# First Principles Thinking: The Elon Musk Method (And Where It Originally Came From) Aristotle described the first principle as the first basis from which a thing is known. By that he meant the foundational truth that cannot be derived from something more basic, the bedrock assumption that the rest of the reasoning has to sit on. Physicists inherited the term to describe calculations done from fundamental laws rather than from empirical fitting. Philosophers used it for the axioms of a system. The phrase was dry academic vocabulary for most of its history. Then it got famous. Elon Musk, in a series of widely circulated interviews around 2013, described his approach to engineering problems as reasoning from first principles rather than by analogy. The illustration was battery costs. The industry told him rechargeable battery packs cost $600 per kilowatt-hour and would not drop much. Musk broke the battery into its constituent elements: cobalt, nickel, aluminum, carbon, polymer separators, sealed steel cans. He computed the commodity price of each element at a London Metal Exchange level and found that the material cost was roughly $80 per kWh. The remaining $520 was the combined cost of manufacturing structure, supply chain margin, and industry convention. That calculation motivated the Gigafactory strategy and, by extension, much of what made electric vehicles economically viable at scale. The interview clip made first principles thinking famous as a method for producing outsized results in industries where most participants accept received wisdom. The framing is not original to Musk. It is not even original to the twentieth century. But the popularization was consequential because it moved the concept from academic physics into the toolkit of anyone trying to find the ground under a problem that everyone claims is already solved. Expert-written and research-backed, this piece walks through what first principles thinking actually is, where it comes from, when it works, and when it does not. > "The normal way we conduct our lives is we reason by analogy. We are doing this because it is like something else that was done, or it is like what other people are doing. With first principles, you boil things down to the most fundamental truths you can imagine. Then you reason up from there." -- Elon Musk, interview, 2013 --- ## The Intellectual Genealogy Aristotles *Posterior Analytics* establishes first principles as the starting points of demonstrative knowledge. For Aristotle, a science consists of true statements derived from axioms that cannot themselves be demonstrated without circularity. The first principles are grasped through intuition trained by induction from particulars, not themselves derived. This is the classical conception that shaped the Western scientific tradition for two millennia. Descartes radicalized the approach in the 17th century. His method of systematic doubt in *Meditations on First Philosophy* (1641) set out to reach the bedrock certainty on which knowledge could be rebuilt. The famous result, *cogito ergo sum*, was intended as the unshakable first principle from which the rest of knowledge would be reconstructed. Whether or not Descartes succeeded, the procedure, doubt everything until you find what cannot be doubted, was a methodological advance that shaped modern philosophy of science. Newton used the phrase explicitly. The full title of the *Principia Mathematica* (1687) is *Philosophiae Naturalis Principia Mathematica*, the mathematical principles of natural philosophy. Newtons claim was that celestial and terrestrial mechanics could be derived from a small number of first principles (his laws of motion plus universal gravitation). The success of this program, producing accurate predictions across enormous scales from a compact axiomatic base, established first principles reasoning as the ideal form of science. Twentieth-century physics extended the approach. Quantum field theory, general relativity, and the standard model of particle physics are first principles frameworks in the strong sense: they predict experimental outcomes from small sets of mathematical postulates. The term *ab initio* in chemistry and physics refers to calculations done from first principles of quantum mechanics without empirical fitting. What Musk popularized was not the concept but its application to business and engineering decisions where the surrounding industry had substituted received convention for actual analysis. The method translates across domains when the underlying fundamentals are tractable. --- ## The Five-Step Method The practical version of first principles thinking is a specific procedure, not a vague call to think harder. The steps below are compiled from engineering methodology, scientific practice, and the recorded accounts of practitioners like Musk, Jim Keller, and Charlie Munger. **Step 1: State the problem precisely.** Not vaguely. Precisely. What specifically are you trying to achieve, at what cost, against what constraints, by when. Most people skip this step and work on the problem they imagine rather than the problem they actually have. **Step 2: List the assumptions.** Write down everything you currently believe about why the problem is hard or why the current solution is the right one. Include industry conventions, prices, timelines, material requirements, regulatory constraints, and received wisdom. The list should be long. If it is short, you are not digging deeply enough. **Step 3: Test each assumption.** For each item on the list, ask: is this true as a matter of physics, chemistry, logic, or law, or is it true as a matter of convention? Assumptions that reflect physical necessity stay. Assumptions that reflect convention become candidates for challenge. **Step 4: Rebuild from what survives.** Using only the assumptions that survived scrutiny, construct a candidate solution. This construction often looks absurd at first because it ignores conventions that the industry treats as sacred. Absurdity is not evidence of wrongness. **Step 5: Stress-test the rebuild.** Once you have a candidate solution from first principles, look for the ways it could fail. What edge cases break it? What are you missing? Talk to people who know the domain well. Invite disconfirmation. The goal is not to protect the elegance of the derivation. It is to find out whether it survives contact with reality. | Step | What It Produces | Time Required | |---|---|---| | State the problem | A specific, testable goal statement | 30 minutes to a few hours | | List assumptions | An exhaustive inventory of beliefs | Several hours, often across days | | Test each assumption | Separation of necessary from conventional | Hours per assumption for real problems | | Rebuild from survivors | A candidate solution from scratch | Days to weeks depending on complexity | | Stress-test | Evidence the solution can survive | Weeks of feedback, expert review, iteration | --- ## The Tesla Battery Example in Detail The battery example is worth working through carefully because it reveals the method in operation. **Problem statement**: Produce a battery pack for automotive use at a cost that enables competitive electric vehicle pricing. Target at the time was a significant cost reduction from the prevailing $600 per kWh. **Assumptions in the industry**: Lithium-ion battery packs cost $600 per kWh. Cost will decline at historical rates (roughly 8 percent per year). Major cost components include the cathode, anode, electrolyte, separator, and packaging. Manufacturing requires specialized facilities with existing supply chains. Cost reduction requires incremental improvements in chemistry and manufacturing. **Testing assumptions**: The $600 per kWh was real as a market price but was not a physical necessity. The commodity components, priced at London Metal Exchange spot prices, added up to roughly $80 per kWh. The gap between commodity cost and finished pack cost was roughly 7.5x, which is unusual for mature industrial products. The manufacturing step therefore contained most of the cost. That manufacturing cost reflected the structure of the existing supply chain: cells made in Asia, shipped globally, packed in small volumes. None of this was physical necessity. **Rebuild**: If the manufacturing structure contains the cost, compressing the manufacturing structure reduces the cost. Build a single facility that takes raw materials at the front and produces finished packs at the back, eliminating the intermediate shipping, warehousing, and margin layers. Scale the facility large enough that the per-unit fixed costs amortize. This is the Gigafactory concept in its first principles form. **Stress-test**: Is it actually possible to build such a facility? What does it require in terms of capital, supply chain, regulatory approval, technical expertise? How does the cost reduction scale with volume? What happens if lithium prices spike? The stress-testing of this rebuild took years of iteration but the core derivation held. The outcome: battery pack costs dropped below $150 per kWh by the early 2020s and continued downward, validating the rebuild. The industry wisdom that costs would decline 8 percent per year was wrong; costs declined much faster once the manufacturing structure was attacked. > "Most people think in analogies. When you present a truly novel problem, they reach for the closest precedent and apply it. That works when the precedent applies. It fails when the problem is actually new. First principles thinking is the discipline of refusing the analogy when it does not fit." -- Jim Keller, computer architect, interview with Lex Fridman (2020) --- ## Other Real Examples Across Domains The method is not unique to battery engineering. The pattern appears across fields. **Marshall and Warren on peptic ulcers**: In the 1980s, medical consensus held that peptic ulcers were caused by stress, diet, and excess stomach acid, treated with antacids and lifestyle changes. Barry Marshall and Robin Warren, working from first principles of microbiology, observed bacteria (Helicobacter pylori) in stomach biopsies and hypothesized a bacterial cause. Marshall famously drank a flask of H. pylori to induce gastritis in himself and demonstrate causation. Their framework was first principles: if bacteria are present and cause tissue changes, and if antibiotics eliminate bacteria and resolve ulcers, the bacterial hypothesis is correct regardless of how established the stress theory is. They received the 2005 Nobel Prize in Physiology or Medicine. Ulcer treatment shifted globally to antibiotic-based protocols. **Buffett on valuation**: Warren Buffett, trained by Benjamin Graham, reasons about stock valuation from the first principles of business cash flow rather than from market prices. A stock is worth the discounted value of the future cash flows it will produce. The market price may differ from that value for extended periods. Investing from this framework produced decades of superior returns, notably by acting when market prices deviated substantially from first-principles valuations. The method is reproducible but requires the discipline to ignore market prices as the primary signal. **Pharmaceutical pricing**: The first principles view of a medication is its cost of production, regulatory approval, patent structure, and demand elasticity. The market prices of many medications diverge from these fundamentals by factors of 10 to 100x. Some interventions, including generic substitution programs and international reference pricing, work by reducing the conventional markups to closer to first principles costs. **Fitness and nutrition**: Conventional dietary recommendations often reflect historical food industry pressures more than physiology. First principles reasoning from thermodynamics (energy balance) and endocrinology (insulin, glucagon, ghrelin, leptin signaling) often produces guidance that differs from conventional recommendations and tracks outcomes more accurately for weight loss and metabolic health. **Startup cost structures**: The conventional view of starting a software company includes dedicated office space, large teams, fundraising rounds, and specific legal structures. First principles reasoning asks what the business actually requires: code, customers, cash flow, and legal protection. Many modern software companies run on structures dramatically different from the conventional view, with outsized outcomes. Our coverage at [corpy.xyz](https://corpy.xyz/) on business formation applies first principles reasoning to the legal structure question specifically. | Domain | Convention | First Principles View | Outcome When Applied | |---|---|---|---| | Auto industry batteries | $600/kWh with 8%/year decline | Commodity-cost-plus-manufacturing at lower bound | Gigafactory, 5x cost compression | | Medicine (peptic ulcers) | Stress-induced, acid-treated | Bacterial, antibiotic-treated | Nobel Prize, treatment revolution | | Finance (valuation) | Market prices reflect value | DCF from business cash flows | Buffetts compound returns | | Space launch | $10k+/kg to orbit, fixed | Propellant cost is tiny fraction | SpaceX reusability, 10x cost drop | | Software pricing | Hourly billing or per-seat | Value-based, cost-unbundled | SaaS economics transformation | --- ## The SpaceX Case in Detail The battery example is well known. The rocket case is better because the contrast between convention and physics is starker. Conventional launch costs at the time SpaceX was founded ran around $10,000 per kilogram to low Earth orbit. The implicit assumption was that rockets are expensive because rocket engineering is hard, materials are exotic, and reliability requirements drive cost. First principles analysis: What does it actually cost to put mass in orbit? The propellant (liquid oxygen and kerosene, or later methane) for a Falcon-class launch costs in the range of hundreds of thousands of dollars, a small fraction of the vehicle cost. The vehicle itself, as a collection of materials and machined parts, has a raw material cost in the low single-digit millions. The rest is manufacturing labor, testing, and the fact that the vehicle is thrown away after one use. The rebuild: If the vehicle is the dominant cost and the vehicle is being thrown away, recovering and reusing the vehicle reduces the per-flight cost substantially. The question is whether the recovery is technically possible and economically worthwhile. The stress-testing of this rebuild occupied roughly 15 years of engineering work at SpaceX, including the dramatic failures and eventual successes of vertical booster landings. The outcome: launch costs to orbit dropped by roughly an order of magnitude with reusable Falcon 9 flights, with further reductions targeted as Starship matures. The industry that said the cost was fixed was accepting a convention that did not survive first principles analysis. ## Where First Principles Thinking Fails The method has real limitations. The popular framing often presents it as a universal solvent. It is not. **When the first principles are wrong.** Reasoning up from incorrect foundations produces confident errors. Ancient astronomy used geocentric first principles and derived internally consistent but wrong celestial predictions. Modern pseudoscience uses first principles reasoning from wrong premises (homeopathic dilution principles, for example) to reach confidently incorrect conclusions. The method amplifies whatever ground it stands on. **When the domain is irreducibly complex.** Weather, most ecological systems, financial markets at the micro level, human social dynamics at scale. These systems contain so many interacting variables with nonlinear coupling that reasoning from fundamentals does not produce useful predictions. Expert pattern-matching and statistical approaches outperform first-principles reasoning in these domains. Philip Tetlocks forecasting research documents the limits of expert analysis in complex domains. **When the solver lacks the needed domain knowledge.** First principles reasoning requires knowing which assumptions are actually fundamental. This is a domain expertise question. A layperson reasoning from first principles in medicine without understanding physiology will often challenge the wrong assumptions and rebuild on bad ground. Gary Kleins research on naturalistic decision-making shows that in well-practiced domains, expert intuition often encodes first-principles understanding that laypeople cannot access analytically. **When speed matters more than depth.** First principles reasoning takes hours to weeks to months for serious problems. In contexts requiring rapid decisions, pattern-matching from analogies outperforms reconstruction from fundamentals. A surgeon in the operating room, a fighter pilot in combat, a trader in a fast-moving market cannot derive responses from first principles. They execute rehearsed patterns. **When social coordination matters.** Some conventions persist not because they are physically necessary but because coordination around them has value. Reasoning from first principles can correctly identify the convention as arbitrary and still be wrong to abandon it if the coordination value exceeds the efficiency loss. Network effects, standards, and protocols often fall in this category. For readers evaluating when to apply which mode of thinking, the practical guidance is: use pattern-matching for familiar problems in your domain of expertise, and shift to first principles reasoning when the usual patterns do not fit or when you have reason to believe the conventional wisdom is distorted by interests, inertia, or status. Our coverage at [whats-your-iq.com](https://whats-your-iq.com/) on cognitive modes and [pass4-sure.us](https://pass4-sure.us/) on structured analytical preparation provides more depth on the pattern-matching side. --- ## First Principles and Analogical Reasoning: A False Dichotomy Musks framing puts first principles thinking in opposition to reasoning by analogy. The research literature is more nuanced. Dedre Gentners work at Northwestern on analogical reasoning shows that analogies are how humans transfer knowledge across domains and that structural analogies (mapping relational patterns rather than surface features) produce many of sciences most important advances. Kepler used the analogy of light to conceive of gravity acting at a distance. Darwin used the analogy of artificial breeding to conceive of natural selection. Bohr used the analogy of the solar system to conceive of atomic structure (later replaced but useful at the time). The scientific revolution runs on both first principles derivation and structural analogy, often in the same minds on the same problems. The more accurate framing is that good reasoners blend the two. They use analogies to generate hypotheses and first principles to test and refine them. They use first principles to identify which features of a domain are necessary and analogies to suggest how solutions from other domains might translate. The opposition between the two, while rhetorically useful for critiquing lazy analogical reasoning, misrepresents how productive thinking actually works. > "The latticework of mental models is the key. You cannot really know anything if you just remember isolated facts. If the facts do not hang together on a latticework of theory, you do not have them in usable form." -- Charlie Munger, *Poor Charlies Almanack* (2005) --- ## Practicing First Principles Thinking Like any skill, first principles thinking improves with practice. The exercises below produce meaningful gains over months. **Exercise 1: Weekly decomposition.** Pick one item of conventional wisdom you encounter in a given week. Sit with it for 30 to 60 minutes. Write down the assumptions embedded in it. Test each assumption. Sometimes the wisdom survives; sometimes it falls apart under scrutiny. The practice trains the habit of decomposition. **Exercise 2: Cost analysis.** Take something you buy and work out what it should cost from fundamentals. Materials, labor, distribution, reasonable margin. Compare to the actual price. The gap reveals the convention premium. Coffee, clothing, software, and services often have large gaps. **Exercise 3: Rebuild exercises.** Pick a field in which you have some knowledge. Imagine you had to rebuild it from scratch with no reference to current practice. What would survive? What would change? The exercise is theoretical but surfaces assumptions reliably. **Exercise 4: Devils advocate with yourself.** When you find yourself defending a position, try to construct the strongest case against it, explicitly assuming your current view is wrong. This is not first principles thinking directly but it loosens the grip of conventional assumptions enough for first principles analysis to take hold. For readers tracking this practice over time, the habit formation research applies directly. The timestamp and calendar tools at [file-converter-free.com](https://file-converter-free.com/timestamp-converter) help schedule these exercises across weeks. The writing craft needed to articulate assumptions precisely is a separate skill worth developing; our coverage at [evolang.info](https://evolang.info/) on professional writing applies. See also: [Pareto Principle: The 80/20 Rule in Real Life](/articles/ideas/decision-making/pareto-principle-80-20-rule-in-real-life) | [Feynman Technique: Learn Anything Faster](/articles/concepts/learning-science-knowledge/feynman-technique-learn-anything-faster) --- ## References 1. Musk, E. (2013). Various interviews including Kevin Roses Foundation series. Transcripts archived at multiple sources. 2. Aristotle. *Posterior Analytics*. Oxford Classical Texts. Available via https://doi.org/10.1093/oseo/instance.00262048 3. Descartes, R. (1641). *Meditations on First Philosophy*. Cambridge University Press (modern edition, 1986). 4. Newton, I. (1687). *Philosophiae Naturalis Principia Mathematica*. Royal Society. 5. Marshall, B. J., & Warren, J. R. (1984). "Unidentified Curved Bacilli in the Stomach of Patients with Gastritis and Peptic Ulceration." *The Lancet*, 323(8390), 1311-1315. https://doi.org/10.1016/S0140-6736(84)91816-6 6. Klein, G. A. (1998). *Sources of Power: How People Make Decisions*. MIT Press. 7. Gentner, D., & Markman, A. B. (1997). "Structure Mapping in Analogy and Similarity." *American Psychologist*, 52(1), 45-56. https://doi.org/10.1037/0003-066X.52.1.45 8. Munger, C. T. (2005). *Poor Charlies Almanack: The Wit and Wisdom of Charles T. Munger*. Donning Company.

Frequently Asked Questions

What is first principles thinking in simple terms?

First principles thinking is the practice of breaking a problem down to the most basic, verified components and reasoning up from those components rather than from received opinions or analogies to other cases. The term traces back to Aristotle, who called a first principle the first basis from which a thing is known. In practice, it means asking what is actually true and necessary versus what is assumed because of convention, precedent, or convenience.

How does Elon Musk actually use first principles?

The widely cited example is the Tesla battery cost analysis. Rather than accepting the industry price of around \(600 per kilowatt-hour in 2012 as fixed, Musk broke the battery into its raw material components (cobalt, nickel, aluminum, carbon, separators, sealed cans) and calculated the material cost at roughly \)80 per kWh. The gap revealed that the industry price reflected manufacturing structure and supply chain markups, not physical necessity. This motivated the Gigafactory strategy to compress the manufacturing supply chain. The method generalizes: decompose the apparent fact into its physical constituents, verify what is actually required, and question everything above that baseline.

Is first principles thinking actually better than expert intuition?

The research suggests it depends on the domain. In stable, well-understood domains where expert intuition has been refined over long careers, expert pattern-matching often beats analytical reconstruction (Gary Klein's research on naturalistic decision-making). In novel, rapidly changing, or deeply cross-disciplinary problems, first principles reasoning can find solutions that pattern-matching misses because the available patterns do not fit the new problem. The best performers typically alternate: pattern-match for speed in known domains, reason from first principles when the usual patterns fail.

What are the steps to apply first principles thinking?

The practical method has five steps. One, state the problem precisely. Two, identify and write down the assumptions you are making about why the problem is hard or the current answer is right. Three, test each assumption against verifiable evidence or physical constraints. Four, rebuild the solution from the assumptions that survived. Five, stress-test the rebuilt solution against counter-examples and edge cases before committing. The process typically takes hours for serious problems, not minutes, and benefits from writing rather than thinking silently.

What are examples of first principles thinking outside of engineering?

The framework applies broadly. In medicine, first principles reasoning from physiology led to the discovery that peptic ulcers are bacterial rather than stress-induced (Marshall and Warren, Nobel Prize 2005). In finance, Buffett's approach to valuation reasons from business fundamentals rather than market prices. In diet and exercise science, first principles from thermodynamics and endocrinology often produce better predictions than tradition-based recommendations. In public policy, cost-benefit analyses built from measurable outcomes rather than political preferences often diverge from intuition-driven approaches.

When does first principles thinking fail?

The method fails in three common conditions. First, when the first principles themselves are wrong: bad physics produces bad engineering regardless of the reasoning quality. Second, when the domain contains irreducible complexity or chaotic dynamics where small deviations from assumptions produce large errors. Third, when the solver lacks sufficient expertise in the underlying science to evaluate which assumptions are actually fundamental versus which are conventional. First principles thinking is not a substitute for domain knowledge. It works best as a layer on top of domain knowledge, not instead of it.

How is first principles thinking different from Charlie Mungers mental models?

They are complementary rather than competing. Mungers mental models approach, from his Poor Charlies Almanack, advocates applying a latticework of models from different disciplines to decompose problems. First principles thinking is one model within Mungers latticework: the physics-and-engineering mode of reducing to physical constituents. Munger also advocates analogical reasoning, inversion, probability frameworks, and discipline-specific heuristics. The best performers blend both: first principles for problems where the physical or fundamental structure is tractable, mental models broadly for problems where diverse framings reveal different facets.