David Epstein’s new book, Range: Why Generalists Triumph in a Specialized World, is a remarkable meditation on the nature of talent. Epstein’s objective in Range is to critically examine an assumption many of us have about how truly exceptional people come to occupy the highest perches in their respective professions. That assumption goes something like this:
Excelling at the highest level in a professional or academic endeavor requires (i) narrow specialization with a firm, unwavering commitment to practicing activities that enhance the skills commonly accepted to be part of that specialization and (ii) skepticism about non-specialist activities, such as exploration, tinkering, and training commonly thought to be irrelevant to one’s primary area of specialization.
Epstein convincingly attacks both prongs of this assumption. Narrow specialization turns out to be a poor wager for a talented human being seeking to truly excel. The reason? It’s incredible simple: Earth shattering innovation — in science and sports alike — tends to come from folks who spent a long time nurturing different skill sets by exploring disparate domains in an unstructured way.
When I first picked up Range, I couldn’t believe what I was reading. Here was a well-respected journalist who had recognized something that always seemed intuitively obvious to me, namely that:
Lasting innovation in any human endeavor requires a mind that is unshackled by bias, empirically-dubious tradition and free floating risk-aversion.
I will call this intuitive axiom the Creativity Principle. I use it in my own consulting practice to evaluate whether a professional organization is encouraging or destroying the ability of internal teams to identify and implement optimal business and legal solutions.
Two notes: First, not all traditions are “empirically-dubious”; sometimes rules of thumb emerge from commercial experience which slowly cement into prosocial traditions within modern firms. Second, risk-aversion is not always “free floating”; sometimes a company’s business model requires focused risk-aversion in order to reach a broad, long-term objective. In cases like these, risk-aversion is principled, not free floating.
Without any further ado, let’s get to Epstein’s incredible book.
I promised we’d jump right in, but just one more thing. Most book reviews are linear, taking the gentle reader through an author’s arguments in a step-by-step fashion. I don’t think that will do with Range, and I suspect that Epstein himself would find a linear, color-within-the-lines approach to presenting the themes in his book rather boring.
With this final qualification in mind, allow me to riff on a few themes that take center stage in Range.
Wicked versus Kind Environments: The best place to begin to understand Range’s core argument is to think about the environment in which innovation occurs. Specifically, we need to ask whether these environments are based on rules we have been provided beforehand (for example, in structured games like chess) or whether these environments are characterized by either (a) an absence of rules or (b) limited information about what the rules might be (as is the case in life generally).
Epstein’s key insight is that the environments within which humans seek to innovate do not provide human beings with rules beforehand. These are called wicked environments, or wicked domains. On the flip side of that insight, Epstein reminds us that only some environments are “kind” — that is, characterized by rules that are disclosed to human beings beforehand.
One example of a kind environment is golf. As Epstein points out, “the learning environment is kind because a learner improves simply by engaging in the activity and trying to do better.” Golf has rules, and those rules are readily shared with all participants. By contrast, in wicked domains (a) the rules that structure the domain are “often unclear or incomplete, (b) there may or may not be repetitive patterns and they may not be obvious and (c) feedback is often delayed, inaccurate or both.” Thus, in wicked environments there is no such thing as complete prior knowledge of the rules of the game. By definition human beings are to some degree always flying blind in such environs.
What follows from this? Or to put the question more precisely: If an environment is wicked (rather than kind), how does the absence of prior knowledge about that environment impact those trying to innovate within it?
For starters, the causal structure of a wicked environment is unknown. By “causal structure,” I don’t mean anything fancy; I am only referring to the fact that in a complex environment there will be more than one model available to explain and predict observable phenomena. Weather is a good example of a wicked environment — we still do not have a model that perfectly explains and predicts weather phenomena. Physics is also an incredibly wicked environment—in physics, we still do not have a well-accepted theoretical understanding of gravity.
Second, in a wicked environment human beings must be acutely aware of the unintended consequences of picking this or that model to explain or predict certain phenomena. Here I merely want to highlight the fact that every causal explanation or explanatory model will include some consequences within it that may or may not be helpful for future innovators who are interested in solving a particularly thorny problem.
For example, in physics there are constants for the pull of gravity and the speed of light. These constants help physicists make an enormous number of predictions work out properly, but these constants also make theoretical physicists wonder why the constants are what they are. These theorists assume that they don’t have a final theory because these constants appear to be arbitrary. Thus, the theoretical constructs that employ these constants, along with the constants themselves, create questions (costs) that for some theorists require deeper answers.
Of Specialists and Integrators: In the previous section, we learned two important insights about wicked environments. First, the causal structure of such environs is up for grabs. This means that humans are cursed with choice; since the causal structure is unclear, humans are forced to pick from a wide array of different models. Second, every model we use has unintended theoretical and practical consequences.
These two important insights make it very hard for narrowly trained specialists to innovate in one very common context — that is, when a well-understood explanatory model has run out of gas and can no longer explain everything that human beings want to know about a particular wicked domain. Why is that? Well, specialists are gifted with a set of prescribed skills that they earned at great cost, typically by investing large amounts of time and effort being trained within recognized institutions of higher learning.
Such intense training makes it natural for specialists to privilege what psychologists Daniel Kahneman and Amos Tversky call the “inside view”. The inside view refers to an individual’s proclivity to conceptualize a problem in accordance with the concepts, tools and practices internal to her training and expertise. Specialists are optimized for working within a particular model. When a model’s limitations become a fetter on scientific progress in not something specialists are emotionally preprared to do.
This is not to say that specialists’ skills aren’t valuable. Quite the contrary, their skills are incredibly valuable and absolutely necessary for (i) calculating solutions to problems in kind environments and (ii) rigorously working out all of the implications of a particular explanatory model. Yet when a new problem arises, one that is outside the ambit of their received training, specialists are loathe to step back from the problem. As Epstein puts it:
Successful problem solvers are more able to determine the deep structure of a problem before they proceed to match a strategy to it.
So who are the problem solvers? Epstein calls these folks integrators, an apt term for individuals who graft insights from one domain to exploit them in another. (They don’t just graft: They configure insight x from Domain A carefully so that x applies properly to Domain B.) Integrators are incredibly curious about many different things, and they tend to be open communicators. But perhaps the most important trait these folks possess is the willingness to collaborate with others to rigorously test new ideas.
Perhaps the best historical example of an integrator is Charles Darwin. While many have the impression that his theory of natural selection just one day exited his synapses in complete form, the truth is much messier. Darwin was something of a professional outsider, always running multiple projects at the same time. The man had more than 200 pen pals with whom he traded scientific conjectures and ideas, a fact that reveals the high degree to which Darwin was a lateral thinker interested in absorbing everything he could — regardless of the provenance of a particular piece of evidence. Absorb, analyze and repurpose — this is one ready-to-hand way of thinking about how integrators try to make progress when tackling a hard problem.
If you zoom out just a bit, it’s obvious that Darwin had to be an integrator. History needed someone with enough knowledge of geology and biology to see that carbon-based organisms went through successive stages of change (evolution) similar to slow transformations in the natural environment. Without the ability and courage to graft one model over to a new domain, Darwin never would have written On the Origin of Species. He would just have been another curious dude with a lot of pen pals and a penchant for taking long, dangerous journeys on the high seas.
True Learning is Hard: When I first got the gist of Range, I wondered what Epstein might have to say about our educational system. It seemed obvious he would find our educational system unhealthy and broken. Moreover, I assumed he’d be unhappy with the fact that the current system is heavily tilted toward only producing specialists. Beyond that, I wasn’t sure what to expect.
Epstein paints a picture of our current educational predicament that is both familiar and disturbing. We now live in an era where education is both highly commodified and entirely too easy.
What do I mean by too easy? We have an educational infrastructure that is designed to make it easy for people to rehearse rather than learn. The difference between rehearsing and learning is best captured in the work of Nate Kornell, a cognitive psychologist at Williams College. Kornell has formulated something he calls the “generation effect”. The generation effect is the positive impact on an individual’s cognitive development that is due to generate answers to difficult problems without getting answers, hints or help from third parties. According to Kornell, even when a person gets the wrong solution, by struggling to generate an answer on her own, she has amplified her ability to learn in the future.
Precisely why struggling is so important for cognitive development is a complex question, one too large for this short piece. Yet Epstein makes a strong case that learning which looks inefficient in the short run is actually optimal in the long term. As Epstein puts it, “frustration is not a sign you are not learning, but ease is.”
Here’s the way I like to think of this insight: If you want to learn how to ride a bike, you can take two routes. One, you can learn how to ride and also minimize your anxiety around getting hurt by affixing training wheels to your bike. Two, you can learn how to ride a bike and face your anxiety around falling down by learning without training wheels.
If you take the first option, riding a bike will feel easier but you’ll actually be putting off the task of learning how to ride a bike; the training wheels give you the illusion of learning. If you take the second option, you’ll have to manage (face) your anxiety about falling from day one. However, because you do not have the benefit of training wheels, you will be slowly learning how to balance on your bike; the absence of training wheels is frustrating, but it also forces you to actually learn how to ride the bike. Without training wheels, there is no illusion of learning — there’s just actual incremental improvement in bike riding ability.
Integrators and other effective learners are primed to take the second route. They know that shortcuts are counterproductive so they don’t take them. In fact, I’d even go as far as to say that integrators are always hungry for a real slog through a wicked environment.
I believe by covering these three themes we are in a great place to illuminate the deep structure of Epstein’s argument in Range.
The first theme is about the environment — there we discovered that most innovation environments are wicked domains. These environments are wicked in the sense that their causal structure is not provided to us in advance. Such environments cannot be understood or even shaped by people who have an emotional need to know the “rules of the game” in advance.
Then in the second theme we saw that certain agents within wicked domains — the integrators — are best equipped to create lasting innovation and excellence. This is the case because integrators have fecund, elastic minds, and are therefore able to quickly pressure test different ways of looking at the world before committing to any one approach.
Finally, in the third theme, we learned what integrators actively participate in their own intellectual growth by facing tough problems without a safety net. This third theme may seem obvious, but as I argued earlier, it’s really important to recognize that integrators are incredibly mentally tough and unyielding. They develop these qualities in a way that many find surprising: though rigorous trial and error without the benefit of a scientific authority figure patting them on the head and announcing that they’ve made it.
So here’s the nutshell:
In any environment with poorly defined or unknown rules, you need people with plastic, nimble minds to integrate insights across disciplines if you want to achieve anything of lasting value. To encourage the development of those dynamic minds, we need an educational infrastructure without training wheels — an environment where each individual has no choice but to wholeheartedly own the process and outcome associated with any problem-solving endeavor.
Range is an outstanding book. Epstein not only manages to articulate a compelling vision of why generalists are so innovative, he also provides some profoundly important clues about the form modern education must take to encourage the development of people capable of thriving in wicked environments.