Title: The Book of Why
Author: Judea Pearl and Dana Mackenzie
Category: Non-fiction
Rating: 3/5
10-word summary: Understanding causation is important and humans are great at this.
About The Book of Why
The Book of Why is a book about the study of cause and effect or causal inference. Judea Pearl explains how this field is different from statistics, how necessary it is and how it was created – despite the resistance of some statisticians.
He explains how important causal diagrams are and how to use them to represent certain phenomena and the causes that influence them. He also talks about the fact that humans are the best at causal inference. But he believes we will be able to create computers that understand cause and effect and think like humans do. When this happens, we may need to reassess how we understand notions like consciousness and free will.
The book covers a very interesting field, but to be honest, I didn’t really enjoy reading it. It felt like a complicated textbook for an advanced course – only I never took the beginner’s course in the first place. There were many things that I did not understand – something that has seldom happened in the past. But maybe you still want to read this book and your experience could be very different from mine.
Lessons from The Book of Why
Causal inference is the study of cause and effect.
Just because correlation doesn’t mean causation this does not mean we cannot study causation too – we can and we should.
Humans are the best at understanding causal relationships.
Data is not intelligence. Data does not understand cause and effect.
We should be able to one day build computers that can think like humans do and understand concepts such as cause and effect, morality, blame and so on.
What I like about The Book of Why
1. The study of cause and effect seems very interesting
To be honest, I didn’t even know this was a field before reading this book. But I quickly came to realize that the study of cause and effect seems quite fascinating and necessary in many cases.
I believe we should rely on science as much as we can to better understand the world we live in and to make good decisions and causal inference can definitely help us do that more effectively.
2. Judea Pearl is obviously very knowledgeable
If you read this book, I think it is quite obvious that Pearl understands how causation works, how to represent causation and how to test his assumptions. He also talked a lot about the origins of the study of causation, how it differs from statistics and how predominant figures reacted to the idea of causation in the past.
As a rule, it’s great to learn about a field or a topic from someone who has studied it and contributed to it as it happened in this case. If you want to learn more about causation, I think this book should definitely be on your list.
3. I learned a bit about some basic concepts
This was the first time I heard about or understood concepts such as correlation, causation, confounders, spurious correlation, controlling for certain factors and more.
What I don’t like about The Book of Why
1. This book is definitely not for everyone
While reading this book, I got the feeling that I was reading a textbook for a university course – and a difficult one. And despite all the years I spent in school, this book made me feel uneducated.
I think this was the first time when I read a book and felt like I cannot understand most of it. And you probably know that I read mostly non-fiction books that are informative and science-based, so I doubt the problem is my education level or reading habits.
This book seemed much more complicated and “dry” than any other book I’ve read before. So it felt like it was not written for a broad audience, rather for someone interested in causal inference or someone with a background in statistics. So if you have no problem with concepts such as confounders, the Simpson paradox, counterfactuals, the front-door criterion and you can read and understand mathematical formulas, I’m sure you will get much more value from this book than I did.
Quotes from The Book of Why
“If I could sum up the message of this book in one pithy phrase, it would be that you are smarter than your data. Data do not understand causes and effects; humans do.”
“Causal reasoning is easy for you because you are human, and you were once a three-year-old, and you had a marvelous three-year-old brain that understood causation better than any animal or computer.”
“The goal of strong AI is to produce machines with humanlike intelligence, able to converse with and guide humans. Deep learning has instead given us machines with truly impressive abilities but no intelligence. The difference is profound and lies in the absence of a model of reality.”
“There is no reason to refrain from building machines that are better able to distinguish good from evil than we are, better able to resist temptation, better able to resist guilt and credit.”
Should You Read The Book of Why?
Maybe not. I think that this book is probably too complicated for most people. If you do not know how to use mathematical formulas or you haven’t studied statistics before, you may not understand most of it – as I did.
But if you are interested in the science of causation (how it started, how to use formulas and causal diagrams and more), then read this book. You may learn a lot from it or this book could inspire you to learn more about the field.
And even though I think this is a complicated book, maybe this is a good reason to read it. It’s an exercise in humility, curiosity and perseverance. It showed me how much I still do not understand and cannot understand and how much there is to learn! It was definitely an interesting experience.