In 1736 Benjamin Franklin sat in the Pennsylvania Assembly, and one of the other members did not like him. Franklin never says what the quarrel was about, only that the man opposed him at every opportunity, and that a few polite overtures had gone nowhere. Then Franklin heard the man owned a very scarce and curious book. He wrote a note asking whether the gentleman would do him the favor of lending it for a few days. The book arrived immediately. Franklin returned it about a week later with a second note saying how much the favor had meant to him. The next time the two met in the House, the man spoke to him — something he had never done before — and from that day forward he was ready to help Franklin in any matter. They stayed friends for the rest of his life. The whole turn came from borrowing a book.
🧠 The Maxim He Lifted From an Old Book
Franklin wrote the story up in his autobiography and reduced it to a maxim he claimed to have borrowed from somewhere older: “He that has once done you a kindness will be more ready to do you another, than he whom you yourself have obliged.” Read it twice, because the order is odd. The man who already helped you is a better bet than the man you helped.
The experiment that turned the anecdote into a finding came from Jon Jecker and David Landy in 1969, published in Human Relations. Students took part in a question-and-answer competition run by a researcher, and some of them won money. Then the researchers did three different things to the winners. One third were approached by the researcher himself, who asked for the money back on the grounds that the prize pool had come out of his own pocket and he was short. One third were asked by a secretary instead, with the reason that the department’s funds were low. One third were left alone. Everyone was then asked how much they liked the researcher. The group the researcher had asked in person liked him the most. The group the secretary had asked liked him the least.
🤔 Two Things That Should Not Work
The first strange thing is the direction of causation. We assume the sequence is like-then-help: I like you, therefore I help you. The Franklin effect claims the arrow can run the other way. I help you, and my attitude has to catch up with what I did — so I decide I must like you. In the lab the size of the favor was tiny and the whole exchange lasted a week, and it still moved how people felt about a stranger.
The second strange thing is the middleman. The same request, for the same amount of money, on the same afternoon, flipped from making the researcher more likable to making him less, depending only on who made it. That detail is worth more than the headline. If the mechanism were simple guilt, the secretary’s version should have worked too — the student still had to give the money back. What the secretary could not supply was the sense that a specific person had reached out and asked for something. That is why one reading of the effect credits cognitive dissonance and another credits something warmer: the request itself is read as an opening bid for a relationship, and people respond to being approached. Both explanations are live, and the second one notes that a middleman cannot be read that way.
The effect also runs backwards into something colder. In a 1971 study from the same era, John Schopler and John Compere had participants teach another student — actually an accomplice — and assigned them to either encourage the learner or insult him. Asked afterward to rate the learners, the ones who had been insulted came out less likable, across the board. The participants’ own behavior toward these people had rewritten how they saw them. That pattern scales upward into one of the uglier ideas in social psychology: we do not like the people we hurt, so we construct a reason not to, and the construction makes the hurting possible again. The scholars who write about it reach for jailers, camp guards, slurs invented for enemies, and long family feuds that neither side can end.
The honest caveat, if you are going to repeat any of this at dinner, is that the evidence base is thin. The 1969 study is small, the 1971 study is a follow-up rather than a replication, and the one attempt to reproduce the effect with both Japanese and American participants was smaller still. This is not an effect with a famous public replication failure, and it is also not an effect that has been put through the kind of large preregistered test that power posing and ego depletion got. Treat it as a real pattern with a modest amount of paper under it.
🔗 The Ask That Builds the Bond
There is a design lesson hiding in the secretary detail. Products that sit with people over time — the assistant, the companion, the teacher, the app — are usually built to give. They answer, they help, they anticipate. But giving is not the only lever. Let me ask you something small. Pick which of these three versions sounds right. Tell me what you think the market is doing next. The user who has just supplied an opinion has done something, and doing something shapes how the relationship gets filed. That is also why mentorships survive on asymmetrical asking, why the salesperson who asks for advice outperforms the one who offers help, and why the reader who leaves a comment on chapter three is the one still reading at chapter ninety. Investment is not just the cost of caring. It is a cause.
🎲 Borrow a Book From Someone Who Dislikes You
Two versions. The version for other people is a party trick with a long tail: next time someone at work is cool toward you, do not send a small gift. Ask them for one small thing they can actually deliver — a recommendation, a book, fifteen minutes about something they know and you do not. Then thank them specifically. If it works the way Franklin’s did, you may notice the temperature change the next time you meet. The one catch you already know: the ask has to come from you, in your own words, not through a colleague.
The version for yourself is less comfortable. Bring up the person you currently have the least patience for, and check whether you have ever, at some point, done them a favor or taken a hit for them. The effect’s best predictor is not that you are shallow. It is that your feelings about people are partly a receipt for what you have already done, and you rarely notice the receipt being written.
Sources: Benjamin Franklin, The Autobiography of Benjamin Franklin (the borrowed book and the maxim); Jecker and Landy, Human Relations, 1969, 22(4), 371–378 (the three-group return-the-money design and the researcher-versus-secretary result); Schopler and Compere, 1971 (encouraged versus insulted learners, and likability ratings); Niiya (a smaller reproduction with Japanese and American participants, and the reading of the request as an attempt to open a relationship, via Takeo Doi’s theory of amae); Dale Carnegie, How to Win Friends and Influence People (the request as flattery). Mechanism remains disputed between cognitive dissonance and the relationship-opening account; evidence base is small and has not been through a large preregistered replication.