Case Study: The Five-Year Difference — When Faster Correction Saves Lives

The Calculation

The asymmetry of correction is not abstract. It can be estimated — roughly, imprecisely, but meaningfully — for specific cases.

H. pylori: 30 Years vs. 5 Years

Marshall and Warren demonstrated the bacterial cause of peptic ulcers in the early 1980s. Full medical acceptance came in the mid-1990s — approximately 15 years later (with the Nobel Prize in 2005, more than 20 years later).

During those 15 years of delayed correction: - Millions of patients received acid-suppression therapy that managed symptoms without curing the disease - Hundreds of thousands underwent unnecessary surgeries (vagotomies, partial gastrectomies) - An unknown number developed gastric cancer from chronic untreated H. pylori infection

If the correction had taken 5 years instead of 15, the unnecessary suffering would have been reduced by approximately two-thirds. The surgeries, the chronic medication, the preventable cancers — all scaled with the duration of the wrong consensus.

Forensic Science: Ongoing

The NAS report on forensic science was published in 2009 — over 15 years ago. Many of its recommendations have not been implemented. Unvalidated forensic techniques continue to be used in courts. People continue to be convicted based on evidence that has no scientific basis.

Every year of continued delay adds to the total: additional wrongful convictions, additional years of imprisonment for innocent people, additional erosion of public trust in the justice system.

The General Principle

For any wrong consensus:

Total cost = (annual cost of being wrong) x (number of years the error persists)

The tools in this book target the second factor — the duration. They cannot prevent all errors (that would require perfect knowledge, which is impossible). But they can reduce the time between error and correction — and the human cost scales directly with that time.

The Motivational Frame

This calculation is the answer to the nihilist objection — "if we're always wrong about something, why bother trying to be right?"

Because the speed of correction matters. A field that corrects in 5 years is not perfect, but it is dramatically better than a field that corrects in 50 years. The difference is measured not in philosophical satisfaction but in human welfare.

Every Red Flag question asked, every Epistemic Health dimension scored, every dissent strategy designed, every institutional reform proposed — each one is a potential input to faster correction. And faster correction saves lives.

Analysis Questions

1. Estimate the asymmetry of correction for one case in your own field. How long has a known (or suspected) error persisted? What is the approximate annual cost? What is the total cost of the delay?

2. Apply the Correction Speed Model (Chapter 22) to the same case. Which variable, if changed, would produce the largest reduction in correction time? What would it take to change that variable?

3. The chapter argues that "less wrong" is sufficient. But is there a threshold below which "less wrong" is not meaningfully different from "wrong"? When does partial correction fail to produce meaningful benefit?