The Half-Life of Facts is a provocative book that makes us realize that a lot of facts that we believe are factually true get overturned over time. It gives several examples of facts that were later proven wrong or became more precise as our measurements and understanding improved.

Half-life of Facts

Knowledge, like radioactivity, decays over time. In aggregate, we can talk about the half-life of the decay of knowledge.

  1. From 1912 to 1953, it was believed that humans had 48 chromosomes. Then the improved experimentation reduced it to 46!
  2. Doctors used to recommend smoking. Now, it is deadly!
  3. Meat used to be good to eat. Then became bad. And now, it is a matter of opinion.

A lot of scientific knowledge, especially that’s based on measurement, is meso-fact. It slowly changes over decades and centuries. For example, the number of elements in the periodic table. Or the number of planets in the solar system. The half-life of books in various fields ranges from ~13 years (Physics) to ~7 years (History).

Sometimes, as our technology improves, so do associated facts. For example, the thermal conductivity of Iron changed with our ability to produce a more pure form of Iron and our ability to measure its thermal conductivity more accurately. The Earth was flat then it became a sphere (~8 inches/mile curvature) and now it’s a spheroid (~7.97-8.03 inches/mile).

Measuring workers’ productivity leads to an increase in workers’ productivity! This is known as the Hawthorne Effect. This makes it harder to decide whether an intervention was genuinely beneficial.

Population plays an important role in discoveries and inventions. When the land bridge between Tasmania and Australia broke ~10,000 years ago, the small population of Tasmania slowly degraded technically over time.

Over time, the speed at which information spreads has gone up as well. In 1812, two weeks after the war between the United States and the British was over, the battle of New Orleans started as the news of the first war hadn’t reached New Orleans. In 1860, it took one week for California to learn about Lincoln’s election.

Sometimes, when the facts permeate society, it becomes harder to update them later. For example, the stereotypical dinosaur is known as Brontosaurus. It has now been linked to Apatosaurus. So, the name Brontosaurus is no longer the valid one but that’s what the public knows at large. Similarly, a calculation mistake made Spinach appear to have 35 mg/100 gm Iron even though it only has 3.5 mg/100 gm Iron, this mistake was corrected by the 1930s, but Spinach’s popular perception remains.

There is a lot of knowledge that is public but hidden. Scientists don’t know about it. Many ideas like “rich get richer” have been published many times in academic journals in unrelated fields such as preferential attached, Gibrat Law, and Matthew Effect. In 1988, a large medical demonstrated that Streptokinase treats heart attacks. However, if one looks at cumulative data starting from 1959, then by 1973 the same fact could have been established. However, most scientists were not combining their results with existing results as they weren’t even fully aware of them.

Facts go through phase transitions where they change rapidly. In 1953, by plotting past speed increases, US Air Force predicted that speed faster than Earth’s escape velocity was possible by 1957. And 1957 is when Sputnik went into orbit. So, such state transitions even though abrupt can be predicted from the past data. Looking at the data, the author predicted in 2010 that the first habitable planet will be discovered by 2013. It was discovered in 2010. Similarly, the author predicts that the P versus NP problem will be 53 years old in 2024 and has a high chance of being solved by then.

As our ability to measure improves, some facts can disappear entirely. There used to be Planet X, a placeholder for a planet that should have existed in the solar system. Calculated estimates of its mass kept going down. In 1901, it was 9 times the size of Earth, then 5 times (1909) then 2 times (1919). Over time, the consensus changed to the belief that planet X does not exist. This is a well-known issue in many fields and is known as The Decline Effect. Apart from better measurements, another factor that contributes to the decline effect is Publication Bias. “Statistics in the science of doing 20 experiments a year and publishing one false one in Nature”, is not merely a statement, it has been widely accepted that most research findings are false. Small studies have a higher likelihood of false outcomes. Another such bias that happens in Biology is Taxonomic bias, which is the bias towards studying creatures that humans are interested in or are easier to find.

Humans suffer from factual inertia, sometimes just to appear contrarian, and sometimes out of desire. Change blindness prevents us from noticing changes that might invalidate the facts that we know. One way to avoid this is to rely less on your memory and more on the Internet to find up-to-date information when required.