Clear and Precise Scientific Communication

Because it is New Year’s Eve, I indulged in some Twitter. One exchange might be worth unscrambling from some others. Dani Rodrik triggered it with a post that offered this advice to non-economists:

Do not let math scare you; economists use math not because they are smart, but because they are not smart enough.

Lukas Freund responded on Twitter:

Agree on many points, which in classic Rodrik-fashion are pithy but insightful. But e.g. “economists use math not because they are smart, but because they are not smart enough” sounds like a misleading maths-bashing generalisation. — Lukas Freund (@LukasFreund) December 31, 2017

I read it as praising math, not maths-bashing. He’s saying we need the help that math can give us? — Nick Rowe (@MacRoweNick) December 31, 2017

Agree with that as a conclusion – math aids economists (and others) in clarifying and expressing their thoughts succinctly & prevents unwarranted jumps in reasoning – but it doesn’t strike me as the most natural reading. Glad you flagged the alternative reading! — Lukas Freund (@LukasFreund) December 31, 2017

Soren Kerndrup and Lukas referred to my work. (Note that Soren meant the “Warsh” book.)

Romer’s ( @paulmromer ) work on growth theory one of the best examples on how math is important “in clarifying and expressing thoughts succinctly & prevents unwarranted jumps in reasoning ” see david harsh book @rodrikdani — Soeren Kerndrup (@kerndrup) December 31, 2017

…which is why his ‘mathiness’ critique has weight, of course. I don’t think anyone doubts that math can be useful — problem is disconnect of natural language and policy conclusions from math, and that critical treatment of math not sufficiently taught in grad school. — Lukas Freund (@LukasFreund) December 31, 2017

I jointed in with 4 tweets (combined here).

1/ The test is whether math adds to or detracts from clarity and precision. A writer can use either the words of everyday language or the symbols from math to make assertions that are clear and precise; or opaque and vague.

2/ The deep problem is intent, not ability or skill.

3/ Writers who want to make predictions use words and math to be clear and precise. Writers who want to make excuses use words and math to be opaque and vague.

4/ Compared to words, math and code tend to be both more precise and more opaque. See, e.g. the international obfuscated C contest.
— Paul Romer (@paulmromer)

Lukas responded with two tweets, that judging from likes and retweets, were the most widely appreciated of the entire exchange.

Put differently, I suppose, we need a version of Orwell’s “6 Rules for Writing” for Economics, perhaps including the following:

  1. Never use mathematical symbols where it does not add to clarity and precision relative to natural language.
  2. Never use math unless you intend to use it for sake of clarity and precision.
  3. Never provide policy recommendations based on a mathematical model unless you understand what assumptions underpin that recommendation.
    — Lukas Freund (@LukasFreund)

By this point we were out of sync. In parallel, Lukas took issue with my use of “intent”, and I accepted his point.

Intent plus, I would say, career concerns and the pressure to publish papers of a particular character, optimally with strong policy conclusions. This might, of course, be purely due to small-n sample of economists I’ve interacted with, but those I’ve met generally do have ‘good intentions’ (to produce research developing policies that promote public welfare) – suggests factors other than intentions are at work. — Lukas Freund (@LukasFreund)

Perhaps I should say “problem of attention” rather than “problem of intention.” But either way, worth emphasizing that clarity and precision take effort, commitment, and self-awareness. — Paul Romer (@paulmromer)

I suppose “problem of attention” also makes a defensive gut reaction by certain economists less likely…

One of the difficulties is that math-skills can be taught, whereas “commitment, and self-awareness” are more fundamental — and are perhaps best instilled by example. — Lukas Freund (@LukasFreund)

One place to start is Do I Make Myself Clear by Harry Evans, the greatest living editor of the English language: “This book on clear writing is … concerned with how words confuse and mislead, with or without malice aforethought …” — Paul Romer (@paulmromer)

Meanwhile, in another attempt at steering the discussion toward all writing, whether in words or math, I wrote:

More generally:

  1. When an assertion is vague, ask for specifics.
  2. When an assertion is opaque, assume that it is the author’s fault, not yours, and ask for clarification.
    — Paul Romer (@paulmromer)

As I put this thread together, I went to Dani’s blog to get the quote that set the discussion in motion. Then I had pleasing sense that we had closed the circle. Dani’s advice, that non-economists should not to be scared by math, is close in spirit to my closing point, that readers should insist that author’s be clear.

Along the way, Lukas, took a chance on online humor and it seemed to turn out ok.

Perhaps also 4. Never write such that “and” accounts for more than 2.6% of the text? 😉 — Lukas Freund (@LukasFreund)

Or, use assertions about the frequency of “and” as an attention getting gimmick. 🙂 — Paul Romer (@paulmromer)

Happy New Year!