Why You Can't Explain What You Read

Main ideas / concepts / lessons

• Smart-sounding nonsense can pass for intelligence

  • In 1996, physicist Alan Sokal set up an academic hoax: he submitted a paper intentionally filled with nonsense to test whether a journal would publish it because it sounded impressive and matched what editors expected.
  • The paper was titled “Transgressing the Boundaries Towards a Transformative Hermeneutics of Quantum Gravity.”
  • The article combined real physics ideas with false claims, unsupported logical leaps, and sentences later admitted to be meaningless.
  • A cultural studies journal, Social Text, published it as part of a special issue on the tension between science and cultural studies.
  • Sokal revealed the hoax the same day, sparking a major controversy.
  • Social Text’s editors argued they weren’t claiming to be doing peer-reviewed science; they viewed it as an unusual contribution meant to engage with their field.
  • Takeaway: how we recognize intelligence can be vulnerable to style, jargon, and signals of “expertise.”

• The reverse problem: genuine intelligence can sound like nonsense

  • The narrator asks: how often do good ideas get communicated so badly that they sound nonsensical?
  • Personal context: the narrator describes struggling to understand John Rawls’ A Theory of Justice, suggesting the difficulty might come from:
    • complexity of the ideas, and/or
    • the author’s ability to explain them clearly (as opposed to merely knowing them).

• Experts often fail to explain because they can’t “un-know”

  • Steven Pinker describes this as the “curse of knowledge”:
    • experts struggle to imagine what novices don’t know,
    • so they skip basic steps unintentionally.

• Evidence for the “curse of knowledge” (song-tapping experiment)

  • A Stanford graduate student, Elizabeth Newton, conducted an experiment later popularized by Chip and Dan Heath (in Made to Stick):
    • One person taps the rhythm of an easy, familiar song (e.g., “Happy Birthday”) for another to guess.
    • The tapper predicts listeners will identify the song about 50% of the time.
    • Actual results: listeners correctly identified only 3 out of 120 songs (about 2.5%).
  • Reason given:
    • the tapper can hear the melody in their head (making the answer feel obvious),
    • the listener hears only disconnected taps without the melody, so the communicated signal is much weaker than the tapper realizes.

• Knowledge changes how people perceive and process information

  • In many domains, experts think in patterns/whole arrangements, not separate pieces.
  • Example: chess
    • beginners: focus on individual pieces and single moves,
    • experts: recognize configurations spanning multiple pieces/moves.
  • Example: reading
    • with practice, decoding (word parts, syntax, grammar, punctuation) becomes automatic,
    • which makes it harder to explain precisely because the process feels invisible even though it’s foundational.

• Another way smart people can sound stupid: needless complexity for status

  • Sometimes writers make language more complicated on purpose (big words/jargon) to appear serious—an assumption the narrator connects to Sokal’s hoax strategy.
  • Daniel Oppenheimer’s experiments:
    • when the same ideas are rewritten with unnecessarily long/complex language, readers judge the writers as less intelligent, not more.

• Clear writing isn’t always easy, but writers still have responsibility

  • The narrator clarifies:
    • some fields require technical language,
    • some ideas truly are complex and take time to learn,
    • readers must also do effort.
  • Still, writers have to meet readers halfway.

• The reader-writer responsibility model (baseball analogy)

  • From Adler and Van Doran’s How to Read a Book:
    • writer = pitcher
    • writing = ball
    • reader = catcher
  • The catcher must be active/attentive.
  • But the pitcher must throw something catchable:
    • if the ball is thrown too far (e.g., 10 feet over the catcher’s head), effort won’t fix it.

• A philosophy of word choice (Faulkner vs. Hemingway)

  • A reported exchange between William Faulkner and Ernest Hemingway:
    • Faulkner criticized Hemingway for avoiding words that require a dictionary.
    • Hemingway replied (quoted): big emotions don’t come from big words; he uses older, simpler, better words.
  • The narrator’s conclusion:
    • you don’t need complicated language to express something deep and meaningful.

Methods / instructions presented (detailed)

• How to think about misunderstanding (frameworks implied by the talk)

  • When an explanation sounds unclear, consider two possibilities:
    • the ideas may be inherently complex, and/or
    • the explainer may be skipping steps due to expertise (curse of knowledge).
  • Evaluate whether the communicator:
    • assumes the audience already has prerequisites,
    • replaces basic steps with jargon because it feels impressive,
    • relies on patterns the expert can “see” but the novice cannot.

• Experiment described to test “curse of knowledge”

  • Setup
    • Choose a very familiar, easy song.
    • Have one participant tap out the rhythm so another can guess.
  • Predictions
    • tappers estimate high listener success (around 50% in the reported setup).
  • Outcome
    • listeners identify the song extremely rarely (~2.5%).
  • Interpretation
    • tappers hear extra internal context (melody) that listeners do not.

• Writing-evaluation experiment described

  • Keep the ideas constant.
  • Change only the language complexity:
    • version A: plain/basic vocabulary,
    • version B: unnecessarily long/complex wording.
  • Measure
    • reader judgments of intelligence.
  • Result
    • complex language versions are judged as less intelligent when ideas are genuinely good.

• Writer/reader interaction model (from How to Read a Book)

  • Reader actions
    • be active,
    • pay attention,
    • exert effort to “catch” the writing.
  • Writer constraints
    • present material that is “catchable,” i.e., appropriately pitched and accessible.
    • don’t throw so far beyond the reader’s reach that effort can’t compensate.

Speakers / sources featured

• Alan Sokal (physicist; author of the hoax)
• John Rawls (author of A Theory of Justice)
• Steven Pinker (author of The Sense of Style; “curse of knowledge”)
• Elizabeth Newton (Stanford graduate student; study described)
• Chip Heath (co-author; popularized the Newton study in Made to Stick)
• Dan Heath (co-author; popularized the Newton study in Made to Stick)
• Daniel Oppenheimer (conducted writing/jargon experiments)
• Adler and Van Doran (authors of How to Read a Book)
• William Faulkner (source of the criticism)
• Ernest Hemingway (source of the reported reply and writing style)
• “Social Text” (the journal that published the Sokal hoax; represented via its editors)