If I asked you to close your eyes and picture a scene from the words “rational thinking,” what would rise in your mind?
Before reading The Mystery of Rationality, two images might have surfaced in my mind: a detective puffing on a pipe, crouched over a desk and reasoning out a criminal’s identity; and a scientist at a blackboard working through equations and arguing for a new discovery.
And you probably thought of them too. Among all the familiar images we have, few seem more skilled at calling on rational thinking to discover the truth and solve difficult problems than detectives and scientists. Yet these two images share another feature in my mind: they are both lone heroes.
Even if Holmes was allowed to bring Watson along while following clues, he ultimately pieced them together with the “thinking palace” in his own head. What detective would need an assistant to help organize his thoughts while making a crucial deduction?
You can probably also recall stories about scientists solving scientific problems through solitary, profound thought. Physics has its famous “lone genius” story: the German physicist Heisenberg retreats to an island, thinks alone for several months, and derives the foundations of quantum mechanics — the uncertainty principle — by himself.
Rational thinking seems to be something that happens to “one person.” The less disturbed we are and the more deeply we think things through alone, the more fully rationality can show its power. Right?
If that is what you think, you really should read The Mystery of Rationality. Only after finishing it did I realize that this impression may be completely wrong!
In fact, rational thinking is not especially friendly to lone heroes. It is at its most powerful not when we are alone, but when people interact with one another.
The two authors of The Mystery of Rationality (both leading cognitive scientists) propose a strikingly counterintuitive new theory — the interaction theory of rationality — to summarize this point. From the perspective of interaction theory, humans evolved rationality not because it helps us think through problems more clearly, but because it helps us interact more effectively.
In other words, rationality originally evolved for doing society.
Rational thinking helps solve two major problems in social interaction: how to prove to others that we are trustworthy collaborators, and how to expose untrustworthy collaborators. Defending our own actions — ignoring doubts — and questioning other people’s actions — ignoring their defenses — are rationality’s two basic functions. Rationality has certainly helped humanity solve countless problems, but those achievements were nearly all trophies won by people in a collective ignore-and-counterignore battle.
This theoretical framework explains some peculiar features of rationality especially well. For example, why does rationality have such a strange dual nature? It is “lenient” toward ourselves, but “strict” toward others.
When we use rationality to reflect on our own views and actions, it seems always eager to defend us. No matter how many holes there are, we do not easily notice them; we are readily persuaded by our own shallow reasoning.
The deductions a detective makes alone inside his “thinking palace” may not be all that brilliant. The Mystery of Rationality contains a little passage about Holmes that goes like this —
Holmes: My dear Watson, you have such an enchanting female companion with you. I won’t interrupt! I’ve been watching the two of you for quite a while, and I strongly suggest that you ask this lady out again. She suits you perfectly. She may be younger than you, but not by much. I know you like women with dark hair, while she is blonde, but surely we can overlook that difference. I have also observed that when you discuss private matters, there has been no intimate physical contact between you. But I believe that will happen very soon. Watson, I guarantee it: this woman is perfect for you!
Watson: She’s my cousin.
If Holmes were a real person, his deductions would probably look much more like that exchange than the divinely precise reasoning in the novels. When people reason alone, the more likely scenario is Holmes in that passage: even with obvious clues in front of him, he can infer that Watson’s companion is not his lover, but perhaps he is so eager for Watson to find happiness that he first decides Watson has a girlfriend and then refuses to see every contradictory clue.
Why does this happen? From the perspective of interaction theory, the explanation is simple — when people are alone, they are closer to the situation in which they defend their own actions during interaction. At that point, people lean toward a lenient standard, because they only ask their collaborator to accept the reason they offer (that is, successful ignoring), not whether the reasoning is genuinely sound.
Scientists often make this mistake too. A study cited in The Mystery of Rationality found that when scientists think through their own research results alone, they are quite good at finding reasons to dismiss data that challenge their hypotheses. When experimental results are disappointing, scientists do not necessarily reason impartially and objectively about whether their original hypothesis may be wrong. Instead, they readily enjoy all kinds of flimsy explanations that rescue it, such as “the result was bad because a technical problem occurred” or “the experiment had an operational flaw.”
That sounds pretty flimsy, doesn’t it? So how do we explain all the reliable discoveries humanity has made through scientific research?
The answer is that once people turn the object of rational thinking toward someone else, they switch from “ignore mode” to “defensive-ignore mode.” Since trusting an unreliable collaborator may cost us dearly, the standards of rational thinking suddenly become strict.
Science makes powerful use of this fact to correct errors. Academia has a familiar rule — peer review: after a paper is submitted to a journal, it is sent to the author’s peers for review. In plain English, this means your peers raise a bar against you, then you raise it back; only when you successfully counter the bar can the paper be published.
Whenever people spar with someone else in this way, their rational thinking immediately becomes sharper. Errors that are hard to detect while reasoning alone are often exposed when someone else pushes back. In the actual history of science, lone-genius stories are extremely rare. The vast majority of important scientific discoveries were made through high-quality argument.
As for the story of Heisenberg discovering the uncertainty principle, a careful look at the historical record shows that his whole process of solving the problem was actually full of interaction. Most important was his constant, intensely close correspondence with another great physicist, Wolfgang Pauli. By continually facing Pauli’s challenges, Heisenberg was able to develop his ideas fully. In fact, the first draft of his paper on the uncertainty principle appeared in a letter to Pauli. Even geniuses need to argue with other people before they can develop their best ideas.
In that earlier passage, if Holmes’s archrival Professor Moriarty had also been present, and the absurd deduction had come from Moriarty’s mouth, I imagine Holmes could have picked out a whole pile of holes in it.
That is why rational thinking does not work well inside the head of a lone hero. Only when minds cross swords can rationality display its full power.
This feature of rationality reminds me of a concept that has been very popular in the startup world in recent years — lean startup.
Lean startup is a philosophy like this: when developing a product, a company should avoid shutting itself behind closed doors and polishing the product to perfection before launching it. That is a lot like Holmes racking his brain alone for half a day — exhausting, and not very productive. The right approach is to develop a basically usable prototype as quickly as possible, put it on the market, let feedback tell you what is worth improving, and then iterate rapidly. Many innovative companies have successfully used this approach to iterate their products. Take WeChat: its first release was an extremely stripped-down version, and its quality was nowhere near that of its competitors. After several rounds of iteration, however, it evolved to a level far ahead of the competition.
Holmes: My dear Watson, I don’t think I’ve met this enchanting companion of yours. Would you introduce us?
Watson: She’s my cousin.
Lean startup is really just this kind of back-and-forth interaction in action: make full use of everyone’s strengths during interaction and solve problems efficiently with less wasted effort.
Ultimately, the perspective of interaction theory on rationality can be carried over into many situations in everyday life. You can experience it in detail in The Mystery of Rationality.
After reading The Mystery of Rationality, when I think of “rational thinking” again, two images still rise in my mind: Holmes puffing on his pipe … and Moriarty arguing with him; a scientist working equations on a blackboard … so he can pull out evidence and push back against his peers.
Human rationality is not found in calm seas, but amid a cacophony of voices.
Zhichao Wei
Faculty member, Department of Psychology, Ningbo University
Host of Ximalaya FM’s The New Psychology Lessons with Zhichao Wei
Original article
Leave a Reply