Thursday, January 21, 2010

The Social Psuedosciences

The study of culture, ideas, history or the economy is a study of, in the essence, complexity. Rarely, if ever, does the focused study of these fields reveal much more than either an axiom of siloed knowledge or a very approximate model of its causes and effects.

Coming from the world of engineering I understand very well the gross inaccuracies of either a top down (taking effects and working backwards) or bottom up model (taking fundamental causes and working towards effects) and that often these models must be used, but should not be relied upon. In my specific field of material science, there are many theories and explanations for phenomena at the atomic level, but as of yet no model which uses such atomic causes has been able to produce results anything like that in real life. Luckily, in engineering, we only need a working knowledge and not a true understanding. I eventually came to the conclusion that almost all engineering is just a working knowledge, with very little to any true understanding. We leave that to the physicists (and perhaps computer scientists?).

This problem with our ability to understand complex phenomena becomes much more dangerous and troublesome when dealing with the social sciences, where very little, if any, working knowledge exists. It is perhaps because of this lack of working knowledge that we tend to rely so heavily models to explain and attempt to, often very badly, predict future occurrences. I should explain here that working knowledge generally is a direct result of highly controlled scientific observation, which may be possible with an aluminum bracket, but not any activity that involves people. I realize this is a contentious point as a great many social experiments are conducted at universities everyday. I contest that no human can ever be impartial when studying another human being. For any true science to be conducted there must be a removal of impartiality on the part of the observer, or what I think is the more important part of the experiment, on the part of the designer.

The second issue I have with the reverence the social sciences receive in modern culture is the siloed nature of its study. I have yet to know of an academic who has the vast knowledge, training and resources to actually be a sociologist/ economist/ psychologist/ historian/ anthropologist/political scientist, but so many of the problems they study could easily fall under any of these categories. A sociologist, historian or economist will all look at the Great Depression from vastly different viewpoints, but which of their arguments is the most valid? The truth is all of their arguments are likely very valid and that each of them has revealed a cause, while at the same time each likely missed many more. This is a battle against complexity and should never been one we so easily dismiss as being one. We are almost never right, just slightly closer to the truth. (Truth here is in italics as in many cases it may not exist, but that is another blog)

NOTE: I recently finished reading the book The Black Swan: The Impact of the Highly Improbable by Nassim N. Taleb and it has undoubtedly influenced my thinking. At the same time I have had these thoughts for some time, but must thank Mr. Taleb for helping me form them into more concise arguments.

-DG

1 comment:

  1. One of my profs often commented that the "science" in political science is more about attempting some respectability when it comes to grants. Interestingly, the "silos" are really more of a recent trend; before economics became an exercise in scientism ("laws", "equilibriums", etc.), it was considered inseparable from politics - Adam Smith did not write a treatise on economics, but on political economy. Later on we have the likes of Friedman and other monetarist/libertarian types who rather insist that a certain type of market economics is "natural" and most be protected from the "interventions" of politics. Earlier thinkers never came up with such arbitrary divisions.

    ReplyDelete

Thanks