Whether it be number crunching or a gut feel, having a reason for predicting the future so that one can take a position is what conjecture is all about. Placing a probability on a future event and having a knowledge of its magnitude permits "putting your money where your mouth is".
James Franklin has written a book that has been put together backwards. His Review of Work on Probability before 1660 which appears at the tail-end of his book, is actually the justification for his having produced this very well composed and thoughtful tome. In the "Review" he points out that the sole reason (and a very good one indeed) is that a previous book by Ian Hacking which was titled, The Emergence of Probability, raised many objections from the learned ones whom Franklin lists in great detail, along with their objections. Then, Franklin sets out to justify chapter by chapter and point by point, his conceptual image of the Science of Conjecture. This he called the Appendix, but let's call it Chapter I.
Then, in the Epilogue, he gets into the meat of the argument that this is all about probability. That is to say, what's the best guess that can be made given the evidence the judge (or scientist) has at hand. And with this he discusses a bit of the "logic" which was introduced in the mid - 1660s. This he calls "The Survival of Unquantified Probability"; lets just call it Chapter II.
Now we are ready for "Soft Science and History", and here we get a nice bit of head-feeling to determine the lumps and bumps that come along when you start to try to make educated guesses with the most bare of evidence. But it can be, will be, and is done!
Statistics can be introduced to show that the value of number collection is to aid the researcher in weighing past information so that a reasoned answer can be arrived at and perhaps give insight into the future(?). With this in mind it is necessary to set up ground rules so that the numbers have some actual value. As example, if you have a nice thermometer and you take measurement of the temperature of boiling water and that of water near freezing; and then average the value you have an estimate of what the range in temperature might be. Woe be unto you if you have no knowledge of whether its Fahrenheit or Centigrade (Celsius) that the device measures. Take the average of 100 degrees (that of boiling water) and 0 degrees (that of freezing water) and you calculate a value of 50 degrees. Not an altogether unpleasant temperature, just a sweater will do if you are thinking in Fahrenheit. But as you were measuring with a Celsius yardstick, that is equivalent to 122 degrees Fahrenheit, about what you get in the desert Southwest on occasion. If you happen to be using a Fahrenheit thermometer, which is what most of the America's use, then you have 212 degrees for boiling water and 32 degrees for ice water. Averaging gives a value of 122 degrees (same number as before- surprise), but if you are thinking in Celsius, then 122 is way above the boiling point of water and you die! Call it averaging, regression, correlation, least squares, or any of the other acceptable names for putting the past findings to paper and you have arrived at the use of numeric logic to arrive at a finite decision, but do be careful!
Missing from this chapter (and the book as well) are attempts of individuals, most notably Sir Thomas Browne who took to task those who thought that all knowledge had already been arrived at and it was just the matter of shuffling the facts (deck) to arrive at new conclusions. Browne in Pseudodoxia Epidemica, cited errors that were accepted by scientist of the day. What Browne and others were laying the ground work for, was experimenting to determine facts. One has to remember that this is the time of the Plague, Great Fire in London, Charles II reclaiming the throne, removal of Cromwell and all his effects, the Restoration, the great poets (Butler, Milton, etc.), beginning of medicine and departing from bleeding and materia medica. There is much meat in this section and while lacking much, it should be Chapter III.
Ready for Philosophy: Action and Induction? Then we are ready to proceed into the dark abys of the church (Any church will do, but Franklin writes about the Catholic Church.) Religion has to have a moral code and what better than to give it one based on the written word. And of course a few pagan observations help to build a creed. Here then begins the structuring of a decision based on the best evidence available or upon the best authority that can be cited. The parallel between this stage and current acceptance of scientific dogma is pronounced. Not only is acceptance demanded, but the infidels who dance to a different drummer, while not put to the stake, are relegated to an even more horrible fate. They are placed in exile among their own.
Examples from today's ventures into the dark world of science:
1) the controversy in Utah regarding "nuclear" reactions in a test-tube, or
2) "Biological Transformations" by Kervran
3) abiotic formation of gas and oil may not be all that unlikely after all, or
4) perhaps more recently the "hubbub over bubble fusion".
There are lots of others.
Graunt's work in deciphering the magnitude of the Plague of 1340, or so and Ockham's "KISS", i.e., keep it simple stupid or as it is called when a shade of antiquity is desired, the "Razor" are early example of the appreciation of the value of numbers. Of course, Franklin correctly points out that there was a bit of, shall we call it plagiarism in those days since, Aristotle and others probably said it in different ways in Greek or Latin or whatever (pp 141).
This avenue was actively pursued by Herbert Spencer in the 1880's who firmly believed that he had a direct line to God in understanding the Universe and all it contained. Others have had the same feeling, but disqualified Spencer since after all he was of philosophical bend and not a "true" scientist.. Regardless, let's just call this Chapter IV and V of Franklin's book..
"Aleatory Contracts" didn't ring a bell, so checking with Webster, I find that Franklin meant to say, "gambling". In this he walks us through risk taking which is what one either does or does not do. The object of money lenders is to place bets in their favor. So how to do this? Figure the odds of failure. Is this any different from playing with dice to which a bit of lead or mercury has been added to give the owner a bit of an edge? This has a very modern application as there is a scramble by insurers to find ways of estimating their exposure to risk from suicide and terrorist bombers and such. (See the Wall Street Journal, April 8, 2002.) In today's terms, "Many insurers are convinced that terrorism insurance could be profitable if the risk can be quantified... But such estimates are little more than guesses." (Not even educated guesses!)
Since it makes little difference whether the bet is on a wheel, die, boats afloat, outcome of war, etc., this is good stuff and deserves to be Chapter VI and VII.
Which now brings us to the really interesting things that Franklin has to contribute. Ancient Law. Well, it's not really ancient law since it's the justice we see meted out in our courts everyday. But here he goes into his view of history and the way in which different notables were able to be swayed in their decision. No better is he than when he describes the putting together of several partial truths to yield a half truth which then can be used to decide the fate of a hapless victim. Drowning, can be used to determine guilt. If the person is a witch, then she should have been able to save herself. It's a bit of damned if you do and damned if you don't. But the innocent is dead regardless. Such was medieval justice.
Here we find the nuptials of Henry VIII brought into focus as arguments were made to put Catherine of Aragon in her place, and make way for the parade of ladies more to Henry's liking (Franklin says many couplings were in public. This was probably added, and although most likely true, to raise the temperature of the reader.) A bit of history springs forth regarding Martin Luther who believed in witches and corporal punishment of them. And, there is much more which fills some hundred or so pages and makes the book worth having if only for Franklin's meticulous research and footnotes (actually endnotes of one-hundred and one pages).
I suspect Franklin put this group of chapters up front for fear if he didn't the casual reader would never encounter the blood and gore that was so typical of civilization in years past. Nevertheless, if this group is logically assembled at the end, the result is more in keeping with a review of the science of making the best possible guess or when guessing, it's best to have a scapegoat to protect your backside (sort of like weather forecasters.)
***
Science of Conjecture, a book I recommend highly!
Simply put, the science of conjecture is trying to guess what is going to happen given a splattering of "facts" on which to base a decision. In the 20th century, the heyday of statistics, every pocket calculator seems to have a program that permits one to become a forecaster based on the input of a series of numbers. Then on that base, an extrapolation can be made into the future(?)
But before you put your trust in the future, based on numbers you enter, consider the case of a noted forecaster. Al was an advocate of linear plotting and ignored clues that the data was not fit for (to fit) a straight line. His answer, there is no evidence that another model will be more satisfactory, with the limited information available. The very best forecast was just to use the very latest data point as the best estimate of what the future would hold. He was right; in an unpredictable market or business, the past has little sway on how volume, price, and supply will respond to circumstances not predictable. His best example was the impact of a large ocean going vessel loaded with goods (be they oil, toys, fertilizer or dollar-bills). The vessel when it sets sail must make landfall somewhere and so long as the market for the product has been previously established a normal business climate exist. However, as that ship is on the seas and must make port and disgorge its contents, regardless of need or price, a chaotic result is inevitable. There is no predicting the result.
Al employed an economist with a PhD who just so happened to have a crippled left arm, the result of childhood polio. He often said that Graham was hired because he wanted an economist who wouldn't say, "on the other hand".
Al headed a large organization, "Research and Long Range Planning". With passage of time, research became a dirty word and those engaged in it were folded into the "Long Range Planning" department. All went well until a group of forecast based on Al's reasoning went badly astray. With the stroke of a pen, the group's name was shortened to "Planning" Shortly thereafter, even that title disappeared along with those who had worked under the banner.
Al's mistake - refusing to acknowledge that business cycles exist. With this mind set, how was he to know where the company was within a business cycle or how to predict the future?
****
Joe Wortham's Home Page , About Joe Wortham , Directory
Questions? Comments? [email protected]