Hacking+Book+Review

Adam Hagelgans Cultures of Scientific Revolutions Review: The Emergence of Probability, Ian Hacking

Probability is a concept that comes up quite regularly in our daily lives. We wake up and check the weather outside. If we are in Troy then it is probably going to be cloudy and chilly. We take the chance and prepare for what seems to be most probable to occur. The meteorologist gives us a percentage chance of rain each day. Insurance companies use it to figure out your premium, based off the chances of you having an accident. Probability is engrained into everyone’s lives but few realize it. The Emergence of Probability by Ian Hacking walks us through the beginning days of the science and how it was developed. First off this book is not for the non-mathematically inclined. I took AP Statistics & Probability in high school and most of the math discussed in the book was light years beyond that. There are many pieces of probability that you can take into your daily life to improve it but there is also lots of boring material inside. Hacking goes through the process that probability “emerged” and that was a very bumpy road. It was far from an easy read. Time and time again, I found myself wondering what I had just read on the page before. Hacking broke the book into chapters based off the concepts of probability and statistics which makes it nice if you are interested in a particular subject. Some people would say that probability is the relationship between a hypotheses and the evidence supporting it. The more evidence that you can collect supporting a hypothesis, the higher probability there is that it is true or will happen. Hacking says that “Evidence, in my usage, is a matter of inferring one thing from another thing…The kind of evidence that I have in mind consists in one thing pointing beyond itself. This must be further clarified. It is non-deductive pointing. A single observation that is inconsistent with some generalization points to the falsehood of the generalization and thereby ‘points beyond itself.’” Lorenzo Valla managed to expose the Donation of Constantine without any of the aforementioned evidence. He exposed it just by claiming that one word “satrap” would never have been used by Constantine. The “low sciences”, defined by Hacking to be alchemy, geology, astrology, and medicine, were forced to use internal evidence because there was no way scientists could perform demonstrations for the public. In order to gain the public’s approval, they had to use the evidence of things, by showing the results that came from their work. “No particular event, no single student, is responsible for the emergence of probability” says Hacking. The first people to study probability were also very proficient in many other subjects. Only the wealthy got real educations so the educated just hopped around from subject to subject. “Cardano, the author of the first book on probability, was famed for both his skill in medicine and his talent at mathematics.” In present time, it is common for a person to specialize in one subject, certainly not two as diverse as medicine and math. Some of the greatest advances in history came about because of the scientists’ breadth of knowledge. The bringing together of different, seemingly unrelated ideas is what really created probability. Decision theory, combinations, logic, and statistics all came to hold a piece of probability. “ Decision theory is the theory of deciding what to do when it is uncertain what will happen.” Before the Renaissance and even during it, probable meant approved by the wise. In the context in which probable is used today, I suppose it was used because it was the most likely to occur. It would like the people who set Vegas’ betting lines today. Probability was first used to denote something measureable in 1662 in the Port Royal Logic. “In [1713] probability came before the public with a brilliant portent of all the things we know about it now: its mathematical profundity, its unbounded practical applications, its squirming duality, and its constant invitation for philosophizing. Probability had fully emerged.” Making a decision is a lot easier when you bring probability into the fold because it places a rough estimate on the uncertainty. A basic example of decision theory would be choosing which horse will win the Kentucky Derby. When all the horses are standing at the gate, an ignorant observer has no idea which will win. Probability brings into consideration many facts and can predict which has the highest chance of winning, based primarily off of past history and evidence. It is not an exact science by any means but it is sure a hell of a lot better than taking a wild stab at guessing. Hacking brings makes a good example, made by Blaise Pascal, of decision theory with religion. “But if God exists, then wagering that there is no God brings damnation. Wagering that God exists can bring salvation; Salvation is better than damnation. Hence the wager, ‘God is’ dominates the wage ‘He is not.’ He is trying to show us that you must weigh the consequences of being wrong when you are weighing a situation. Daniel Huet partially refutes this statement by pointing out that it can cover any religion and just shows the need for a belief in God, but in any particular one. Raymond Lulle is cited in the book as the founder of the theory of combinations. He wanted to show “all the elements of the world by their true signs and then by generating all possible combination of signs, to produce true signs for all possible compounds in the universe.” This statement is representative of the great wording Ian Hacking uses. Throughout the book he referred to true signs but I could not comprehend just what he was trying to get at. The theory of combinations does play a big part in probability. Elementary probability problems involve you finding the number of certain situations in a larger group. Like if you were to flip a coin five times, what are the chances of getting 2 heads? You have to calculate the number of combinations to find your chances. Quantitative considerations are addressed by Hacking in a chapter. These are where it is safe to consider a certain outcome based off sheer numbers. If a certain outcome occurs 999 times out of 1000 then without knowing anything else about the situation, we can safely say that it is “incomparably more probable that [it] is one of the 999 rather than the single one of the 1000.” St. Augustine points out that the above logic also applies to the topic of miracles. “It is impossible that a judicious man would have attempted to lie about such a public matter, since many people would have found out his lie and so brought nothing but disgrace on the Christian religion.” Statistics began as London began collecting data on births and deaths weekly. It is very likely that the data was far from accurate but with this store of data, they began to manipulate it. John Graunt tried to extrapolate the weekly trends into the population of London. The standard estimate without statistics was around 2 million. Graunt inferred the number of women of child bearing age by fertility charts from the number of births, number of families from that point and then onto the whole population by mean family size. His predicted population of London was 384,000, quite a bit different from 2 million. We may not know exactly how many people actually lived there but it is safe to say Graunt erred on the side of caution. The Emergence of Probability by Ian Hacking is a great book on the history of statistics but nowhere near as interesting as I thought/hoped it would be. I probably (no pun intended) would not recommend this book to anyone whose company I enjoy. The history of probability and statistics is one filled with twists and turns which, if it is your field, would make for a great read. Ian Hacking is a thought provoking brilliant man, just not an enjoyable read for me at this time.