Dan Lindley

How to Make a Theoretically Informed Argument

November 25, 2000, v. .8 (early version)

This handout is designed for those learning how to make theoretically informed (political science) arguments. It assumes that the task facing the student is to explain a historical or current event using their course=s concepts, models, and theories. There is a section on how to write a policy prescription paper below under notes. Unless specifically asked, do not write a normative paper, a prescriptive paper, or a straight historical narrative.

  1. First, find a question to answer or puzzle to resolve.
    1. There are examples below, but the general form is to ask : Why did this event happen? What caused what so that this event unfolded the way it did?
  2. To answer your puzzle and make your argument, you need 3 ingredients:
    1. A theory
    2. Predictions
    3. Evidence
  3. A theory (or theories) is the basis of your explanation or answer to your puzzle. A theory is:
    1. A hypothesized causal statement (AB>B) with an explanation of why A causes B.
    2. More generally, a theory is a hypothesized pattern of behavior for (for IR/AFP classes) individuals, groups, states, and/or the international system.
  4. Predictions, or observable implications, predict what the world would look like if your theory was true. They tell you what evidence would confirm your theory.
    1. Several examples are given below. Predictions are not about the future or the weather.
  5. Evidence is the data that supports your main theory, its complements, or counterarguments.
    1. Evidence ranges from statistical correlations to public opinion polls to detailed histories of meetings to direct quotes from actors.
    2. What evidence is most convincing depends on your question and theory.
      1. In the example below, public opinion polls are crucial for addressing the issue of whether or not public opinion led to the Somali intervention. Records of what President Bush was thinking (meetings, quotes, psychological profiles, etc) are helpful in determining whether or not Bush had humanitarian motives in ordering US forces to Somalia.
  6. General Example of Making an Argument:
    1. Puzzle: Why did the U.S. send military forces to Somalia in 1992?
    2. Primary explanation (ie your argument):
      1. humanitarian motivation to save lives
      2. Arrow diagram: humanitarian motivation C> US military intervention (because President and commander in chief Bush was deeply affected by the plight of the starving Somalis)
    3. Complementary (but less important) or competing explanations:
      1. US economic interests demanded an intervention
      2. US public opinion demanded an intervention
      3. To flex US military might in the post-Cold War world to:
          1. impress others and build credibility
          2. increase/defend defense budget
    4. Your argument PREDICTS that the evidence will show the Bush had humanitarian motives. Predictions/evidence for the primary explanation include records indicating Bush's angst at the tragedy, statements of angst at planning meetings, angst-filled reactions while watching the news with his family, lack of other motivations when he discussed the options and made his decision, etc. Strong evidence would be transcripts of meetings in which others argued that we had no strategic or economic interests in Somalia, that intervention would be bad the Department of Defense, that public opinion didn't care, and in which Bush said that he didn't care about these arguments, that he was President and that he was not going to let those Somalis die.
    5. What are the predictions for the complementary or competing arguments?
  7. For advice on how to sew this all together, please refer to my handout:
  8. http://www.nd.edu/~dlindley/handouts/writingtips.html 


    *********Some questions, issues, and problems that arise********

  9. PROBLEM: how do we know what really caused any event in political science?
    1. Answer: it is very hard, and sometimes leaders themselves can't fully explain their actions. Yet, life and death issues are involved and we must do our best to make wise policy. We must be modest and ambitious at the same time.
  10. NOTE: If something causes something, the cause must precede the effect.
    1. Logical, isn't it? Yet many students mess this up.
      1. Example: the shift in public opinion in favor of Somali intervention must precede the intervention if there is to be any chance that public opinion helped cause the intervention.
  11. PROBLEM: Ignoring counterarguments and evidence that goes against your arguments.
    1. Solution: be honest with the data. If you are wrong, switch arguments and be right.
    2. Solution: be aware of counterarguments. Acknowledge them and explain why they are weak and yours is better.
  12. PROBLEM: if multiple theories predict the same thing, your argument will lose persuasiveness.
    1. SOLUTION: add predictions and make them more specific
    2. Example:
      1. Puzzle: why did the dinosaurs die?
      2. Theory 1: An asteroid hit earth, and kicked up enough dust to cool the climate and kill the dinosaurs.
        1. Prediction: evidence of lots of dust in atmosphere when dinosaurs were estimated to have died.
      3. Theory 2: A huge volcano erupted, and threw up enough dust to cool the climate and kill the dinosaurs.
        1. Prediction: evidence of lots of dust in atmosphere when dinosaurs were estimated to have died.
      4. PROBLEM: finding evidence of dust is not enough to determine which theory is true.
      5. SOLUTION: make more specific predictions about what kind of dust each event generates, make predictions about asteroids causing indentations and volcanoes causing mountains, etc.
  13. For opinion or policy arguments, find an argument to make. These arguments should be theoretically informed.
    1. Theories are a good basis for predictions because theories are hypothesized patterns. Patterns are the basis for prediction (think about it: if behavior was random, nothing could be predicted). Policy arguments are predictions in two ways: they predict what problems will arise and they predict what solutions will work. To do this successfully, one must be aware of patterns applicable to the case at hand. An example of such a pattern is balance of threat theory, which might be well applied to the question of future great power relations, but less well applied to the issue of whaling agreements.
    2. Instead of explaining history with a theory, the task here is to make an argument about wise policy. You must make predictions about current/future problems and their solutions. Instead of spending much time marshaling evidence for your explanation, you have to justify why you think your theory is well fitted to make predictions about the issue at hand. For example, you would have to justify why balance of threat theory was appropriate for assessing great power relations. You have to use theory to diagnose the problem at hand, and to argue why your solution will work and why barriers to your solution are surmountable.
    3. The meaning of the word prediction in this section accords more with the informal understanding of the word meaning a forecast of future events. When first used above, prediction had a more formal social science meaning of past events that if found would confirm the theory (ie prediction as an observable implication of the theory).
  14. A note on complementary and competing theories.
    1. Complementary theories are those that help explain the event in question, but do not explain as much as the primary theory. Think of it as a pie chart: one explanation covers 60%, the next 25%, the next 8%, etc.
      1. How many theories you wish to deal with depends on your question and your willingness to trade parsimony for richness. This is an art and a judgement call. My rule of thumb: do what is most persuasive. If the best theories only explain a bit of your event, then you may need to incorporate several theories to make your case. On the other hand, if the best theory explains most of your event, then you can mention complementary theories in passing.
        1. Parsimony is achieved when you can explain a lot with a little.
        2. Richness adds detail, but at what cost in clarity?
    2. Competing theories are similar to complementary theories except that they are not part of the same pie. They are separate alternatives.  Many people use the term competing when they really mean complementary.
    3. NOTE: often it is helpful to think of theories as contending or competing when in fact they are complementary. This is because debates about what caused what are often about which theory explains the 60% and the other the 25%. They are indeed complementary, but it makes a big difference which is the more important. When writing, use the terms competing, contending, or counterargument when arguing about which theory is king of the hill and explains the most. Use the term complementary when talking about theories that occupy support roles of secondary explanatory power.