Corruption: Is It Measurable?
by Dean
One of the bigger misconceptions about the social sciences is that they are not "real" science because they use arbitrary standards, don't use math, or don't make predictions. In some areas this is probably a legitimate criticism (especially in some areas of psychology), but in many cases it's simply wrong. Demography, for example, is a science that has been one of the most influential in the last 100 years, right up there with discoveries in physics and medicine in changing how we live our lives. We see instances of demography every day without even being aware of it: the very concept of an "ethnic minority" depends upon it, as do public opinion polls (which influence how politicians govern) and census data, which provide an enormously useful look at at populations and trends over time.
Another area that really is a science is political science. One of the most easily accessible data sets in political science is the Freedom House data set, which measures freedom around the world. You think it can't be measured? They've been using the same measurement standards for well over 30 years, and that data set has been used to make remarkably vigorous predictions.
Yes, of course, there's some arbitrariness and subjectivity to some of the measurements used in political science, but so what? If you think other fields don't have that problem, try some time to get a biologist to give you a specific, concrete, universally understood and agreed-upon definition of "species" that all of his colleagues will agree with. Or ask a physicist how you determine both the position and velocity of a subatomic particle.
The Freedom House data has been used to make some remarkable predictions: for example, no two nations ranked at least 4,4 or better has ever gone to war with any other nation ranked 4,4 or better, and the Democratic Peace Theory holds that they never will. So far there have been literally hundreds of thousands of opportunities for that to be falsified, and it never has been. Thus what you have here is a strong theory. (There's a weak version which says merely that it will be rare, but so far the proposed exceptions are very weak indeed.)
Interestingly, political scientists have also put together indexes to measure corruption. While it might surprise a few folks, political scientist Rudy Rummel notes that his fellow scientists have found that the more democratic a nation is, the less corruption there is in government.
This should be no surprise if you think hard about it. Free speech, free press, and free elections make corruption much riskier. When we see evidence of corruption in the new, it's not evidence of how corrupt our system is, it's evidence that it's working properly.
Think of it another way: if you come down with a fever, is that proof your body is hopelessly corrupted with infection? In some rare cases yes, especially if the fever goes completely out of control. But most often, in the vast majority of cases, it's just a symptom of your body successfully fighting off an infection.
Rudy also makes the interesting observation that sometimes in science you get a theory that people seem obsessed over. The Democratic Peace Theory is just such a theory, because it's shown through rigorous data analysis and specific predictions that democracy reduces war, reduces poverty, reduces starvation, and, yes, reduces corruption. People hear you talk like that and it sounds like you think you've found a panacea, but all you have to do is look at the data, folks.
I had a similar experience some years ago experimenting with low-carbohydrate diets. I was part of a community of people who used them. Those who advocated low-carb diets routinely said they weren't for everybody but that in a substantial number of people they could a) help lose weight, b) help diabetics attain blood sugar control, c) reduce blood pressure, and d) improve serum cholesterol numbers.
This by eating a diet consisting of things like red meat, bacon, eggs, and cheese on a daily basis.
I remember the howls of rage I would get from some people, the incredible scorn, and yes, the warnings that I was advocating something dangerous that would kill myself and others. Dr. Robert Atkins was a flake, a liar, a con artist, and even possibly a murderer.
It was amusing to watch over the last decade as controlled study after controlled study showed that, indeed, for at least some people, those were the exactly predictable results of such diets. For years Atkins had advocated that such studies be done because of the results he was seeing in his patients, but no one wanted to do them. They preferred to abuse him and a few of his colleagues who said they had the same results instead.
I think that was my first experience in dealing with fulminating bullies in science and medicine. It was most instructive, and is something I've never forgotten. When the data is on his side, even a mouse can stand up to a lion.
