Friday, December 24, 2010

Should Political Science Be Scientific?

One of the preliminary chapters of my political science methodology textbook, entitled "Studying Politics Scientifically," states: "Knowledge that is evaluative, value laden, and concerned with prescribing what ought to be is known as normative knowledge. Knowledge that is concerned not with evaluation or prescription but with factual or objective determinations is known as nonnormative knowledge. Most scientists would agree that science is (or should attempt to be) a nonnormative enterprise." (Johnson and Reynolds, p.31) Aside from the irony of the textbook's statement that science should be nonnormative (which itself is a normative statement "prescribing what ought to be"), I'd like to contrast that quote with some of Strauss' arguments from What Is Political Philosophy? In the first part of the essay ("The Problem of Political Philosophy"), Strauss states, "'Political science'... designates such investigations of political things as guided by the model of natural science..." However, Strauss also makes some interesting observations about the nature of political things: "Political things are by their nature subject to approval and disapproval, to choice and rejection, to praise and blame. It is of their essence not to be neutral but to raise a claim to men's obedience, allegiance, decision or judgment. One does not understand them as what they are, as political things, if one does not take seriously their explicit or implicit claim to be judged in terms of goodness or badness, of justice or injustice, i.e., if one does not measure them by some standard of goodness or justice." (p.12)

Of course, if true, Strauss' description of political things poses a problem for modern political science. If Strauss' analysis is accurate, and if political science is the application of the methodology and assumptions of natural science to political phenomena (see p.1 of Johnson and Reynolds), then there's the very real possibility of what philosophers call a "category mistake" or "categorical error". For instance, when people argue about abortion, do they debate whether abortion is true or false? No, the argument is about whether abortion is right or wrong. To discuss abortion in terms of truth and falsity is to commit a categorical error -- to confuse two different categories: the logical and the moral. If political things or (in more academic language) political phenomena are as Strauss describes them, then it would seem wholly inappropriate to study them as you would the subject matter of natural science or to study them using the framework of natural science. Does the law of gravity demand praise or blame? Does photosynthesis demand allegiance? Does mitosis require choice or rejection? Does one approve or disapprove of the quadratic equation? Rhetoric aside, this observation at least might cause us to hesitate before accepting the premises of modern political science as laid out by my textbook.

One of the more interesting implications of this problem is that if political phenomena do make claims on us as humans and demand evaluation, then attempting to study them in a "nonnormative" way actual would lead us to a less realistic understanding of them. If political phenomena inherently present themselves in terms of goodness or justice, then attempting to be "nonnormative" endangers objectivity and gives us a distorted, filtered, and sterilized view of political phenomena. If the purpose of empirical science is to understand things as they are, then a nonnormative, value-free approach to political things actually prevents us from understanding them as they are; empirical science turns out to be self-defeating when applied to political things. Only an evaluating or evaluative political science can be truly scientific.

There are some (perhaps obvious) parallels that can be drawn to theology and theological things. I plan to cover a few of those in subsequent posts. Until then, do you think Strauss' analysis of political things is accurate? If so, does that really pose problems for the way political science is currently conducted? If it does, is there anything that we can or should do about it?


Johnson, Janet Buttolph and H.T. Reynolds. Political Science Research Methods
Strauss, Leo. What Is Political Philosophy? And Other Studies


  1. What a load of high-sounding nonsense! (Seriously, though, I had to read that twice before I understood it).

    Unfortunately, not knowing much (anything) about political science, this is all a bit theoretical for me. What is it, exactly, that you study? Can you not break down your inquiry into the political process into, say, "Political Science" and "Political Philosophy"? First, study the science and logic of the political process, and then use the nonnormative knowledge you have learned, and evaluate its justice and morality?

    As an example, study the principles of democracy from a scientific standpoint. Is this an efficient mode of government? How does the percentage of the population living in poverty compare to those populations under other modes? Answer this line of questions, and then use this knowledge as a background for ethical inquiry: Is democracy a more ethically sound form of government? Is it just?

    Or, perhaps that's just more nonsense.

  2. Heh, seems like I did choose an appropriate title for this blog...

    You bring up a good point, and it's actually almost the same point that the professor of the class in which I used that textbook brought up. However, an adequate response to that question would pretty much require another blog post of its own... Which I will start working on. ;)

    Until then, I can kind of summarize the direction I'll be arguing in... by using an analogy to medicine. If science as science is to be value-free and nonnormative, then medicine as a science should also be so. That being the case, medicine must study medical phenomenon while refusing to pass judgment on whether or not humans should have long and healthy lives, or whether the lives of humans are more valuable than the lives of bacteria and cancerous cells. (Is it even possible to look at cancer while being neutral in the conflict?) Where would medical science even start? You have no basis upon which to call any state of affairs "health" or "disease" because you can't say whether a given configuration of atoms in a human body which is conducive to living is inherently any better than its opposite; that would be making a value judgment, compromising your scientific objectivity. Theoretically, only after making observations under these neutral conditions could you then proceed to make recommendations for treatment for what you've philosophically judged to be "diseases" -- but now you've left the realm of medical science and such prescriptions are made on the basis of medical philosophy...