I

…[T]he ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back. I am sure that the power of vested interests is vastly exaggerated compared with the gradual encroachment of ideas. Not, indeed, immediately, but after a certain interval; for in the field of economic and political philosophy there are not many who are influenced by new theories after they are twenty-five or thirty years of age, so that the ideas which civil servants and politicians and even agitators apply to current events are not likely to be the newest. But, soon or late, it is ideas, not vested interests, which are dangerous for good or evil.

- John Maynard Keynes, The General Theory of Employment, Interest, and Money (1939), the last page.

Theory and Explanation in IR

General v. Idiosyncratic Explanations

Value of General Explanations

- parsimony: explain more with less (Occam’s razor)

- prediction: general theories can predict future events

- control: using a theory, one might be able to prevent unwanted outcomes and generate wanted ones

Fancy terms

- ontology: what exists

- epistemology: how we know what exists

- methodology: how we show how we know what exists

Theories v. Other Knowledge

Supposedly Non-Theoretical Knowledge

- facts, rules of thumb, typologies (organization of facts), empirical laws

o empirical laws: statements of regularity or correspondence between phenomenon

§ rare in the social sciences

- most “facts” are theory-laden

o e.g., democracy v. dictatorships; great powers v. minor powers

Theories provide causal explanations

- explanations are not descriptions

- Relate specific events to general class of events

Theory: set of interrelated generalizations (axioms) about a range of phenomena

- specifies independent (explanans) and dependent variables (explandandum)

Independent Variables → Dependent Variables

X →Y

- axioms are abstractions about empirical world

- deduce hypotheses from axioms

o hypotheses: prediction about relations between variables, usually stated in if, then format

§ If two states are democracies, then they will not fight each other

Ad-hoc hypothesis testing is inferior to theory-generated hypotheses

Theories identify causes v. correlations

Correlation: regular association between variables

- Not all correlations reflect causal relationships

- May be useful to predict (e.g, contiguity and warfare)

Three Views of Causation

1) Covering Law(C. Hempel):

To say that C is the cause of E, is to say that C is regularly followed by E (C → E)

Necessary and sufficient conditions

- C is necessary for E; without C, no E

- C is sufficient for E; if C, then E

2) Probabilistic Causes: C → E some percentage of the time (e.g., smoking as a cause of cancer)

3) Causal Mechanisms: processes (often unobservable) by which C generates E.

Contingency and Causation

- difficult to find necessary and sufficient conditions

- Contingency (chance) plays a large role

Testing Theories

Falsification

Try to falsify theories: what evidence would indicate that the theory is wrong?

Naïve v. Sophisticated ‘Falsification’

Naïve view: a single wrong prediction falsifies a theory

Sophisticated view: despite wrong predictions, a theory should be accepted until a new theory is developed that

- explains what the prior theory explained

- accounts for the wrong predictions

Multiple causation is prevalent

Different causal forces may produce the same effect independently or interactively, so searching for a single or primary cause may be futile

- war may result from misperceptions, aggression, and fear of future weakness

- increased trade, social contacts, institutions and democracy may reduce war

- economic cooperation may result from influence of domestic interests and socialization to international norms

Independent Causes

X → E₁

Y → E₁

Z → E₁

Interactive Causes

Too Many Variables, Too Few Cases

Many events in IR are rare (e.g., wars)

- Between 1816-1997, there were 79 inter-state wars, and 401 wars of all types, or 4 wars per decade (Sarkees, et al)

- The number of states has increased over time so that relative frequency of wars (wars per system member) has declined

- Looking at all state-dyads for each year since World War II generates a data set with 303,814 observations, of which only 0.3 percent, or 1,042 dyads, were at war (King & Zeng)

There are often more plausible causal variables than can accurately be tested given the number of cases.

Methodological Errors and Faulty Inferences

Selecting on the dependent variable:

- examine (select) cases where the event or action occurred, not where it was absent.

Hypothesis: Evil dictators cause wars.

Researchers studies wars; ignores non-wars

Want to find variation in independent and dependent variables.

Post Hoc, Ergo Propter Hoc (False Cause) Fallacy

- conflate temporal sequence with causal sequence

- e.g., arms build-up preceded war, therefore arms build-ups caused the war?

Revealed Preferences

- examine actions taken and impute motives from action

- but action may reflect constraints, not motives

- problem for interest-based theories

Levels of Analysis

Locates level of the causal variable

1. Individual

2. State (political system, bureaucracy)

3. System

Many IR scholars favor systemic explanations

- parsimony: can explain more with less

- interactive or ‘strategic’ component: what is the other individual, bureaucracy, or state attempting to do?