(1) Probabilistic Causal Claims

Review:

• unknown deterministic processes
• recognizing probabilistic claims

(2) Types of Causes

• "structural" causes
• "triggering" events
• "random" causes

When we seek to investigate effects of causes (we look at causes and what effects they produce),

does it make sense to speak deterministically?

e.g., does it make sense to claim:

• "Wood framing causes buildings to burn down (every time)?"
• "Short circuits cause buildings to burn down (every time)?"
• "Presence of indoor gas furnaces cause buildings to burn down (every time)?"

When we are interested in effects of causes, we usually make probabilistic causal claims

probabilistic causal claims

are claims that the presence of a cause \(C\) makes an effect \(E\) more or less likely to occur.

• In contrast to deterministic causal claims this implies
  • effect \(E\) can happen when \(C\) is absent
  • effect \(E\) may not happen when \(C\) is present
• NOT a claim that politics has some inherent randomness (not quantum mechanics)
  • recall, coin flips are deterministic but seem random.

If the social/political world is actually deterministic, why do we need probabilistic claims?

Because causality is complex and not fully known:

Cause \(C\) might produce an effect \(E\)…

• only when multiple other necessary conditions are met (conjunctural)
• that differs depending on other factors \(C_2\) etc. (conjunctural: conditional)
• but so do multiple other sets of necessary conditions (multiple and conjunctural)

\(C\) appears to only cause a change in the probability or likelihood of seeing the effect \(E\).

If the causal relations were inherently probabilistic:

the same exact cause happening to the same exact unit under exactly the same conditions would produce different results with some probability

This seems unreasonable, except in quantum physics.

Determinism may imply no free will (still an open debate in philosophy)

Voter turn out:

• Low in many countries
• Low in many local, off-year elections.

How can you get people to vote?

Exposure "shame" increases probability of voting by 2/3

1. because multiple, conjunctural causation
   • (shame AND degree) OR (shame AND urban)
2. because necessary condtions present for only 2/3 of people.
   • some people both lack degree, live in rural area.

Which is a probabilistic causal claim?

A) It's probably true that leftwing government reduce student tuition fees

B) Electing a leftwing, rather than rightwing, government increases the probability likely that tuition fees wil be reduced

C) Tuition fees are reduced more frequently under leftwing governments than rightwing governments

Which is a probabilistic causal claim?

A) It's probably true that leftwing government reduce student tuition fees

B) Electing a leftwing, rather than rightwing, government increases the probability likely that tuition fees wil be reduced

C) Tuition fees are reduced more frequently under leftwing governments than rightwing governments

Not every probabilistic statement is causal

1. "Oppression is likely to cause a rebellion"

• Says oppression is probably a cause out rebellion
• Should say: cause \(C\) changes likelihood of outcome \(E\)

2. "A rebellion is more likely to occur if the population is oppressed"

• Says we are more likely to see rebellion where population is oppressed
• Not clearly causal; just a descriptive claim.

probabilistic causal claims follow from a focus on effects of causes

Implies different "varieties" of causes

"structural causes"

and

"triggering events" or "triggers"

structural causes

From the perspective of complex causality (multiple and conjunctural), structural causes are those causes that

• are necessary in many sets of sufficient conditions
• thus, structural causes contributed to an effect/outcome under many different conditions

triggering events

From the perspective of complex causality (multiple and conjunctural), triggering events are those causes that

• are only necessary in one or a few sets of sufficient conditions
• thus, they generate the outcome in combination with some structural causes
• thus, triggering events cause an effect/outcome under highly specific conditions
• triggers are substitutable: some trigger needed, specific trigger may not be
  • may fail to meet counterfactual definition of causality (I'm skeptical of this argument)

Which are structural causes? Which are triggers?

Global Financial Crisis of 2008

1. Growing Wealth Inequality

2. Weak Regulatory Structures

3. Slow Wage Growth

4. Collapse of Lehman Brothers investment bank

Which are structural causes? Which are triggers?

Global Financial Crisis of 2008

1. Growing Wealth Inequality (structural)

2. Weak Regulatory Structures (structural)

3. Slow Wage Growth (structural)

4. Collapse of Lehman Brothers investment bank (trigger)

Social science generally interested in structural causes.

We want prediction

• This requires finding patterns/regularities in different contexts
• triggering events may not be relevant across contexts
• structural causes are.

Want causal patterns/regularities