First Principles Thinking

Where this came from

The term and the method originate with Aristotle, who wrote in the "Posterior Analytics" that first principles are the foundational propositions from which all other knowledge in a domain derives. You cannot prove a first principle; you simply identify what is irreducibly true and use it as your starting point. Descartes applied a similar method philosophically in the 17th century with his method of radical doubt: strip away everything that can be questioned until you reach what cannot be denied, then rebuild.

In modern business, the framework is most closely associated with Elon Musk, who has described it explicitly in interviews as central to how he approaches engineering and business problems. His worked example is instructive: when evaluating whether a private rocket company was viable, the conventional wisdom was that space launch costs $65 million per flight. Musk asked instead what the constituent materials cost. The answer was approximately $2 million. Everything between $2M and $65M was cost created by convention, inefficiency, and monopoly pricing, not physical necessity. That gap was SpaceX's opportunity.

How it works

The process has three steps.

First, identify the problem you are trying to solve and list every assumption embedded in how you are currently thinking about it. Write them down explicitly. "This is how it has always been done." "This is what it costs." "This is who does this kind of work." "This is the timeline." Every assumption is a candidate for examination.

Second, challenge each assumption. The test is simple: is this assumption based on a physical, logical, or legal constraint, or is it based on convention and prior practice? Physical constraints are real. Legal constraints are real, though they can change. Conventional practice often reflects historical conditions that no longer apply. When costs were higher, when certain technology did not exist, when different regulations were in force. Separate real constraints from inherited ones.

Third, reason upward from what remains. Once you have stripped away the convention, ask: given what is genuinely true, what is the best solution I can construct? This is where creative thinking enters. The analysis is not the solution; it clears the ground for the solution.

Charlie Munger, who built a career on reasoning from fundamental principles in investment analysis, describes this as inverting the problem: instead of asking how to succeed, ask what would guarantee failure and eliminate those things. Both moves share the same logic: get back to what is actually true before deciding what to do.

When to use it — and when not to

First principles thinking is most valuable when you are entering territory where the existing playbook was written for different conditions. Starting a company in a new market. Pursuing a career path that did not exist a decade ago. Solving a technical problem that prior solutions have handled inadequately. Anywhere that the cost of carrying forward conventional assumptions is high.

It is also useful as a diagnostic for stalled decisions. If you cannot work out why a decision feels hard, mapping out the assumptions underneath it often reveals that you are treating inherited constraints as fixed when they are not.

Where it should not be applied uncritically: safety-critical domains, regulated industries, and anywhere convention encodes hard-won failure analysis. A surgeon questioning the first principles of sterilisation protocol is not displaying intellectual courage; they are taking an unnecessary risk with accumulated institutional knowledge. The appropriate question is whether the conventional practice is based on evidence or inertia. In medicine, engineering safety, and legal procedure, it is usually evidence.

How to apply it in practice

Take a decision you are currently stuck on. Open a blank document and write the decision at the top. Then write: "Assumptions I am treating as fixed." List everything you have not explicitly chosen but are accepting as given.

Go through each item and mark it with one of three labels: Physical/Legal (this constraint is genuinely non-negotiable), Evidence-based (this constraint is supported by strong empirical record and changing it has known costs), or Conventional (this is how it has been done, but there is no strong reason it must be done this way).

Keep the first two categories. Question every item in the third. For each conventional assumption, ask: if this constraint did not exist, what would I do? That question often reveals the actual preferred path, which was obscured by inherited assumptions.

The process does not take long. Most decisions have five to ten assumptions worth examining. An hour of this kind of structured questioning often produces more clarity than weeks of deliberation under unexamined assumptions.

---