Experts vs Trial and Error

Writing in the Wall Street Journal, Nina Teicholz casts doubt on the ‘conventional wisdom’ that saturated fat causes heart disease (thanks to The Pretense of Knowledge for the pointer).

Of course, Gary Taubes laid out much of the same story line in his book, Good Calories, Bad Calories. I mention it here and here.

Teicholz mentions President Eisenhower’s heart attack. She didn’t mention the additional detail that Taubes provided. His doctor cut his cholesterol intake and his cholesterol levels went up.

Teicholz, perhaps, summarizes the beginning of the Type II diabetic and obesity trends when unreliable health studies were used to guide the American diet:

As Harvard nutrition professor Mark Hegsted said in 1977, after successfully persuading the U.S. Senate to recommend Dr. Keys’s diet for the entire nation, the question wasn’t whether Americans should change their diets, but why not? Important benefits could be expected, he argued. And the risks? “None can be identified,” he said.

This is where I’ve gained much appreciation for what Nassim Taleb identified as the expert problem, as he describes here.

Enhanced by Zemanta

6 thoughts on “Experts vs Trial and Error

  1. I liked the post discussing Taleb’s criticism of the “Expert Problem”. It seems to me that reliance on experts and academics, is by definition unscientific. Unfortunately, our nature is to seek short-cuts — so reliance on experts allows us to avoid the rigor necessary to become proficient in many topics.

  2. Also, I dislike the way published studies jump from “increased risk factor” to “cause”.

    For example: smoking does not CAUSE lung cancer, it seems to increase the risk factor. For it to be a CAUSE, everybody who smoked would get lung cancer, but there are numerous cases where that doesn’t happen.

    From everything I’ve read, the causes and results of elevated serum cholesterol reside in a tangled web. Some parts are partially understood, but others are still a mystery. One part I thought had been cleared up though is that there is no link between dietary cholesterol and serum cholesterol. Test subjects eating vast amounts of dietary cholesterol showed no statistical increase in serum cholesterol, but did show elevated testosterone levels.

    Hmmm… Better go eat some eggs. 😉

    • That’s a pet peeve of mind too. It amazes me how researchers let their work be misrepresented and don’t try too hard to make sure the caveats or potential problems are highlighted.

      Incentives matter, even with experts. Some like noteriety more than the truth.

  3. Experts – The don’t know (in Taleb’s sense). They can’t know (in Hayek’s sense). And even if the did know, they would lack the incentives to act on that knowledge

  4. I had another thought. James “The Amazing” Randi used to say that experts were the easiest to fool. As a magician, he had lots of experience to back up the idea, and used it to explain the many brilliant scientists and academics who bought into Uri Geller’s parlor tricks.

    There’s a further trap of granting an audience to experts when they discuss areas in which they have no expertise. The media does that crap all the time, drives me nuts.

  5. Pingback: A good synapses of how we were duped into getting fat | Our Dinner Table


Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s