I have written about Sherman Kent for years as someone who grappled with uncertainty and the language we use to discuss it. There is imprecision in our words, such as “likely” and “probable.” Sherman Kent was a professor at Yale University who was called to co-head the CIA’s National Office of Estimates and improve their forecasting skill. Getting the chance of a bad event wrong has real effects. During his tenure, Kent wrote an essential piece on using ambiguous words to describe our probable estimates. He worked to end squishy vague language that provided political cover for assessment authors.
It now looks like Sherman is getting his due with a recent article in the Harvard Business Review, “If You Say Something Is “Likely,” How Likely Do People Think It Is?” The authors of the HBR piece extended the sample used by Kent to show the attached probability for words. While many may agree on relative probity weights, language has vagueness, and that has not changed over the last few decades. So if you don’t believe me, you can take a survey and see how your assessment of probabilities associated with words differs from others.
The conclusions we can draw from all of this work are important:
- Words have imprecise meaning and can lead to sloppy thinking;
- Precision in language is critical to make the right decisions;
- Don’t use words when numbers are needed.
- Precision through probabilities will not solve forecasting errors but will eliminate ambiguity.
- Quantitative thinking is more running statistical models – It is about learning to be systematic and precise in decision-making.
For more on Sherman Kent, see: