Want to learn the ideas in The Book Of Why better than ever? Read the world’s #1 book summary of The Book Of Why by Judea Pearl, Dana Mackenzie here.
Read a brief 1-Page Summary or watch video summaries curated by our expert team. Note: this book guide is not affiliated with or endorsed by the publisher or author, and we always encourage you to purchase and read the full book.
Video Summaries of The Book Of Why
We’ve scoured the Internet for the very best videos on The Book Of Why, from high-quality videos summaries to interviews or commentary by Judea Pearl, Dana Mackenzie.
1-Page Summary of The Book Of Why
Overview
The Book of Why: The New Science of Cause and Effect (2018) explains a science called causal inference, which mimics the natural ability that humans have to determine causation. By applying this science, we can mathematically find out what causes something to happen. With help from journalist Dana Mackenzie, renowned computer scientist Judea Pearl explains how he and his students at UCLA developed models and mathematical language for causality. Even if researchers don’t have data on all the variables affecting a phenomenon, they can use causal inference to create an estimate with confidence about cause and effect. Learning a mathematical language for causality will allow scientists to prove cause and effect for complicated phenomena; it might even allow them to devise theories without conducting experiments or creating machines with human intelligence.
Scientists first started using causal inference in the late 19th century. The concept was developed around the same time that modern statistics were being formulated as a science. Scientists of this era desired to understand heredity and genetics, but they couldn’t figure out how these things worked. Eventually, statisticians like Francis Galton and Karl Pearson concluded that it wasn’t possible to determine causality through scientific means alone. Modern scientists who work with statistics have come to rely on data as a way of getting closer to truth. They believe that if you collect enough data and interpret it correctly, then you can answer any question science might pose (provided there’s no bias). However, relying solely on data isn’t always helpful when pursuing scientific truths because questions need context too; otherwise they’ll skew your results in unexpected ways.
One reason causal inference is so useful is because it resembles higher-level thinking. Humans are capable of three levels of thinking and reasoning; these levels are sometimes called the ladder of causation. The first level involves the ability to see and observe changes in a pattern and draw conclusions based on associations. Animals, as well as some machines programmed with artificial intelligence, demonstrate the ability to utilize first-level thinking. The second level is the ability to recognize factors that might interfere with or intervene in how a process usually functions, and to draw conclusions about how these interactions will affect the process. A caveman would have been able to use step two when he imagined what would happen if more tribe members accompanied him on a mammoth hunt. What sets modern humans apart from animals, early humans,and machines is their third rung: imagining alternate scenarios. In an example during court proceedings,a prosecutor can explain that without someone firing a gun at an individual,he wouldn’t be dead. Imagining counterfactuals has played a crucial role in civilization’s development.
Although people can intuitively determine cause and effect in some situations, there are many complex systems that even the best thinkers cannot figure out. Scientists need to embrace causal reasoning if they want to make breakthroughs about why things happen the way they do. To do this, scientists should use a mathematical language that proves causality when analyzing data from experiments or surveys. Although statistics-based methods are still preferred by researchers, it may be possible for causal inference to lead to scientific discoveries that would otherwise be impossible with traditional statistical analysis alone.
