Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

News Pollitical science is not a science

  1. Mar 8, 2005 #1
    "Pollitical science" is not a science

    For something to be a science it must follow the scientific method. In particular it must be falsifiable.

    Political science does not follow the scientific method and is not falsifiable because it is not practical or moral to create political experiments. To create an experiment it is necessary that all factors other than those you want to study be eliminated. This is impossible in the domain of human interaction, where you cannot arrange people into political units as you like, eliminating the effects of culture; therefore there is no science in politics.

    So it's all a matter of what seems most reasonable. There are no experts in politics; no testability means no way to determine validity, except for reasoning through what seems to make the most sense from one's own point of view.

    We see the effects of this untestability in the multiplicity of opposing political theses. Were political science a science, fundamental opposition of that nature could not arise; one view would be demonstrably valid, and one would be demonstrably false. The explanation is that there is no way to demonstrate falsehood, and therefore no way to distinguish, except through each individual's unaugmented, natural power of reason.
  2. jcsd
  3. Mar 8, 2005 #2


    User Avatar

    Staff: Mentor

    Here is the flaw in your reasoning:
    There is no such thing as a perfect experiment. Every paper you write for your science lab classes will include a section where you discuss the possible sources of error and all professional papers discuss the possible sources of error.

    The experiments (and there are social experiments) need not be perfect to be scientific.
    The word "expert" is not an arbitrary distinction. It has professional and legal implications. An expert in a field is recognized as such by other experts and by law (there is a special legal significance to an "expert witness" for example). An expert in political science is a phd political scientist.
    One example: the USSR. The USSR tried to impliment a system and that experiment failed. Hence, that system is not a valid form of government.
    Can you name another field of science where such harmony exists? Check out the cosmology or the string theory forum and see how there is disagreement over the validity of certain ideas. Political Science may be tougher in that regard, but that doesn't mean it isn't a science.

    You're holding on to an idealized version of science that is not reality. You don't understand what science really is and thats why you don't understand why Political Science is a science.
  4. Mar 8, 2005 #3
    But that's exactly why it is a science, and a beautiful one.
  5. Mar 9, 2005 #4


    User Avatar
    Gold Member

    We had a board where students could write an oxymoron for the day. One day someone wrote "Political Science." We had a good laugh, though this has been debated forever, hasn't it? Suffice it to say, scientific methods are used by Political Scientists, but since I don't have the strength to do battle on this one, I'll just thank those of you above for your good replies.
  6. Mar 9, 2005 #5
    I think Russ made very good comments and I only have two things to add:

    Regarding falsifiability, Popper himself thought it was important to distingwish between the logical meaning of the word and its methodological applications:


    With his theory of Verisimilitude he went on to explain how theories' strengths could be compared with respect to their 'truth'-value and 'false'-value. In short, he said that theories can only be more or less true and we must pick the more true ones.

    Another thing is that political questions are such precisely because they can not be solved scientifically. However, political science is not directly about those issues, it is about the their context, interpretations, interdependence, and the application of a theoretical framework on them.
  7. Mar 9, 2005 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I think the others here have already addressed the flaws well. I just want to point out these two statements:

    By this logic, study of ecology would not be a science either, as would many aspects of medical research (and there would be plenty of other examples, these are just the ones I know best). It is not always practical or ethical to create the perfect experiment, and this occurs in any field of science. Instead, one must carefully observe and compare "natural experiments." The use of statistics to analyze the results and the interpretation of the findings will be adjusted and applied cautiously to address the known limitations of the experiment, but these limitations do not make it less of a science, just a more challenging one to study.

    First, you never eliminate other factors, they will always be present. In the ideal experiment, as you would be taught in elementary science classes, you would control for all variables other than the one you are testing. However, we do not live in an ideal world, and in practice, every variable cannot be practically controlled for, and very often, not even identified.
  8. Mar 10, 2005 #7
    Though scientific methods are used in many fields of study, such as political science, it is not considered to be a pure science. In looking to see how "pure science" is defined, I googled into Wikpedia, and saw the "Dispute Resolution" section. Perhaps just as interesting is the creationists and mainstream science posts on that site...?
  9. Mar 12, 2005 #8
    This is strangely the first time I have looked at the replies here since I posted it.

    Dispute resolution is a good point. You never have that conclusively in political "science."

    You never have good experiments. All right, not all experiments are perfect, but some are good and some are not. In political "science" no experiments are sufficiently controllable to demonstrate or disprove any idea with any reasonable certainty. Name one--one bona-fide _experiment_. Observing conditions and drawing what seems to you to be conclusions is not an experiment.

    In other sciences people tend not to disagree over the basics. In the most abstruse areas of discourse there can be some disagreement, though it is nearly always resolvable through more experiments. In political "science" people disagree over the very basics all the time.

    Also in political "science," the conditions are never the same way twice. There are no twenty cities just like Boston, like there are twenty rats just like each other or twenty electrons just like each other. Sample size is ludicrously small. The lessons of the past--to the extent such lessons can be gleaned, which is not large--become increasingly irrelevant to various degrees, often large degrees, as technology and hundreds of other factors shift in ways not understood.

    Political scientists cannot predict the stock market. Did you know that the amount of variation in stock market indicators over time remains approximately constant no matter what the resolution you're looking at the indicators with? If you look at a stock market graph over 1 year it statistically looks much the same as a stock market graph over 10 years. Makes you think that the factors determining it are chaotic.

    If political scientists were worth anything, they would be able to predict the stock market and become rich.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook