A while back, Elon Musk revealed that his success is based on a special way of solving problems called first principles thinking.
In this type of problem-solving method, you don’t rely on old ways of doing thinks (analogies), but instead you rethink the problem from the ground up. You go back to the fundamental assumptions that underpin the problem and try to see what other possible directions could be followed.
However first principles type thinking is not natural to most humans. The human mind evolved in order to promote survival. This favored speed and efficiency in the thinking process and not a slow, thorough evaluation.
Barriers to first principles thinking:
In order to reach quick decisions in a world of uncertainty, the human mind uses heuristics, or shortcut methods to solving problems. In most cases, these result in good enough answers, but there are instances when heuristics fail. These failures are called cognitive biases.
If you want to know more about cognitive biases, check out my series on critical thinking and my cognitive biases framework:
There are two main cognitive biases that make it harder for you to solve problems using the first principles method: the Einstellung effect and functional fixedness.
These two effects are not bad things in themselves. They do promote fast thinking and an efficient way of doing things. For most tasks this is the best way to proceed. In most cases, you want to follow best practices and do things using the tried and tested method.
Daily practice of the same moves reinforces them in the brain. More and more synapses between specific neurons are built up, which helps you to pass on the information faster between them and to be able to execute these moves faster and better (and after a significant amount of practice often even without thinking about them, just by instinct).
This is what makes experts so good at what they do. Their heuristics become automatic and since they have been in these situations many times before, they can execute almost effortlessly.
However there is also a problem. Once these neural pathways connecting different neurons are built up, it is hard to pass around them and to do things differently. When one neuron lights up, the other ones connected to it light up too.
That’s why for example functional fixedness is so strong. If you have a hammer and the only thing you have ever done with it is to pound in nails, then you will only see one use for the hammer. You might become really good at pounding in nails with a hammer, but you will fail to see the other potential uses of the hammer.
What techniques can you use in order to be able to think in first principles?
1) Set up the right problem to solve.
The way you set up your problem will have a huge influence on what answers you will arrive at. Each framing of the problem can imply a different set of assumptions and first principles.
This is oftentimes the hardest thing to do. The big question you are asking and trying to find the solution to needs to be clear and precise. That’s why you should spend a significant amount of time doing this.
Once you have established the problem you want to solve, things will flow from there.
2) Ask lots of questions.
Throughout the entire process, you need to be asking a lot of questions. What? Why? How?
Question everything. Question your assumptions, question your choices and question your solutions. Try to find holes in everything you do. Create counterfactuals and develop alternatives.
If you want to find out more about how to phrase the right questions, read the part on questioning here:
3) Find your assumptions -> find first principles.
Any type of argument or solution is built upon a series of assumptions. What you need to do in order to find those first principles is to go through the different assumptions you are making and see which ones can be broken down further.
At every step of the way, don’t forget to question these assumptions. If your assumption is not correct, then your conclusion will also be flawed. That’s why it is very important to keep an open mind through the entire process.
A first principle is a basic statement that cannot be broken down further and upon which all other statements are built. In the first principles method, this is your starting point.
4) Use the generic parts technique.
One technique that you can use in order to find the first principles of something and to free your mind of functional fixedness is the generic parts technique. There are two basic questions that you ask as part of this process:
- Can it be broken down further?
- Does my description of the object imply a use?
The first question is quite straight-forward. You take a look at an assumption and try to determine whether it can be broken down further.
When explaining his use of the first principles method, Elon Musk used the example of a battery and how he broke it down further to find the first principles.
First he asked what are batteries made of? That resulted in a series of parts that make up a battery. After that he went one level further down and so on and so on until he came to the real first principles of the composition of batteries, the materials themselves: cobalt, nicked, aluminum, carbon and polymers.
The second question in the generic parts technique is the one that really helps when you are stuck and functional fixedness is blinding you.
Tony McCaffrey, the researcher who came up with the technique, illustrated how this works with an example from an experiment.
There is a candle and a match. You also have two steel rings. Your goal is to join the two rings together only using the candle and the match. How would you do this?
The key to solving this problem is the second question. You need to be able to get rid of functional fixedness and see different, non-traditional uses of the candle and the match.
One way to solve the problem is by breaking down the candle to its constituent parts. The two parts of a candle are wax and a wick.
The problem with these labels is that they still imply a use. What is a different way to call a wick?
Well, it is a string!
This is the generic term that you can use and suddenly you see a world of possibilities. In this way, you can solve not just this problem, but numerous other problems you might come across.
5) Be agile.
Another thing that you need to remember is that this is not always a linear process. You might need to return to different stages at different times and some parts you can run in parallel.
If you are getting nowhere, don’t be afraid to go back and rephrase your initial question and change the problem you are trying to solve.
The key to everything is keeping an open mind.
If you want to know how to manage any type of project in an agile way, then read this:
–Agile Project Management and Productivity
This lesson was just a short teaser summarizing first principles thinking. I have written a series of more detailed articles on what it is, what barriers there are to thinking that way, and how to overcome them. If you want to learn more, read the series of articles below.
If you want to read an introduction on first principles thinking and a discussion on barriers to thinking in first principles, click below:
Introduction to first principles thinking and barriers to thinking in first principles.
If you want to read more on what types of techniques you can use in order to overcome those barriers and solve problems using first principles, click below:
The techniques for first principles thinking.
If you want to read more on the applications of first principles thinking, then click below:
The applications of first principles thinking.