Heavy going, marred pedagogically by poor choice of notation, too many typographical errors, and overly-complex examples — yet nonetheless Causal Inference in Statistics: A Primer by Judea Pearl, Madelyn Glymour, and Nicholas Jewell offers an introductory glimpse of how to think about cause and not just correlation in collections of data. Bayesian graphs are Giant Step #1, and midway through the journey, a glimpse:

...There is a powerful symbolic machinery, called the do-calculus, that allows analysis of such intricate structures. In fact, the do-calculus uncovers all causal effects that can be identified from a given graph. Unfortunately, it is beyond the scope of this book ...

Perhaps there's a gentler pre-primer introduction to this topic? Perhaps it is yet to be written? Or perhaps Robert Tucci's 2013 "Introduction to Judea Pearl's Do-Calculus" is one such gateway, as recommended by Ferenc Huszár's "ML beyond Curve Fitting: An Intro to Causal Inference and do-Calculus" earlier this year?

(cf Statistics - A Bayesian Perspective (2010-08-13), Introduction to Bayesian Statistics (2010-11-20), Doing Bayesian Data Analysis (2013-11-02), Probability Theory, the Logic of Science (2013-11-18), Statistical Hypothesis Inference Testing (2013-12-01), ...) - ^z - 2018-09-16