My Main Page with Links to My Other Book Reviews

Thinking and Deciding by Jonathan Baron

Thinking too often stinks. Jonathan Baron aims to help. A general and wide-ranging look at the subject, Baron emphasizes psychological research--the best part of the book. Baron poorly defines rational thinking as any thinking that “best helps people achieve their goals.” Rationality, he writes, is not “cold calculation of self-interest." (A better definition of rational thinking is thinking aimed at finding the truth.) Baron argues we often search poorly for information and make bad inferences from the information we have found. Baron argues for the teaching of heuristics, standards, belief formation, truth seeking, transfer ability, and avoidance of psychological obstacles. Baron concludes with a consequentialist vision.

Due to hedonistic and other reasons, consequentialism is unpopular.

Thinking and Deciding offers numerous heuristics, sometimes helpful rules of thumb. Baron writes that we should look at problems from numerous perspectives. We should list important information. We should brainstorm and employ the Delphi technique. We should play the devil's advocate, thinking up counterarguments and counter-counterarguments to the counterarguments, creating a benign spiral of argumentation. We should be careful in choosing both long-term and short-term goals, whether we have too many goals or too few goals. We should restate problems, break problems into parts, and break goals into parts. We should experiment with various methods from utility theory such as value trees. We should find out how problem solvers solved similar problems in the past and experiment with a variety of problem solving methods.

It is extremely important, he argues, to reason in specific terms about the comparative importance of claims. One might, for example, rate the harm from a broken fibula as 40 times greater than the harms from a broken pinky. If you hate mowing the lawn, and I mildly dislike mowing, we might agree that for every hour I mow, you perform a chore I hate for an hour.

The best thinkers, he writes, have lots of thoughts. They know how to edit worse thoughts. Baron argues that worldwide political opinions are atrocious, often due to groupthink.

His excellent section on groupthink covers Janis’ study of decisions made before the Bay of Pigs Invasion. Some of the errors arising from groupthink include assuming one's group is automatically right, ignoring the potential for terrible outcomes, selecting immoral ends or means, suppressing counterevidence, and failing to find better alternatives and claims.

Good thinking requires hard work. It is attuned to cognitive dissonance. It goes where the best evidence leads despite social pressures. But individuals would rather read for pleasure and believe what they want to believe. Many great arguments have fallacies, but no great arguments leave out important, relevant claims. We should not cynically and fanatically reject arguments because they have minor flaws. A fallacy is a claim that fails to support a conclusion it is not a claim that wrecks or undermines a conclusion. The weight of an argument depends on the weight of its good points. And those good points should be weighed against the weight of good points for alternative arguments. Look for and find the best available arguments and conclusions with an open mind.

Numerous fascinating studies pepper Thinking and Deciding. One hedonic study fed individuals plain low-fat yogurt for eight days. Surprisingly, instead of growing to loath the yogurt, they grew to like it, suggesting that individuals learn to enjoy or at least tolerate activities they once disliked. Of course, that study may not be relevant to many non-food situations. Or it could be junk science for hundreds of other reasones. Learning to like something depends heavily on beliefs and how horrible the actually thing is. It is easy to learn to like plain low-fat yogurt because it is only mildly distasteful, it doesn't poison you, and there is little cultural pressure to dislike yogurt.

Another study indicates individuals preferred immersing their hands in cold water for one minute followed by 30 more seconds in slightly warmer water over holding their hands in cold water for one minute, then warming them, a result that suggests memories of events are over influenced by endings or best points or worst points.

One often noted, and more often ignored, point is that direct efforts to obtain positive subjective states--drugs, overeating, voyeurism, edutainment, infotainment--reduce the disposition to have positive subjective states in the future, which then leads to more direct attempts at pleasure in vicious spirals of pleasure seeking that produce less pleasure, sacrificing more important matters.

He also notes that humans overestimate the probabilities of both rare events and highly probable events. Humans think the minuscule probability of dying from an accident at a nuclear power plant, for example, is much higher than it actually is. We are also likely to think an 85 percent probability event is closer to 99 percent probable. But we tend to underestimate semi-low probability events. An event that one thinks has a probability of ten percent may actually have a probability of 20 percent.

Baron gives us a fascinating study on framing effects: Most individuals think it is wrong for a company to cut wages seven percent when inflation is zero percent. Most people think it is acceptable for a company to increase wages only five percent when inflation is twelve percent, all other factors held constant in both cases. Both cases are morally and economically similar for the workers, yet people think they differ. (For the economy as a whole, they are not identical. Both zero and 12 percent inflation can cause serious economic problems.)

The omission bias--judging harmful actions as worse than harmful inactions--plays a major role in Baron's worldview, but omissions as harmful as actions are often not as wrong or evil. An omission of not stopping at a stop sign and killing someone is not as morally wrong as first-degree murder, though they may both produce the same result. Results are not everything. Motives and efforts count too. Worth skimming. Book review article by J.T. Fournier, last updated April 24, 2023.

(H) 1994

My Main Page with Links to My Other Book Reviews