While reading about the Marijuana debate in the last several days, I am overwhelmed by the need for a discussion about complexity in science. First this leads me to question which sciences are relatively complex and which ones are pretty straight forward. I resolve to advocate for more appreciation of the role human involvement in a field plays in introducing complexity.
We, in academia, generally think of something as complex when it is difficult to acquire the technical skill to compete at the highest level in the field. This generally explains why physics is called the queen of the sciences. To learn mathematical skills you have to have talent, hard work, and the fortune to have both good teachers and the circumstances to avail yourself of them. I would argue that the truth is the opposite of this instinct, that social science is much more complex. Social science is something that people can learn though reading, and everyone thinks they are an expert. If you don't believe me, try studying politics and then go out in public and debate people on the street -- their ignorance of the literature will likely not stop them from dismissing the nuanced model you are trying to present.
It is important to consider how error is identified. For me the natural and social sciences differ in how they deal with error. Take engineering for example, error is identified rather quickly, in what I would call a "tight" feedback loop. In the social science, by contrast, error is in a loose feedback loop (and in fact many times error can be dismissed completely as resulting from flawed measurement). Economics has taken the position as "queen of the social sciences" though its ability to mathematize its subject and apply statistical tools. The noble pursuit is to isolate single variable analysis, the practice is to isolate the subject from the social world it seeks to study.
If engineering did similarly detached itself from reality, bridges would start collapsing. During a recent visit to NYC, I am overwhelmed that buildings constructed the better part of a century ago can be so solid. How did they build the Empire State building in 16 months without a clear science of engineering? Can we really claim that we have such a social science?
The best skyscrapers of social science, by contrast, seem to only stand a few levels high. I attribute this to poor feedback loops in the disciplines, but it is also a function of the diversity of preferences that are at hand. There is no difficulty in understanding the difference between a bridge and a skyscraper. These are different arrangements of the same materials (concrete and steel). In fact, we can understand the different appearance of the old world trade center and the new quite easily -- but how do these different structures manifest? In some way it is simply fashion. In some ways it is determined by the person or group in charge at the time the plans are selected. But none of these are engineering questions - they simplify no further than a discussion of preferences. That is the realm of the social sciences.
But, we are not content simply to reduce everything to different preferences. Policy is set in order to get coherence on particular topics. There are two articles on marijuana that have been circulating recently. The first was on treatment options that the marijuana derived CBD gives for families suffering with child seizures. Another article talked about the misinformation in the public health debate, a famous doctor changes his mind.
Rather than see these instances the exception, I see them more as the rule. Anything that involves human beings is going to have a degree of complexity, and this includes medicine. Marijuana will work for some folks. I would still defend their prerogative to try it even if the link could not be established in the medical literature. This is what it means to acknowledge complexity in social science, and there are always cases where the one-size-fits-all approach is insufficient for political economy. The history of why marijuana became illegal is interesting, but it is an example of the all-too-common phenomena of folks misunderstanding the nature of scientific discovery and too quickly trusting the scientific consensus.
The less things are like skyscrapers, the more important it is to consider complexity.
