EconTalk — David Epstein

On: Mastery, Specialization, and Range.

Episode: N/A

Date: May 2019

Background: Journalist, author of “The Sports Gene” and “Range“.

Key Subjects:

  • Paths to performance: (early) specialization versus generalization.
    • Need specialists (to advance knowledge), but also generalists (to integrate knowledge).
    • Society tends to overvalue specialists and undervalue generalists.
    • People are forced to specialize too early (education/work).
      • Forced to make early decision about matching your abilities/interests and work.
      • Underestimate how much personality and environment change over time.
    • May benefit from early generalization:
      • Learn who you are in practice, not theory: experiment in order to find best match.
      • Learn how to think, evaluate new information, what is true.
  • Benefits and costs:
    • Specialization: head-start, narrow deep knowledge, limited tool set (the inside view).
    • Generalization: slow-start, sampling, wide range of skills and tools (the outside view).
  • Match with internal traits:
    • Match abilities and interests: if you don’t know, sample / experiment (generalization).
    • Match personality: curious, open-minded = generalization, etc.
  • Match with external environment:
    • Kind: stable, recurring patterns, immediate feedback = instinctive pattern recognition = specialization.
    • Wicked: unpredictable, delayed feedback = apply wide variety of patterns = generalization
  • Learning environments:
    • Kind: increasing automation of knowledge accumulation.
    • Wicked: increasing access to knowledge across domains through better communication.
  • Applied to organizations:
    • Calibrate to culture to minimize two different types of errors:
      • Mindless conformity: follow procedures and use the tools no matter what.
      • Reckless deviance: never following procedures, improvise.
  • Flynn effect:
    • Increase in IQ scores, specifically the more abstract section of IQ tests.
    • Adapted to current environmental requirements:
      • Need to transfer knowledge from one domain to the next as we handle new things.
      • Higher requirement to think in abstract classifications.
  • Context dependent.
    • Limits of specialization:
      • Limits of reductionism: understanding the parts, but not the whole.
      • Surrogate markers: solving only parts a problem, not solving the whole problem.
    • Increasing opportunities to apply knowledge (patterns, solutions) to different domains.

Key Takeaways:

  • Limits of specialization are mirrored in the limits to reductionism 
  • Opportunities in applying existing knowledge to different domains to counter limits of specialization / understanding complexity.

Worth Listening:

Clear and systematic run through the book (Range).


Leave a Reply