Hızlı ve Yavaş Düşünme (Thinking, Fast and Slow)

Kahneman divides the mind in two.

System 1 is fast, automatic, effortless. Recognizing that a face is angry, working out two plus two, driving on a familiar road. It cannot be switched off; it runs even if you do not want it to.

System 2 is slow, effortful, demands attention. Seventeen times twenty-four. A tax form. Looking for a particular face in a crowd.

The subject of the book is how often the first overrides the second.

I got caught again and again

Kahneman's structure is this: first he asks a question, you answer, then he explains why you were wrong.

I was wrong almost every time.

The best known: a baseball bat and a ball cost 1.10 lira in total. The bat costs 1 lira more than the ball. How much is the ball?

The first answer that came to mind was wrong, and to find the right one I had to stop and calculate. What Kahneman describes is exactly this: an answer comes by itself, and checking that answer takes a separate effort.

There is a big difference between reading about a fallacy and seeing it in yourself. The book's strength is here, and this is why reading a summary is not enough.

The ones that resonated most with me

Anchoring. A number sitting in the middle influences your later estimate even if it is entirely unrelated. This is why the first price named in a bargain matters so much.

Availability. You think something that comes to mind easily is more frequent. Being afraid to fly after news of plane crashes.

Loss aversion. Losing a given amount feels about twice as heavy as gaining it. It explains why letting go of something is so hard.

The planning fallacy. People systematically underestimate how long things will take. For me this was already a known problem; learning that everyone has it was strangely comforting.

The experiencing self and the remembering self. What you feel while living an experience is not the same as what you remember afterward. Remembering looks at the peak and the end; the length of the middle hardly counts.

This last one is, I think, the most thought-provoking idea in the book. How a vacation or a week "went" and how it is "remembered" are separated, and it is not clear which one you should decide by.

The confession at the end is what matters

Kahneman says: knowing these things does little to prevent your own mistakes.

System 1's errors go on even though they are known. You read five hundred pages and the next day you fall into the same trap. Kahneman writes that he falls into it too.

This sentence felt very familiar to me. I know what I do wrong. Knowing is not enough.

What he suggests instead

Turning distrust of your own momentary judgment into a system.

Checklists. Predetermined criteria. Written criteria before a decision. In a job interview, writing the questions in advance and asking every candidate the same ones.

And a technique called the "premortem": before a decision is made, asking the team "a year has passed and this project has been a complete disaster; what happened?" People get out of defense mode around the decision and name the real risks.

So the solution is not thinking better but building a structure that does not require thinking.

I already knew this from somewhere else: I write because I cannot trust my memory, I set alarms because I cannot trust my sense of time. Kahneman applies the same logic to decision making, and it was good to see it in another language.

What needs to be read knowingly

Some parts of the book, especially those on the social priming effect, were seriously contested after publication, and some of those studies could not be replicated.

Kahneman himself acknowledged this and wrote plainly that the confidence in the relevant chapter was too high.

The ironic thing is that this is itself an example of the book's own thesis. Believing too much in a strongly told finding; drawing big conclusions from a small sample.

The backbone, the decision-making fallacies part, was largely unaffected by this debate. But you need to read knowing the difference.

Amos Tversky

Behind the book there is a second name: Amos Tversky, with whom Kahneman worked for many years.

Most of these studies are joint work of the two. Tversky died in 1996; Kahneman received the Nobel in 2002, and because the prize is not awarded posthumously, Tversky could not receive it.

Kahneman talks about him in many places in the book, and those sections are the warmest parts of it.

How I read it

I did not try to read the five hundred pages from start to finish.

The chapters are short and each explains a single concept. It is one of the books best suited to reading in stretches.

When I read many chapters back to back, the concepts blurred into each other; one or two chapters a day sat better.