Most thinking is done by analogy. We see a problem, recall a similar one, and apply a variation of the old solution. This is efficient, but it's also a recipe for incremental, not radical, improvement. First Principles Thinking is the alternative. It’s the practice of breaking a problem down to its most fundamental truths and reasoning up from there.

Imagine you want to build a car. Reasoning by analogy, you'd look at existing cars and try to make a slightly better, cheaper version. Reasoning from first principles, you ask: what is a car, fundamentally? It's a way to get from A to B. What materials are absolutely necessary for that? What's the raw cost of those materials? This line of questioning forces you to bypass assumptions and see what's truly possible, not just what's been done before.

The process involves three steps:

1. Identify Assumptions: What are the current beliefs and conventions surrounding your problem? Challenge them relentlessly. Why are things done this way?
2. Break It Down: Deconstruct the problem into its core components or fundamental truths. These are the things you know to be true beyond any doubt, like the laws of physics or the raw cost of a material.
3. Create New Solutions: From these basic building blocks, start to construct a new solution. Because you're not constrained by prior analogies, you're free to find a fundamentally better path.

Musk used this approach with SpaceX. The assumption was that rockets were incredibly expensive. By breaking a rocket down to its material components—aluminium alloys, titanium, copper, carbon fibre—he found the raw material cost was only about 2% of the typical price. The rest was manufacturing complexity and process. By rethinking the process from the ground up, SpaceX drastically cut the cost of spaceflight.

Not all problems are the same. Trying to solve a complex geopolitical crisis with the same methods you'd use to bake a cake is a recipe for disaster. The Cynefin framework is a sense-making tool that helps you diagnose the kind of problem you're facing so you can apply the right strategy.

It divides situations into five domains:

Simple: The relationship between cause and effect is obvious to all. The approach is to Sense, Categorise, Respond. You see what's happening, put it in a known category, and apply a standard best practice. Think of a help desk following a script.

Complicated: There's a clear relationship between cause and effect, but it requires expert analysis to see it. The approach is Sense, Analyse, Respond. You need specialists to diagnose the situation and determine the right course of action, known as good practice. Fixing a jet engine falls into this category.

Complex: Cause and effect can only be understood in hindsight. The system is dynamic and unpredictable. The approach is Probe, Sense, Respond. You run small, safe-to-fail experiments to see what works, and then amplify the successful ones. This is the domain of emergent practice, common in market strategy or cultural change.

Chaotic: There is no discernible relationship between cause and effect. The situation is turbulent. The only approach is to Act, Sense, Respond. You must act immediately to establish order, then sense where stability lies, and respond by moving the situation from chaotic to complex. A natural disaster is a chaotic environment.

Disorder: This is the central space where you're not sure which domain you're in. The danger here is interpreting the situation according to your personal preference, which is often counterproductive. The goal is to move into one of the other four domains as quickly as possible.

Linear thinking is simple: A causes B, which causes C. Systems thinking is different. It's a way of seeing the world as a web of interconnected elements where cause and effect are not always straightforward. A change in one part of the system can have unexpected consequences elsewhere.

At the heart of systems thinking are feedback loops—the engines of system behaviour. There are two main types:

• Reinforcing Loops (Positive Feedback): These are amplifying. A small change leads to an even bigger change in the same direction. Think of a viral video. The more people who share it, the more visible it becomes, leading to even more shares. This creates exponential growth or collapse.
• Balancing Loops (Negative Feedback): These are stabilising. They work to keep a system within a desired range. A thermostat is a classic example. When the room gets too hot, the thermostat turns the heating off. When it gets too cold, it turns it back on. Balancing loops resist change and push the system towards a goal.

By mapping these loops, you can understand why systems behave the way they do. You can find leverage points—small changes that can produce significant results—and avoid policies that have unintended negative consequences.

Our natural tendency is to think forward. We set a goal and then figure out the steps to get there. The Inversion Principle flips this on its head. Instead of asking how to achieve success, you ask: what would guarantee failure?

This mental model was popularised by Charlie Munger, who famously said, "All I want to know is where I'm going to die so I'll never go there." By identifying all the things that could go wrong—all the potential obstacles, stupid decisions, and biases—you can design a strategy to actively avoid them. It's often easier to avoid stupidity than it is to achieve brilliance.

For example, if you're launching a new product, don't just focus on the ideal outcome. Instead, conduct a 'premortem'. Imagine the project has failed spectacularly. Now, work backwards to figure out what could have caused this disaster. Was it a flawed marketing strategy? A buggy product? Poor customer service? By identifying these failure points in advance, you can put safeguards in place to prevent them from ever happening.

Inversion is a powerful tool for risk management and clearer thinking. It forces you to see the blind spots in your plans and to prepare for a wider range of outcomes.

The final advanced model we'll cover is Probabilistic Thinking. The world is not deterministic; it's a messy, uncertain place. Probabilistic thinking means using maths and logic to estimate the likelihood of different outcomes. Instead of thinking in black-and-white certainties, you think in shades of grey. A key tool here is understanding Bayesian inference, which is a method of updating your beliefs as you gather new evidence.

By combining these models—deconstructing problems to their core, understanding the context you're in, seeing the system as a whole, avoiding failure, and weighing probabilities—you move beyond simple logic. You build a versatile toolkit for navigating a complex world.