Chapter 14: Mill’s Method

  

Chapter 14: Mill’s Method

•   In science and everyday life, we think a lot about causes and effects. Accurately determining causes and effects is not a simple task. We can often confuse the two, or misidentify one because we lack sufficient information. Mill’s methods are attempts to isolate a cause from a complex event sequence. 

•   They were formulated by the famous English philosopher John Stuart Mill (1806- 1873), who wrote on a wide range of topics from logic and language to political philosophy. 

•   Suppose one is interested in determining what factors play a role in causing a specific effect, E, under a specific set of circumstances. Let us now look at the five rules one by one and see how they are applied. 

1.      THE METHOD OF AGREEMENT: If two or more situations leading to an effect E have only one event C in common, then C is the cause of E. 

•   Suppose your family went to a dinner buffet and all of you had stomach problems afterwards. Different people might have eaten different things, but suppose the only thing that everybody has eaten was raw oysters. It would be reasonable to infer that the oysters caused stomach problems. 

•   Consider patients who suffer from AIDS, which weakens the immune system and makes deadly infections more likely. Despite their different backgrounds, it was discovered that all AIDS patients have been infected by HIV. So the conclusion was that HIV causes AIDS. 

•   For another example, four students come to Ms Maya with indigestion, and she questions each about what they had for lunch. The first had pizza, coleslaw, orange juice, and a cookie; the second had a hot dog and french fries, coleslaw, and iced tea; the third ate pizza and coleslaw and drank iced tea; and the fourth ate only french fries, coleslaw, and chocolate cake. Ms Maya, of course, concludes that "Eating coleslaw caused indigestion."

Situation	Candidate Causes			Effect
	A	B	C	E
1	✔️	❌	✔️	✔️
2	❌	✔️	✔️	✔️
3	✔️	✔️	✔️	✔️

2.      THE METHOD OF DIFFERENCE: If one group of situations leads to an effect E, but another group does not, and the only difference between the two groups is that C is present in the former but not the latter, then C is the cause of E.

•   Suppose a mobile phone is not working, but it works fine when the battery is replaced. Since the only thing that is different is the battery, it was probably the cause of the problem. 

•   For another example,  suppose that only two students arrive at Maya’s office. The two are roommates who ate together, but one became ill while the other did not. The first had eaten a hot dog, french fries, coleslaw, chocolate cake, and iced tea, while the other had eaten a hot dog, french fries, chocolate cake, and iced tea. Again, Ms Maya concludes that the coleslaw is what made the first roommate ill.

Situation	Candidate Causes			Effect
	A	B	C	E
1	✔️	❌	✔️	✔️
2	✔️	❌	✔️	✔️
3	✔️	❌	❌	❌

3.      THE JOINT METHOD: Suppose one group of situations leads to an effect E, but another group does not. If C is the only factor common to situations in the first group, and it is also the only factor that is absent from all the situations in the second group, then C is the cause of E. 

•   Assume that eight students come to Ms Maya: four of them suffered from indigestion, and with each of these four there is another who did not. Each pair of students had exactly the same lunch, except that everyone in the first group ate coleslaw and no one in the second group did. The Nurse arrives at the same conclusion.

•   This situation is an example of Mill's Joint Method of Agreement and Difference: the first four students are evidence that everyone who got ill had eaten coleslaw, and the four matching pairs are evidence that only those who got ill had eaten coleslaw. This is a powerful combination of the first two methods since it tends to support our notion that genuine causes are necessary and sufficient conditions for their effects.

Situation	Candidate Causes				Effect
	A	B	C	D	E
1	✔️	✔️	✔️	✔️	✔️
2	✔️	❌	✔️	❌	✔️
3	✔️	❌	❌	✔️	❌

4.      THE METHOD OF CONCOMITANT VARIATIONS: If variation in some factor C is followed by variation in an effect E, then C is the cause of E.

•    The general idea behind this principle is that an effect can change by changing its cause. Suppose the number of people suffering from asthma attacks goes up when air pollution becomes very bad. When air quality improves, the number of such attacks drops. Because there is this correlation in variation, we conclude that air pollution causes asthma attacks. 

•   Suppose that the Nurse sees five students: the first ate no coleslaw and feels fine; the second had one bite of coleslaw and felt a little queasy; the third had half a dish of coleslaw and is fairly ill; the fourth ate a whole dish of coleslaw and is violently ill; and the fifth ate two servings of coleslaw and had to be rushed to the hospital. The conclusion is again that coleslaw caused indigestion.

Situation	Candidate Causes	Effect
	C	E
1	✔️	✔️
2	✔️✔️	✔️✔️
3	✔️✔️✔️	✔️✔️✔️

5.      THE METHOD OF RESIDUES: If a set of conditions causes a range of effects, and some of the effects can be explained by some of the earlier conditions, then the remaining effects are caused by the other remaining conditions. 

•   Suppose that Ms Maya, during prior investigations of student illness, has already established that pizza tends to produce a rash and iced tea tends to cause headaches. Today, a student arrives at the Nurse's office complaining of headache, indigestion, and a rash; this student reports having eaten pizza, coleslaw, and iced tea for lunch. Since she can account for most of the student's symptoms as the effects of known causes, Ms Maya concludes that the additional effect of indigestion must be caused by the additional circumstance of eating coleslaw

•   The idea behind the method of residues is to identify causes by elimination. Suppose you discover that two books are missing from your room, and only two people have been there recently. You ask one of them and he confesses that he "borrowed" just one book without asking. If you trust his answer it would then be reasonable to conclude that the other visitor had taken the second book.

•     LIMITATIONS OF MILL’S METHODS

•     Although Mill's Methods are an important component of serious investigation of natural phenomena, they have significant limitations. Careful application of these methods succeeds only when every relevant antecedent (predecessor) circumstance is taken into account, and that is impossible to guarantee in advance.

•     The true cause might not be any of the candidate's causes. Mill's methods start with a preselected set of candidate causes. If the true cause of the effect is not among them, obviously the five rules will not yield the correct result. Of course, we can always expand or change the list of candidate causes. But Mill's methods will be more effective when we already have a good idea about what the possible causes of the effect might be. 

•      The effect might have more than one cause. Mill's methods can fail spectacularly (impressive) when more than one candidate cause can bring about the same effect. It might be true that everybody who ate oysters got sick. But perhaps the oysters were fine, and it was the salad and the noodles that had gone bad. It just so happened that everyone ate either salad or noodles. But the method of the agreement will give us the wrong result.

•     Causation can be indeterministic. Throwing a rock at a window might cause it to break, but not always. A heart attack can result in death, but again not every time. Mill's methods can give the wrong results when dealing with probabilistic causes. Take the joint method for example, where the cause is supposed to be the condition that occurs when and only when the effect is present. This rules out causes that do not determine their effects but that make them highly probable. 

Having noted some of the limitations of Mill's methods, it should be said that they remain important tools of scientific reasoning. They do not guarantee that you will find the real cause of an effect, but they are good formulas to try out in causal investigations.

END OF THE PART