More or Less – "How likely is ‘likely’?"

Podcast: More or Less, BBC Radio 4
Host: Charlotte McDonald
Date: April 4, 2026
Guest: Adam Kucharski, Professor of Epidemiology and author of The Uncertain Science of Certainty

Episode Overview

This episode delves into the ambiguity of probability language—words like "likely" or "possible" that often crop up in news, intelligence reports, and everyday conversation. Host Charlotte McDonald and epidemiologist Adam Kucharski explore how these words are interpreted by experts and the general public, and why miscommunication around them can have real-world consequences.

Key Discussion Points & Insights

1. The Problem with Probabilistic Language

• Charlotte explains that terms like "likely," "probable," or "possible" are used to convey probability but are inherently ambiguous. People's interpretations of these words often vary widely.
  "Do you know how likely likely is and what probability probable is meant to convey? In some cases, it seems you probably don't." (01:58)

• Adam Kucharski defines probabilities in numeric terms:
  "If there's 10 events, all of which that have an 80% chance of happening […] we'd expect eight of those 10 to happen. If it's a 10% chance, we'd only expect on average one of those 10 events to actually occur." (03:10)

2. The History of Ambiguity: Sherman Kent and the CIA

• Adam recounts a famous incident from 1951, when CIA analyst Sherman Kent wrote of a "serious possibility" of Soviet action and realized colleagues had wildly different ideas of what that meant numerically.
  "He […] realised that even in the team who are co writing these things, they had very different numbers in mind for the kind of language they were using." (04:08)

• This ambiguity persisted for decades until recent efforts to standardize the language, especially in intelligence settings.

3. Modern Efforts to Standardize Probability Terms

• Intelligence agencies have adopted "probability yardsticks":

  • "Highly likely" means an 80-90% chance.
  • "Realistic possibility" indicates a 40-50% chance. (04:54)

• Despite these efforts, Charlotte points out that these phrases now frequently show up in public domains but lack consistency in interpretation.

4. Kucharski’s Large-Scale Probability Language Quiz

• Adam developed a quiz to test the public's understanding of these words, with over 5,000 respondents.

• The quiz had two parts:

  1. Comparing pairs of phrases to judge which implies higher probability.
  2. Assigning a numerical probability to a single phrase, e.g. "may happen." (05:52)

• Findings:

  • "The most consistent was the phrase 'about even', which you'd hope people would land on about 50%." (06:33)
  • High clarity for:
    • "Will happen" ≈ 100%
    • "Almost certain" ≈ 95%
    • "Almost no chance" = 1-5%
    • "Remote chance" and "highly unlikely" = 1-10%
  • Ambiguous phrases:
    • "Realistic possibility" had the widest spread, interpreted as anywhere from 10% to nearly 100%, average just over 50%. (07:42)
    • "Probable" and "likely": broad range (50%-100%), means just under 75%. (08:00)

• Charlotte demonstrates with a mini-quiz:

  • "Which is more likely: 'might happen' or 'could happen'?"—most people did not answer consistently. (08:54)

5. Sociodemographic Patterns

• Adam notes minor trends from the data:
  • More educated respondents were slightly more pessimistic.
  • US respondents were marginally more optimistic than UK respondents.
  • However, these were small effects; ambiguity was the dominant theme. (09:17)

6. Addressing Ambiguity: What Should Be Done?

• Adam’s recommendations:
  • "Just ensuring that what the person coming up with an estimate is saying is what the person is hearing at the other end." (10:00)
  • Avoid hiding behind vague words to dodge accountability:
    • "All of us, I think, can have some benefit in actually challenging ourselves to make a judgment that we can really go back and ask if we genuinely got that right, rather than having these kind of ambiguous words where we can always wriggle out of it later if we want." (10:15)

Notable Quotes & Memorable Moments

• On the crux of the problem:

  “Do you know how likely likely is […] In some cases, it seems you probably don't.”
  — Charlotte MacDonald, 01:58

• On historical confusion in intelligence:

  “Even in the team who are co writing these things, they had very different numbers in mind for the kind of language they were using.”
  — Adam Kucharski, 04:08

• On the ambiguity of ‘realistic possibility’:

  “It really does seem like a word that means a lot of different things to a lot of different people.”
  — Adam Kucharski, 07:56

• On accountability in communication:

  “All of us, I think, can have some benefit in actually challenging ourselves to make a judgment that we can really go back and ask if we genuinely got that right, rather than having these kind of ambiguous words where we can always wriggle out of it.”
  — Adam Kucharski, 10:15

Important Timestamps

• [01:48] — Episode begins; Charlotte introduces topic of ambiguous probability language in news and reports.
• [03:10] — Adam explains the basics of probabilistic thinking using numeric examples.
• [04:03] — Story of Sherman Kent and the “serious possibility” CIA report.
• [04:54] — Intelligence agencies introduce “probability yardsticks.”
• [05:52] — Adam describes his large-scale public quiz.
• [06:33] — Results: clarity and ambiguity in phrase interpretations.
• [07:31] — “Realistic possibility” as the most variable phrase.
• [08:54] — Charlotte’s on-air quiz: “might happen” vs. “could happen.”
• [09:17] — Demographic patterns in responses.
• [10:00] — Adam’s advice for communicators of risk.
• [10:15] — The importance of accountability in using probabilistic language.

Summary

This episode underscores how words intended to convey risk or likelihood are open to wide interpretation and miscommunication, sometimes with critical implications. Adam Kucharski’s research reveals that while some probability phrases are well-understood, many—including commonly used ones like "likely" or "realistic possibility"—are not. The hosts urge greater precision and accountability in communicating risk, especially by experts, to ensure vital information is clear and actionable.