of each commitment.
A. Alchian, "Uncertainty, Evolution, and Economic Theory,'
Journal of Economic Theory, 195
Sometimes our best response to a causal claim is
• Did you ever think that might just be coincidence?
• Just because it followed doesn't mean it was caused by . . . .
• Have you thought about another possible cause, namely . . .
• Maybe you've got the cause and the effect reversed.
• Not always, but maybe under some conditions . . . .
310 CHAPTER 15 Cause and Effect
Exercises for Section A
1. What criteria are necessary for there to be cause and effect?
2. Why isn't a perfect correlation enough to justify cause and effect? Give an example.
3. Comparable to the unstated premises of an argument, what do we call the claims that
must be true for a causal claim to be true?
4. What real problem in establishing cause and effect is usually mis-stated as:
"That's not close enough in space and time to be the cause"?
5. Dick makes a causal claim. Zoe says it's just post hoc ergo propter hoc reasoning.
How can he show her that he's right?
6. Explain why it's not amazing that every day a few dream predictions come true.
7. When should we trust authorities rather than figure out a cause for ourselves?
B. Examples
We have necessary conditions for there to be cause and effect. What about sufficient
conditions? In practice, all we can do is check that the necessary conditions hold,
being careful not to make one of the obvious mistakes, even if we're not satisfied
that we can exactly state sufficient conditions for there to be cause and effect.
Are the following examples of cause and effect?
Example 1
The cat made Spot run away.
Cause: What is the cause? It's not just "the cat." How can we describe it with a
claim? Perhaps "A cat meowed close to Spot."
Effect: Spot ran away.
Cause and effect each happened: The effect is clearly true. The cause is highly
plausible: Almost all things that meow are cats.
Cause precedes effect: Yes.
SECTION B Examples 311
It is (nearly) impossible for the cause to be true and effect false: What needs to be
assumed as "normal" here? Spot is on a walk with Dick. Dick is holding the leash
loosely enough for Spot to get away. Spot chases cats. Spot heard the cat meow.
We could go on, but this seems enough to guarantee that it's unlikely that the cat
could meow near Spot and Spot not chase it.
The cause makes a difference: Would Spot have run away even if the cat had not
meowed near him? Apparently not, given those normal conditions, since Dick seems
surprised that he ran off. But perhaps he would have even if he hadn't heard the cat,
if he'd seen it. But that apparently wasn't the case. So let's revise the cause to be
"Spot wasn't aware a cat was near him, and the cat meowed close to Spot, and Spot
heard it." Now we can reasonably believe that the cause made a difference.
Is there a common cause! Perhaps the cat was hit by a meat truck and lots of meat
fell out, and Spot ran away for that? No, Spot wouldn't have barked. Nor would he
have growled.
Perhaps the cat is a hapless bystander in a fight between dogs, one of whom is
Spot's friend. We do not know if this is the case. So it is possible that there is a
common cause, but it seems unlikely.
Evaluation: We have good reason to believe the original claim on the revised
interpretation that the cause is "Spot wasn't aware a cat was near him, and the cat
meowed close to Spot, and Spot heard it."
These are the steps we should go through in establishing a causal claim.
If we can show that one of them fails, though, there's no need to check all the others.
Example 2 Maria caused the traffic accident.
Analysis We're interested in who or what was involved in the cause when we go
about assigning blame or fault. But it's not just that Maria exists. Rather:
Cause: Maria didn't pay attention.
Effect: The cars collided.
Is this really cause and effect? Let's assume that these claims are true.
It seems the cause preceded the effect. But did the cause make a difference?
If Maria had been paying attention, would the cars still have collided? Since she
was broadsided by a car running through a red light where a line of cars blocked her
vision, we would say that it didn't matter that she was changing a CD at the time:
The cars would have collided even if she had been paying attention, or so we all
imagine. The purported cause didn't make a difference. It's not cause and effect.
Example 3 Lack of rain caused the crops to fail in the Midwest in 2000.
Analysis We've talked about causes as if something active has to happen. But
almost any claim that describes the world could qualify as a cause.
312 CHAPTER 15 Cause and Effect
Cause: There was no rain in the first part of the year 2000.
Effect: The crops failed.
Is this cause and effect? We better check the meteorological records and ask some
farmers if there wasn't some other cause, perhaps locusts.
Example 4 Oxygen in the laboratory caused an explosion.
Analysis This seems right, but what are the normal conditions? Harry works in a
laboratory where there's not supposed to be any oxygen. The materials are highly
flammable. He has to wear breathing gear. Harry knows that matches won't light in
the laboratory.
But it certainly isn't "normal" that Harry carried matches with him into the
laboratory for a joke with a friend and struck a match. Nor is it normal that there
was a leak in his face mask.
When several claims together are taken jointly as the cause, we say that each is
(describes) a cause or that each is a causal factor.
Example 5 Running over nails causes your tires to go flat.
Analysis This is a general causal claim. But it's false. Lots of times we run over
nails and our tires don't go flat.
But sometimes they do. What's correct is:
Running over nails can cause your tires to go flat.
That is, if the conditions are right, running over a nail will cause your tire to go flat.
The difference between causes and can cause is the difference between the
normal conditions. We'll look at how to evaluate claims like this in Section D.
Example 6 When more and more people are thrown out of work, unemployment results.
Calvin Coolidge
Analysis You don't have to be smart to be President. This isn't cause and effect;
it's a definition.
Example 7 Birth causes death.
Analysis In some sense this is right. But it seems wrong. Why?
SECTION B Examples 313
What's the cause? What's the effect? The example is a general causal claim
covering every particular claim like "That this creature was born caused it to die."
We have lots of inductive evidence: Socrates died. My dog Juney died. My
teacher in high school died. President Kennedy died. . . .
The problem seems to be that though this is true, it's uninteresting. It's tracing
the cause too far back. Being born should be part of the normal conditions when we
have the effect that someone died.
Example 8 Maria: Fear of gettin
g fired causes me to get to work on time.
Analysis How can we describe the purported cause with a claim?
Cause: Maria is afraid of getting fired.
Effect: Maria gets to work on time.
Is it possible for Maria to be afraid of getting fired and still not get to work on
time? Certainly, but not, perhaps, under normal conditions: Maria sets her alarm;
the electricity doesn't go off; there isn't bad weather; Maria doesn't oversleep; . . . .
But there's something odd in calling these the normal conditions: Isn't it supposed
to be because she's afraid that Maria makes sure these claims will be true, or that
she'll get to work even if one or more is false?
In that case how can we judge whether the relationship between the purported
cause and effect is valid or strong? That Maria gets to work regardless of conditions
that aren't normal is what makes her consider her fear to be the cause.
Subjective causes are often a matter of feeling, some sense that we control what
we do. They are often too vague for us to classify as true or false.
Example 9 Dick: Hold the steering wheel.
Zoe: What are you doing? Stop! Are you crazy?
Dick: I'm just taking my sweater off.
Zoe: I can't believe you did that. It's so dangerous.
Dick: Don't be silly. I've done it a thousand times before.
(crash . . . later. . .)
Dick: You had to turn the steering wheel!? That made us crash.
Analysis The purported cause: Zoe turned the steering wheel. The effect: The car
crashed. The necessary criteria are satisfied.
But as they say in court, Zoe's turning the steering wheel is a foreseeable
consequence of Dick making her take the wheel, which is the real cause. The
normal conditions are not just what has to be true before the cause, but also what
will normally follow the cause.
Example 10 Dick: Wasn't that awful what happened to old Mr. Grzegorczyk?
Zoe: You mean those tree trimmers who dropped a huge branch on
him and killed him?
314 CHAPTER 15 Cause and Effect
Dick: You only got half the story. He'd had a heart attack in his car
and pulled over to the side. He was lying on the pavement
when the branch hit him—he would have died anyway.
Zoe: But I heard his wife is going to collect from the tree company.
Analysis What's the cause of death? Mr. Grzegorczyk would have died anyway.
So the tree branch falling on him wouldn't have made a difference.
But the tree branch falling on him isn't a foreseeable consequence, part of the
normal conditions of his stumbling out of his car with a heart attack. As they say in
court, it's an intervening cause. Juries, usually made up of people like you, will be asked to decide what is the cause of Mr. Grzegorczyk's death. There's no clear
answer, though these kinds of cases have been debated for centuries.
Example 11 Sunspots cause stock prices to rise.
Analysis Suppose your finance teacher tells you this general causal claim, and she
backs it up with data showing a very good correlation between the appearance of
large sunspots and rises in the Dow Jones index. But a correlation, though needed
for a general causal claim, doesn't establish cause and effect by itself. It's hard to
imagine a common cause, but coincidence can't be ruled out. If we look around the
world long enough, we'll eventually find some phenomenon that can be correlated
to the rise and fall in stock prices. Even if there were a very exact correlation
between the size of the sunspots and the percentage of increase in the Dow Jones
average two days later, we still want a theory—normal conditions that give us a way
to trace how the sunspots cause the price rises—before we accept that this is cause
and effect.
Example 12 The Treaty of Versailles caused World War II.
Analysis The cause: The Treaty of Versailles was agreed to and enforced. The
effect: World War II occurred.
To analyze a conjecture like this an historian will write a book. The normal
conditions have to be spelled out. You have to show that it was a foreseeable
consequence of the enforcement of the Treaty of Versailles at the end of World
War I that Germany would re-arm.
But was it foreseeable that Chamberlain would back down over Hitler's
invasion of Czechoslovakia? More plausible is that the signing of the Treaty of
Versailles is a cause, not the cause of World War II.
Example 13 Poltergeists are making the pictures fall down from their hooks.
Analysis To accept this, we have to believe that poltergeists exist. That's dubious.
Worse, it's probably not testable: How could you determine if there are
poltergeists? Dubious claims that aren't testable are the worst candidates for causes.
EXERCISES for Sections A and B 315
Exercises for Sections A and B
Here is an exercise Tom did on cause and effect.
I used Diabolic Grow on my roses and they grew great! I'll always use it.
Causal claim: (unstated) Diabolic Grow caused my roses to grow great.
Cause: The speaker put Diabolic Grow on his roses.
Effect: The roses grew great.
Cause and effect each happened? Apparently so.
Cause precedes efect? Yes.
It's (nearly) impossible for the cause to be true and effect false! Hard to say.
Cause makes a difference! It seems so, but did the cause really make a
difference? Maybe they would have grown great anyway. Some years
that happens when it rains at just the right time in the spring.
Common cause! For sure, no.
Evaluation: You'd need a lot more evidence to believe the claim.
"Excellent! you're thinking critically.
For the exercises here, find the causal claim. Then evaluate it, explaining why it's plausible
or clearly wrong, or whether you need more information to evaluate it by answering:
Causal claim:
Cause:
Effect:
Cause and effect each happened?
Cause precedes efect?
It's (nearly) impossible for the cause to be true and effect false?
Cause makes a difference?
Common cause?
Evaluation:
1. Maria: I had to slam on the brakes because some idiot pulled out in front of me.
2.
Suzy: My feet hurt so bad the other day when I was cheerleading. My feet have never
hurt at the other cheerleading events, but I was wearing new shoes. So it must have been
my new shoes.
3.
Dick: Ooh, my stomach hurts.
Zoe: Serves you right. You really pigged out on the nachos and salsa last night.
They always give you a stomachache.
4. Marriage is the chief cause of divorce.
5. I've got to go to the game. The only time I wasn't in the bleachers this season, they lost.
6. Hazards are one of the main causes of accidents. (OSHA, "Safety with Beef Cattle")
316 CHAPTER 15 Cause and Effect
7. Zoe: The dark sky makes me really depressed today.
8. Dick: Boy, are you red.
Zoe: Ouch! I got a terrible sunburn because the sun was so strong yesterday.
9. The emphasis on Hollywood figures in the media causes people to use drugs because
people want to emulate the stars.
10. Maria: It's awful what's happened to Zeke.
Lee: Why? What happened? I haven
't seen him for ages.
Maria: He started using drugs. It's because he was hanging out with that bad bunch.
11. Lou's college education helped him get a high-paying job the year after he graduated.
12. Dick: Every day I run up this hill and it's no big deal. Why am I so beat today?
Zoe: It's 'cause you stayed out late and didn't get enough sleep.
13. Zoe: My life's a mess. I've never really been happy since all those years ago in high
school you told Sally that I killed Puff. She believed your stupid joke, and made
sure I wasn't a cheerleader. I'll never be a cheerleader. It's your fault I'm so
miserable now.
Dick: There, there.
14. Sex, drugs, and rock 'n roll are the causes of the decline in family values.
15. Suzy: Eating potato chips and sitting on the couch must be healthy. All the guys on
the football team do it.
16. Lee: Yesterday my neighbor said this spring has been the worst season ever for
allergies, but I told her I hadn't had any bad days. Then today I started sneezing.
Darn it—if only she hadn't told me.
17. Dick: Normally my pulse rate is about 130 after exercising on this bike.
Zoe: I can't believe you actually measure your heart rate! You're so obsessive.
Dick: But for the past week or so it's been about 105. That's odd.
Zoe: You stopped drinking coffee two weeks ago, remember?
18. He's stupid because his mother dropped him on his head when he was young.
19. A recent study shows that everyone who uses heroin started with marijuana. So smoking
marijuana causes heroin use.
20. Dr. E: My students don't like the material at the end of this course. That's why so many
have missed class the last two weeks of the course.
21. The swallows never come back to Capistrano except when there are a lot of people
waiting to see them there at the festival they have each year. They must come back
because they like the welcome.
22. Flo: Salad makes you fat. I know 'cause Wanda's really fat and is always eating salad.
23. (An advertisement by the Iowa Egg Council in the Des Moines International Airport)
Children who eat breakfast not only do better academically, but they also behave better.
Archives of Pediatric and Adolescent Medicine
SECTION C How to Look for the Cause 317
24. Gingrich: Liberalism led to Colorado massacre
In his first speech since leaving Congress, former House speaker Newt Gingrich blamed
Richard L Epstein Page 40