the project.” What does the director actually need in order to decide? → “The expected return, the cost, the main risk, and a recommendation.”
Bedrock fact: the director needs three numbers and a recommendation to act. Everything past that is padding wearing the costume of “thoroughness.” The real report might be one page — and better, because the director can act faster.
Complete the test that tells a fact from an assumption:
Pick the right option for each blank, then check.
If the only justification you can find for a claim is 'because ,' you've found an , not a fact — and it's fair game for the why-ladder.
Don’t dig through the floor
The why-ladder is powerful, which means it’s easy to over-swing. Two failure modes to watch for.
Pitfall 1: “I can’t explain it” ≠ “it’s baseless”
Just because you can’t immediately explain why a rule exists does not mean the rule is a baseless assumption. Some conventions are dumb inheritance — but others quietly encode hard-won, expensive lessons. The guardrail on the mountain road has no sign explaining the fatal crash that put it there; remove it because “I don’t see why it’s here” and you may rediscover the reason the hard way. (There’s a whole mental model about this trap — Chesterton’s fence — and we devote lesson 5 to it.) The fix: when a rule resists your why-ladder, that’s a cue to investigate its origin, not a license to bulldoze it.
Pitfall 2: digging past bedrock into paralysis
The ladder is supposed to end. When you hit a law of physics, a verified number, or a genuine human need, you’ve struck rock — stop and build. If you keep asking “but why is that true?” forever (“why does the customer need to decide?” → “why does the project need funding?” → “why does the company exist?” → “why does anything exist?”), you haven’t found deeper insight; you’ve found analysis paralysis with a philosophy degree. Bedrock is a feature, not a failure. Plant your flag there.
Two ways the ladder breaks
Too shallow: you stop at the first answer that sounds reasonable and miss the assumption underneath. Too deep: you blow past genuine bedrock and dissolve every constraint into existential fog. Aim for the middle: dig until you hit something the universe (not a human) enforces, then stop.
You're running a why-ladder on a safety checklist step nobody can explain. What's the wisest move?
When to use the why-ladder
This isn’t a hammer for every nail. Reach for it specifically when:
- A plan is bloated with “we need X” items. Each “need” is a candidate assumption. Frisk them; watch most of the pile collapse.
- You’re stuck. Being stuck often means you’ve accepted a false constraint. The ladder finds the constraint that isn’t real.
- A constraint feels inherited rather than physical. “That’s the process,” “that’s the standard,” “that’s what people expect” — these are flares marking buried assumptions. A constraint backed by a law of physics or a verified number doesn’t need the why-ladder; it is bedrock already.
Conversely, don’t run it on decisions where the cost of digging exceeds the prize, or on bedrock you’ve already verified. Frisking the boiling point of water for the hundredth time is not insight; it’s a hobby.
Match each term to its definition.
Pick a term, then click its definition.
The whole method on one page
Big picture
The why-ladder, at a glance
- Why-Ladder
- The move
- Ask "why is that true?"
- Repeat on each answer
- Stop at bedrock
- Sort into two piles
- Fact = verifiable
- Assumption = inherited
- The frisk
- How do I know this?
- Could it be otherwise?
- When to dig
- Bloated "we need X" plan
- Stuck
- Constraint feels inherited
- When to stop
- Hit a physical law / verified number / real need
- Unexplained ≠ baseless (investigate)
- The move
Climb the ladder: a quick check
In your own words, what is the why-ladder's core move?
Check your answer to continue.
Where this goes next
You now have the tool: take any belief, frisk it with “how do I know this?” and “could it be otherwise?”, run the why-ladder until you hit rock, and rebuild from the facts that survive. It feels almost too simple — until you watch it demolish a constraint that the entire world had quietly agreed was permanent.
In lesson 4, we’ll watch exactly that happen at industrial scale. For decades, everyone “knew” what a battery pack cost — that was just the price, full stop. Then someone refused the inherited number, ran the ladder down to the raw materials a battery is physically made of, and priced those. The gap between “what it costs” and “what the atoms cost” turned out to be enormous — and the why-ladder is what opened it. Bring your shovel.