Recently, I concluded that our understanding of expertise, especially in the sense of the 10,000 hour meme, is seriously flawed. Even though there is something real there. And I don’t just mean the understanding conveyed by Malcolm Gladwell, who popularized the idea. I include the primary researchers such as K. Anders Ericsson on whose work the popular accounts are based.
The problem isn’t what you might think. It’s not the basic model of 10,000 hours of practice coupled with metacognition that’s the problem. The observation that you can get to useful levels of skill short of mastery in less time doesn’t fundamentally challenge the model. Nor is there a serious problem with the ideas that practice is necessary or that practice without metacognition is insufficient. That’s all true enough. Nor is the idea vacuous and tautological as some suggest.
The real problem is that research on expertise focuses on fields where “expertise” is a well-posed and objectively codified notion. This means mature fields that are closed and bounded, and can be easily observed, modeled and studied under laboratory conditions. So it is not surprising that the work of researchers like Ericsson is based on fields like “medicine, music, chess and sports” (Wikipedia) or “music, science, golf and darts” (Ericsson’s own website).
Notice something? They’re all sharply circumscribed and regulated domains.
Open vs Closed Worlds
Sharply circumscribed and regulated domains are what I called closed worlds in Tempo. The opposite kind with fuzzy, permeable boundaries and low regulation are open worlds.
Of course, in reality there are no closed worlds, except in our heads. It takes a certain amount of violence (such as physical violence or regulatory policing) to keep the external world closed and in correspondence with our mental models.
We define closed worlds by consensus on top of fundamentally open ones, by abstracting out a finite game from an infinite one and appointing guardians to police in the real world the boundaries that are mainly in our heads. One effect of this process of closing a world is that it necessarily drives a shift from trader (commerce) values to saint (guardian) values.
So it is not surprising that three of those four items in each list are in fact not significant economic trades at all, and the ones that are (medicine and science) are practiced under such highly regulated conditions (the healthcare sector and academia respectively) that they might as well be artificially closed domains.
These competitively compromised economic sectors, incidentally, suffer from a phenomenon known as Baumol’s disease: low rates of productivity increase through innovation, accompanied by cost increases that create increasingly unsustainable microeconomic conditions. I won’t talk about that in this post, except to note that it is a direct result of the specific model of deliberate practice that prevails (thanks to protection via closure) within those domains.
Science might seem like an outlier in this set, but it isn’t really. The academic study of “science” by psychologists like Ericsson must rely on bureaucratic productivity measures like publications and citations, and the consensus of “leading experts” of a given scientific fields about what “expertise” means (since psychologists can’t be expected to themselves be experts in other scientific disciplines). So you necessarily end up studying Science! the bureaucratically proceduralized game rather than the scientific sensibility that drives significant breakthroughs. I’ll claim without arguing in detail, that the real scientific sensibility is a trader sensibility rather than a guardian sensibility.
The common feature of all these domains is that they create artificially leveled playing fields, within which people compete on the basis of prowess rather than in an open, Darwinian sense of variation and natural selection.
This makes all the difference.
Sustaining Metacognition: Finding Creative Flow
The characteristic feature of closed-world competitions is that there is a clear and unambiguous external error feedback signal from an authoritative source that also supplies extrinsic rewards such as money and status. Learning to incorporate that signal in your ongoing deliberate practice is the purpose of the metacognition.
In domains like sport and music, errors can only count as negative. Which means metacognition serves a particular purpose: getting to a notion of perfection by eliminating one performance error after another, preferably with a Tiger Mom hovering over you.
In that constrained context, the metacognition in deliberate practice can only be used in a sort of customer-driven QA way: you learn to measure up to, and “delight” the refined tastes of judge-like figures (typically retired-with-honors former expert performers). You achieve perfection by eliminating errors in meeting a standard of “perfection” that is somewhat arbitrarily determined by the most conservative customers.
This understanding of expert performance should also lead naturally to a communitarian understanding of expert creativity: one based on external rules and received standards and aesthetics. And this is precisely what you find in the work of Mihalyi Csikszentmihalyi, one of the pioneers of positive psychology. A deeply flawed thinker I took way too seriously about 15 years back.
Csikszentmihalyi is known for two big ideas. The first is the idea of flow: the Zen-like state of peak performance that you get to when you’re operating at the edge of your abilities, with cognition and metacognition blurring together. The second is his model of creativity: which he constructs not as an individual trait, but as an product of a triad: individual, field and domain (as you might expect, for his creativity studies, he used exactly the same sorts of domains Ericsson uses in his expertise studies).
- Individual is a person embedded in a performance domain.
- Field in his sense is a community of experts who determine the worth of your work through peer review, or some analogous process.
- Domain is a particular area of creative performance mediated by symbols.
By his model, you develop expertise by gaining fluency in the symbolic domain, internalizing the standards of your field, and constantly striving to push the envelope just a little beyond your maximum ability, but not too far beyond. The sweet spot of performance is the position that induces flow.
If you’re in flow, you can get flawless performance and lose yourself in it. This is in fact the definition of prowess. That sense of being at your peak of performance, with a subconscious sense of your own virtue, valor and skill combining in a graceful inner symphony. A feeling of being in a psychologically sacred state. If you’re in flow and being “creative” in this sense you’ll know what the judges will say before you even submit your work to them for judgement.
Survive long enough and you can retire with honors and become a steward and guardian in your own right, with younger people looking to you for judgment. Or you might end up dying trying to prove your prowess in fields where that’s possible, and be martyred, which in some sense is an even better outcome for aspiring guardians.
This model is an accurate one in descriptive terms, but a terrible one in normative terms. So let me propose a highly prejudiced contrarian reading of what Csikszentmihalyi is describing.
What we have here is a closed boundary defined by a symbolic domain (rather than raw, unmediated reality), within which there are awestruck beginners and awe-inspiring experts. Expert performance is primarily a beautiful feeling that is derived not from the effects of the performance itself, but from the integration of metacognition and cognition into an internal superego. An inner [Tiger-] parental spectator that supervises performance according to an external standard of error-free perfection, and rewards you psychologically to the extent that you meet that standard. The performance is necessarily an incremental push beyond the edge, where received standards of performance and aesthetics can be reliably extrapolated. You cannot apply standards of violin performance if you suddenly decide to use your violin as a bat in an improvised game of softball (a profane use of a violin that is nevertheless physically possible).
In short, this is sustaining innovation driven by groupthink, divorced from reality by an internal language of symbols, and limited to what doesn’t violate sacred standards of quality or prevailing aesthetic sensibilities. As determined by honored retirees whose expertise is beyond doubt.
The reward for such metacognition is in fact the subjective state of flow: a regime of behavioral sacredness that is valued for its own sake rather than for its effects, and which is rewarded in social ways.
Disruptive Metacognition: Finding Ugly Awkwardness
It’s easy to get to the broader notion of deliberate practice. The base layer is still the same. You’re still practicing the skill for 10,000 hours.
It’s the metacognition that is different. Instead of finding creative flow, you seek out ugly awkwardness that nevertheless intrigues and tempts you. You figure out what feels uncomfortable and “wrong” in some sense, but also alluring, and figure out why. There are no judges to tell you if you’re right. There are no aesthetic standards to internalize. There are no performance standards other than what you’ve yourself done before or the behaviors of people you choose to imitate because you can’t think of anything yourself.
And most importantly, there is no clear understanding of whether variation from your own past behavior or others’ behaviors should be considered error or innovation.
Under such conditions, repetition for 10,000 hours becomes important not as a way to achieve perfection, but as a way to get the law of large numbers to work in your favor. Because with some low, but not vanishingly low probability, some of those “errors” can turn into mutations/innovations that can be developed further. Doorways that lead down very generative forks that have never been explored before.
Being open to these “errors” requires that your experience of what you are doing not be completely mediated by symbols. When playing the violin, it is fine to let the symbolic understanding of music performance guide your understanding of the music, but it should not blind you to, for instance, errors that suggest improvised use of violins as weapons or baseball bats.
So disruptive metacognition is irreverent and transgressive. It does not respect received sacred/profane distinctions. It does not justify extended practice on the basis of “pay your dues” but as a means of exploration. It does not seek flow as an end in itself, divorced from the effects of performance. While sustaining metacognition can be whimsical in an approved way, it cannot be offensively playful in the sense of irreverently crossing the boundary separating sacred and profane. Only disruptive metacognition can do that.
If the reward for effective sustaining metacognition is a sense of your own inner sacredness, experienced as flow, the reward for effective disruptive metacognition is a sense of snowballing absurdity and paradox that miraculously does not unravel. Effective awkwardness that inspires irreverent laughter rather than reverent awe. Instead of approval from honored figures, you get the slightly vicious pleasures of desecration.
While it is possible to do this all this in closed worlds of performance, it takes a kind of sociopathy to ignore expert tastes (or refined customer/audience tastes) and willingness to suffer being punished for being genuinely innovative (customers of cultural products punish straying performers much more than other kinds of customers). This is why early rockers shocked classical musical purists by burning or smashing guitars. Of course, you can also shock aging rockers’ sense of the sacred by not being outrageous (“kids today, they have no rebellion in them!”).
In default-open trader domains, where there isn’t a strong a sense of standards and aesthetics to tug you towards an arbitrary direction you can seek out a personal true north. The possibilities of generative, open-ended innovation are significantly higher because it takes less courage to stray.
That’s the open-world version of the dynamic behind 10,000 hours: mimicking evolution by providing a behavioral substrate for random positive mutations to act upon. This leads to good news and bad news.
Disruption as Violent Transgression
The good news is this. Viewed this way, talent becomes much less important than the mere fact of mindful repetition over a long period of time with a sensitive eye alert to the positive mutations to build upon.
The talent which feeds error elimination in closed domains only needs to be good enough to ensure survival in an open domain. Because all you need from the finite game is sustenance to keep the repetition going. You’re not seeking flow, perfection or victory in the finite game. In fact, perfection is the enemy (this is what ‘great is the enemy of good enough’ means). Seeking flow blinds you to the positive mutations you might stumble across.
The bad news is that success still depends on repeating some skilled behavior in roughly the 10,000 hour range, at “good enough” levels, before you’ll start stumbling across mutations that are both good and haven’t been spotted and explored before. This is why “good ideas” that beginners come up with, even if actually good, aren’t worth much. They lack the behavioral base to actually go down the bunny trail opened up by the idea. The have the idea, but not the idea maze. The genetic mutation without the protein synthesis machinery.
But if you do have the disruptive deliberate practice under your belt you can, well, be disruptive.
If you know the basics of disruption theory, you know it involves attacking a market from a marginal niche. I won’t rehash that. But I will state what might be a new point. What’s disruptive about disruption is that it violates a prevailing sense of the sacred with irreverent profanity.
A disruptor attacks a saintly mindset rather than a market. A mindset that holds certain performance standards and aesthetic considerations to be sacred, and is blind to the potential of what it considers profane. The disruptor wins by being mediocre where it is a sacred duty to be exceptional, and embracing profanity where saints are blinded by their own taboos.