The next installment of my axioms took a bit longer. :-) The development of the next set of axioms that I will use is less certain than those I have developed. They involve the analysis of historical events. By definition, one cannot do experiments to test the validity of statements concerning historical events. This problem is so extensive; logical positivism fell on its inability to address the problem with any internal consistency. This gives us our first axiom:
Accepting a logical positivistic framework precludes any meaningful historical analysis. Even statements like "Julius Caesar's adult height was over 5 feet" have no truth value for logical positivists. Therefore, in discussing history, we will eschew logical positivism. We will assume that meaningful statements can be made about the past. These statements can have rough likelihoods attached to them. For example, the statement given above could have a general likelihood attached to it; studies could be made of the average height of people during that time, taking into account the superior nutrition Caesar would have had growing up, compared to a subsistence farmer. Perhaps there are references in reports that let us put bounds on the statement. Not knowing this information, I'd tend to assign a 50/50 chance to this. Historians who's work focuses on the Roman Empire of the time could assign better odds. Even though neither they nor I know for sure, their judgment should be considered better than mine. In historical analysis, we do not have the ability to do experiments. We also don't have a very large number of examples. In some sense, we have just one example, in others we could have a number (different elections in different countries, for example). But, we don't have, as we do in astronomy, billions of examples that can all be used. So, there are limitations on our ability to separate out the causes and effects. Reasonable people can differ on historical analysis. But, there are still analyses that have verisimilitude and those that don' t. For example, the Lost Cause historians argument that the Civil War was fought over states rights, not slavery does not stand up to rigorous analysis. Although one does not have the rigorous falsification one has for the aether theory of electromagnetic propagation, for example, it is still fairly well discredited. Facts like the strong support of the South for the most intrusive Federal law (the fugitive slave law), the speeches given at the Confederacy constitutional convention, the fact that the Confederate constitution was identical to the US constitution except for the enshrining of the institution of slavery. Given these facts, one would need to understand why people who's prime concern was states rights would push for legislation that incorporated unprecedented intrusiveness in the affairs of various states. If it was states rights, why don't we have that clearly spelled out. If it was, why wasn't the Confederate constitution written in a manner that clearly strengthened the rights of the states vs. the power of the Federal government. Without experiments, of course, it is hard to falsify this concept in the same manner as caloric heat was falsified. In this case, we do not have exact predictions to compare observations against. Rather we have a softer model of behavior against which we can compare actual historical events. But, there is still some value in it. For example, we could explicitly detail some of assumptions: 1) People will act to ensure that their highest priority is met, even to the point of sacrificing lower priority actions. 2) At places like constitutional conventions, people state the reasons for doing things. The exception to this is that the more unsavory reason can be acknowledged with a nod and wink, while other better sounding reasons are given for public consumption. 3) Constitutions represent the political views of those who write them and ratify them. >From these assumptions, predictions can be made. If states rights are more important than slavery, then one would not push for slavery legislation that lessened the independence of the state. If people really thought states rights was the reason for the Confederacy, why did they highlight slavery at the constitutional convention. Was slavery thought to be less controversial than states rights? So, even in the absence of experiments, one can assign rough likelihoods to various historical interpretations. For example, it is near certain that Rome ruled Palestine during the life of Jesus. Chances are very close to zero that Jesus ran off to India instead of being crucified. The next step that can be taken is a bit more problematic, but can still be addressed. That is determining importance of various factors in historical development. The "what if" game is hard to play conclusively because it is impossible to go back and rerun history with a parameter changed. Still, I think one can look at history and answer broad questions. One of which is whether historical progressions are: 1) Chaotic 2) Deterministic 3) Complex (in Gell Mann's sense) The chaotic view of history has been expressed in a number of SF books. Stepping on a butterfly in the distant past changing an election is one example of this. While this cannot be absolutely falsified, there are many difficulties with this. For example, family tall tales to the contrary, my dad's service in San Francisco did not save the West Coast from a Japanese invasion. If my grandma had not died in the flu epidemic of 1920, our family would have been much different, but it's not likely that the Great Depression would have been averted. It makes sense that, in a country of >100 million (as the US has been since about 1915) the lives of most individuals would tend to have minimal impact on the general shape of history. This idea has led to the opposite view: individual actions do not matter at all. The historical dielectic is one example of this: history is thought of in terms of classes, not individuals. Psychohistory is a great example for SF fans. Hari Selden's work is well known in this field. :-) This view also seems to be at odds with observations. If it were to be true, then US actions, like the present Gulf War, also have no impact on history. The reason I say this should be clear: it is very unlikely that Gore would have invaded Iraq. Afghanistan is probable (but debatable), but I think few would argue that a Gore administration would have invaded Iraq. The win by Bush was very narrow. Individuals certainly had an impact. For example, if the butterfly ballot did not move Gore voters to Bucannan, then he probably would have obtained more votes. Certainly, if Clinton had kept himself zipped, he could have campaigned for Gore without the enormous baggage he had at the time. There are numerous other possibilities in the 20th century. The third submarine officer who wouldn't give the authorization to launch missiles during the Cuban missile crisis changed the course of history. (They were being depth charged at the time by the US.) If they had launched their missiles successfully, a number of US cities would have been turned to glass. That certainly would have had a tremendous effect on world history, even if Kennedy and Khrushchev found a way to limit the ensuing war short of full scale. I've given the example of Lincoln's unique skills, and Gautam has given the example of the great difference between Truman's and Wallace's views of the USSR. I would have a hard time believing that Wallace would have acted as aggressively in Europe as Truman. Having given difficulties with the deterministic and chaotic views, I would like to present my view: history is complex. Some random (or quasi-random) factors are involved. Still, the flow of history is not chaotic; only a few specific small changes have gigantic repercussions in history. This type of behavior is what Gell-Mann refers to as complex. It seems straightforward to understand why deterministic, broad sweeps of history (or any broad linear system) would be easier to model. But, as we see with QM, and statistical mechanics, random actions can also result in straightforward rules. In between, determining and predicting results can often be much harder. Now, if you look at the evidence I give above, you will see that it doesn't match the checking of QED by g-2 in precision or in accuracy. The arguments are by counter-examples that are somewhat broad. It is possible that Gore would have lost if Clinton kept his pants on; it is just unlikely. It is possible Gore would have us in the exact same place in Iraq, but few think that likely. It's possible that Bush will announce that Howard Dean will be his new secretary of Defense, but I don't think I could get anyone to bet on it. The purpose of this long discussion is to support one axiom I will use later: Axiom: The development of history is not inevitable. If certain small things had been different, the outcome of a number of different clashes of worldviews could also have been different. I think to falsify this axiom, one would have to show how the relatively small actions given above did not have long term effects on the course of history. (and there's plenty more where those came from.) Dan M. _______________________________________________ http://www.mccmedia.com/mailman/listinfo/brin-l
