A collection of Charlie Munger's speeches, talks and writing, first published in 2005 and edited by Peter Kaufman. It is a compilation rather than a composed book, and repeats itself deliberately.

Multidisciplinary models

Munger's organizing argument is that useful thinking requires the major ideas from several disciplines rather than depth in one. His image for the alternative is the man with a hammer, to whom every problem looks like a nail.

The recommended latticework draws the few big ideas from psychology, economics, mathematics, physics, biology, engineering and accounting, held well enough to be applied. His claim is that most large errors come from analyzing a situation with the one framework you happen to possess, when the situation is governed by another.

Inversion

The most transferable habit in the book is inversion. Rather than asking how to achieve an outcome, ask what would reliably prevent it, then avoid those things. He argued the list of ways to guarantee failure is shorter and more certain than the list of ways to succeed.

The 1986 commencement address applying this to life — a set of prescriptions for guaranteed misery — is the clearest example of the method and the shortest piece worth reading first.

The psychology of human misjudgment

The 1995 talk that closes most editions catalogues around twenty-five tendencies producing systematic error, including incentive-caused bias, commitment and consistency, social proof, and deprival superreaction — his version of loss aversion.

The tendency he returns to most is incentives. His position is that incentives explain behavior better than intentions, and that the first question about any recommendation is how the person making it is paid.

Reading it

The book is long, expensively produced and unstructured. Most readers get the value from the misjudgment talk and the commencement addresses, then use the rest as reference.