One of the most boring episodes of my high school career was the few weeks in fifth form history when we were taught the causes of the French Revolution. To this day I really can't remember much of what we were taught but, looking up Wikipedia today, I read that one cause was the high level of taxation of the peasant class while the nobility led lives of conspicuous consumption.
But what exactly does "X causes Y" mean in the context of history? Can we deduce that if the French peasantry had not been taxed highly then the revolution would not have occurred? Of course not. In fact we cannot know the answer to that question because we cannot rerun history. Does it mean that there is a tendency for Y to occur when X is present in any historical situation? Perhaps so, and this is indeed a hypothesis that could be tested by an examination of historical situations where X was present. But the dataset might still be unconvincingly small.
I was led to these thoughts by two massive books that I've read over the last two years. One was "Guns, germs and steel" by Jared Diamond, the other was "Ideas, a history from fire to Freud" by Peter Watson. Both of them are staggering accomplishments that offer a comprehensive account of how hunter-gatherer societies developed into modern-day societies. Inevitably the authors colour their accounts with opinions that can only be hypotheses, despite their persuasive arguments to support their opinions.
Diamond's main intent is to give an explanation of the factors that contributed to the greater success of European civilisation compared to its historical counterparts. He completely rejects explanations based on racial differences and argues convincingly for geographical differences. Among his explanations are, for example, the greater number of domesticable animals in the Old World compared to the new; and the larger number of plants that could be farmed for food production. I strongly recommend the book for it is thoughtfully presented and hugely informative.
Watson's book offers a number of reasons why some ideas surfaced at a particular moment in history, and why they originated in a particular country. Again it is all plausible coherent stuff and I greatly appreciated how he related the historical contexts to the development of key ideas.
But, just like the causes of the French revolution, what does it mean to say that a particular constellation of geographical factors made the Old World develop more quickly than the New World? What does it mean to say that a particular constellation of social and historical conditions caused a certain set of ideas to be developed? To make these claims is not to say that if X was not present then Y would not have occurred. History is not a repeatable experiment.
Neither in the case of the Diamond and Watson claims can we even test the hypothesis "Y tends to happen in the presence of X". We can't do this because they have already used up all the experimental data. In other words they did too complete a job!
If we were to encounter a plentiful number of alien races not too unlike us we would have more scope to make historical claims of the form "X causes Y". Since this is not likely to happen we appear to be stuck. However there is a possible avenue that might enable us to test this type of causality. We could simulate various worlds and societies in a computer, seed them with various initial conditions, and study how they evolved. Such simulations already exist, some on a very large scale: some online computer games depend for their very fascination on watching how a virtual society develops under the constraints of the game.
So here's the ideal academic project: conduct history and sociology (and other inexact disciplines) through the medium of computer gaming. Wouldn't that be great fun for the young folk of today?