First Principles
How to assess systems
The world makes more sense when you see it as a patchwork of systems—each of which can be reverse-engineered with mental models.
A "system" is simply anything with multiple parts that depend on each other. Every machine and process is a system at some level. For example:
- A business, such as Microsoft or a startup.
- A tool, such as a rocket or a keyboard.
- A process, such as economic growth or sustaining a romantic relationship.
- A state of being, such as your health or happiness.
A mental model of a system is a reduction of how it works. The model identifies the core components that matter and how they interact. This clarity is necessary to drastically improve the system:
- When you improve the system of business, you make more money.
- When you improve the system of relationships, you gain deeper friendships.
Here's a mental model that Musk uses to reduce complex systems:
First Principles — What is the most efficient way to solve a problem if you started from scratch? If you look past humanity’s attempts to solve it, what is the best approach if you reasoned from its fundamental principles?
The SpaceX team looked past decades of incremental rocket improvements to re-examine spaceflight from scratch. They asked, What do the underlying engineering (not historical) principles reveal to be the most cost-efficient and power-efficient way to build a rocket? They worked up from those physical realities to build the world’s most efficient rocket.
They let a simple mental model—one that fits on a cue card—guide them through the deconstruction of a massive system so they could drastically improve it.
That is the leverage of models. Imagine if you looked at every part of the world this way.
Some people do.
How do you use First Principles for your down-to-Earth problems?
For every project, ask:
- What system underlies this project? Is it a relationship, a business, a product, or something else with multiple components relying on each other?
- Is this system already efficient?
- If not, what are the ironclad principles underlying it?
- Can I start over from those principles to identify a remarkably better way to design this system?
Anything that takes up a lot of your time must be investigated like this.
This brings us to our next question: How do we know which systems in our lives are worth significantly improving? To answer this, we'll use yet another mental model: one that I call One Level Higher.