An anniversary edition offers an opportunity to reflect on how a book came into being and the impact it has made.
I did not expect Sources of Power: How People Make Decisions to be deemed successful enough for the MIT Press to publish a 20th anniversary edition. Yet MIT Press editors Amy Brand and Phil Laughlin tell me that Sources of Power regularly appears on the MIT Press’s list of best sellers even after two decades, and that it still sells between 1,500 and 2,000 copies a year. Since its publication, the book has sold over 50,000 copies and has been translated into six languages. A number of people have described the book as a classic, required reading for anyone who wants to study—or better understand—decision making. Twenty years later, it is worth explaining how I came to write the book and the impact it has made.
I wasn’t part of the decision research community and I never took a single course in decision making in graduate school. Maybe that worked to my advantage. My specialty was how to develop expertise, a real problem for the research team I belonged to in 1974 as we set guidelines for Air Force flight simulators. After the Arab oil embargo, jet fuel had become much more expensive and pilots had to do more training in simulators. Our team searched for ways to help them.
Thus, when I started to investigate decision making in the early 1980s, my perspective was on developing expertise—not on calculating probabilities or determining which of several options was optimal.
I had left my job with the Air Force in 1978 to start my research and development company, Klein Associates, and to conduct studies of decision making. Starting in 1985, my colleagues and I published descriptions of the recognition-primed decision (RPD) model which is described in detail in Sources of Power. Our research method was to study critical incidents in which people made difficult decisions under time pressure and uncertainty. The incidents were stories; many of them seemed exciting, at least to us. We never had a chance to tell these stories, however, because articles in professional journals have to be short and their format discourages stories. The only way to present these stories would be in a book, but who had the time to write a book?
Sources of Power got started because of a broken ankle. In 1992, my wife Helen took a six-month sabbatical in Jerusalem to continue her research on cultural differences in cognition. I accompanied her along with our younger daughter, who enrolled in high school. I didn’t have any consulting work lined up, but I didn’t mind because I was busy visiting friends and relatives. I also stayed in shape by running. It seemed suicidal to run on the roads so I used a path along a ravine next to our apartment. It wasn’t a proper running path—it was pretty rocky, and one morning I judged that it was just a matter of time until I sprained an ankle. Prophetically, I did just that about ten minutes later, breaking a bone in my ankle. Somehow I limped back to the apartment. I was pretty well immobilized and would be stuck in the apartment for the next month, after which I was scheduled to fly back to the United States for some lectures.
It seemed like a good time to work on that book idea. I had a 30-day window to churn out a first draft. I had a laptop computer with me (although I had never turned it on until then). I spent a few days outlining the book, mapping out the chapters, deciding which stories went into which chapters.
And then I started. I would read over my notes for the next day’s quota before I went to sleep, and then as soon as I woke up I’d be at it, at least six pages a day, sometimes ten or more, finishing by late afternoon, and then in the evening reviewing the topics for the following day. That’s how I wrote about 250 pages—most of the first draft—in a month.
After I returned to the United States with the draft manuscript, Buzz Reed, the CEO of Klein Associates, marshaled the resources of our company to edit it, track down references, fact check it, set aside my time for rewrites, and produce it. Five years and many drafts later, I submitted the manuscript to the MIT Press. It was published in 1998.
Thanks to the MIT Press marketing director, the book was reviewed in several different outlets, including Nature. I think what attracted readers was that the book described a new model of decision making. The RPD model was different from the existing accounts, focusing on how we use prior experience. It explained the mystery of how people were able to make tough decisions under time pressure and uncertainty. Previous research studies had usually ignored expertise and eliminated its influence. The RPD model demystified intuition.
All of these features made the RPD story fairly compelling. After one of our researchers wrote Tom Petzinger, the Wall Street Journal columnist featured Sources of Power in two columns, bringing it widespread attention. Then Malcolm Gladwell gave the book some prominence in Blink, which came out in 2005.
I think what is happening now is that each new citation the book receives keeps the momentum going, either bringing it to the attention of readers who haven’t heard of it, or encouraging readers who have heard of it to finally get a copy. Google Scholar finds more than 3,000 citations to Sources of Power from a wide range of professional journals including The Bulletin of the American Meteorological Society, Journal of Sport and Exercise, Personnel Journal, Journal of Marketing Research, Review of Policy Research, Journal of Nursing, Journal of Macromarketing, and Computer Methods and Programs in Biomedicine.
Sources of Power has reached a diverse audience. My colleagues and I continually hear from readers in a variety of fields working on topics such as railway safety, special operations, and avalanche avoidance decision making. We also hear about how training departments are replacing old models of decision making with the RPD model in different professional fields. Because of all the interest in applying the RPD model to training, several colleagues and I designed a ShadowBox® training program to foster effective decision making. We have been using it with child protective services caseworkers, petrochemical plant operators, law enforcement training developers, and military units.
The success of Sources of Power created lots of different kinds of opportunities, including very enjoyable collaborations with the Nobel prize winner Daniel Kahneman and the chance to serve on a team redesigning the White House Situation Room so that it could support more effective decision making.
The book described a different way of conducting research, a naturalistic decision-making (NDM) approach that investigated experienced decision makers performing realistic tasks, rather than novices performing artificial tasks in a laboratory. Researchers have subsequently applied this naturalistic strategy to other processes besides decision making—to sensemaking, problem detection and planning. Now there are hundreds of researchers around the world engaging in naturalistic decision-making projects and participating in regular meetings in the U.S. and in Europe.
Inevitably, the success of Sources of Power also encouraged caricatures and oversimplifications. One of the most common, and most annoying, is that the RPD model is just about using intuition and gut instinct, as opposed to more systematic decision strategies. Actually, the RPD model posits a two-stage process, starting with intuition, as decision makers recognize how they need to respond, followed by deliberate evaluation as they mentally simulate a possible response to see if it will work. A blend of intuition and analysis, not just gut feelings.
In reflecting on the impression that Sources of Power has made, I am struck by the primacy of stories over data. Certainly, my colleagues and I had to collect data to make a convincing case for the RPD model, yet when I read any extended citation of the work it is rare to come across any discussion of the numbers. Instead, people cite the stories, and go into detail about the firefighter who thought he had ESP (probably the most popular of the stories) and several others. Data are necessary to convince people, but stories are what they carry with them. Even in an era of Big Data, stories have a unique potential for conveying understanding. I decided to write Sources of Power because I wanted a home for the stories, so it is satisfying to see how strongly readers resonate with them. As you prepare to read or reread the book, I hope you will enjoy the stories as well.