The Value of Crazy Ideas in Science

Minutes

President Collins called the 2154th meeting of the Philosophical Society of Washington to order at 8:20 p.m. on December 6, 2002. The Recording Secretary read the minutes of the 2153rd meeting and they were approved. The speaker for the 2154th meeting was Robert Ehrlich, author and Professor of Physics at George Mason University. The title of his presentation was, “The Value of Crazy Ideas in Science.” Mr. Ehrlich opened his remarks by pointing out that he as not the new governor of Maryland, nor was he the author of the governor's “100 new ideas,” whether they were crazy or not. He then settled into his presentation by pointing out that science often makes great strides when it entertains crazy or unconventional ideas. But, since 99% of new ideas lead nowhere, scientists need ways to judge how fruitful new and sometimes bizarre ideas are likely to be. Similarly, citizens wishing to make informed judgments are in a similar position to scientists. Using the tools of science, they too can decide which of this week's crop of strange ideas are more likely to be correct, without needing to rely on the experts. Mr. Ehrlich described a variety of ideas, and attempted to judge how likely each is to be true. Essentially, he provided a primer in how to evaluate the strength of evidence for a strange-sounding idea, illustrated through a series of case studies. Consider for a moment some successful ideas that were not accepted initially such as the germ theory of illness, or the theory of plate tectonics which grew out of the crazy idea of continental drift. However, not all theories can be proven right or wrong. Some theories are simply not testable such as the theory that rocks have feelings that they can't communicate, faster than light particles or tachyons exist, but they do not interact with ordinary matter, and lastly, the world is only about 5,000 years old, but it was created to look like it's billions of years old. These are examples of some crazy ideas and of some nutty ideas. The latter usually involve internal inconsistencies, vast conspiracies, violation of known physical laws or require empirically false assumptions. Some examples of the latter category include astrology, no lunar landing and time travel to change the past. The speaker described some checks to use when assessing new ideas as being in the crazy ideas column or the nutty idea column. A first check might be to assess the background and qualifications of the person offering the new idea. Use care in implementing this check and keep in mind that Einstein was a patent clerk not a scientist when he developed the theory of relativity. Other checks to show a crazy idea is actually false relate to the method used to present the idea. Consider the care taken by the idea proposer in the use of statistics, and the number of references to the work of others. In the case of statistics recall the remarks of Benjamin Disraeli, “There are three kinds of lies: lies, damn lies and statistics.” An example of an interesting statistical tool that allows one lie convincingly is the idea of “informed choice.” Here, the idea proposer choses a subset out of total data set to enhance the statistical significance of his claim. It's also called “cooking the data.” Mr Ehrlich then offered examples of some current crazy ideas in science and his assessment of the ideas with a subjective “cuckoo” rating scheme. This is a quantitative rating system using a scale of 0 to 4 “cuckoos” based on a subjective assessment of an idea, where zero cuckoos means “why not” and the idea is substantiated by data, up to four cuckoos which means the idea is “certainly false” and the idea is not substantiated by data. The first idea described is, “More guns mean less crime.” Mr. Ehrlich rated this idea at three cuckoos which may make gun control enthusiasts happy. However, they should not rejoice too soon because Mr. Ehrlich also concluded that the rating for the opposite idea, “More guns increase crime” is only slightly smaller at two cuckoos. The next idea discussed was, “AIDS is not caused by HIV,” is rated at three cuckoos. However, the idea of “Sun exposure is beneficial” was considered by Mr. Ehrlich to be plausible and rated it at zero cuckoos. The idea that “Nuclear radiation is beneficial” is less likely to be true and is rated at one cuckoo. Another crazy idea is “Oil, coal and gas are abiogenic in origin, in other words they are not fossil fuels” is considered by Mr. Ehrlich to be plausible and he assigns it a rating of zero cuckoos. The idea that “Time travel is possible” is rated at two cuckoos. On the other hand, the crazy idea “Faster than light particles are possible,” is considered by Mr. Ehrlich to be plausible and rated at zero cuckoos. Lastly, the really crazy idea that there was no big bang is considered to be not true and is thus rated at three cuckoos. Mr. Ehrlich then closed his presentation and kindly answered questions from the floor. President Collins thereupon thanked Mr. Ehrlich for the Society, welcomed him to membership, announced the next meeting and made the usual parking announcement. He then adjourned the 2154th meeting to the 132nd Annual Meeting of the Philosophical Society of Washington at 9:25 p.m. Attendance: 52 Temperature: 16°C Weather: Clear Links: George Mason University: Bob Ehrlich Biography Ehrlich, Robert, “Nine Crazy Ideas in Science: A Few Might Even Be True,” Princeton University Press, May 2001 (available at Amazon Books) Respectfully submitted, Bill Spargo Recording Secretary