Last week I wrote about fallacious reasoning in regard to checking the reasoning that politicians employ during campaign debates and assignments you may have to complete during your collegiate career. I gave an introduction to fallacies by quoting Robert J. Gula; I then proceeded to examine a specific kind of fallacy: the Ad Hominem Fallacy.
For today's post, I will continue to explore fallacies.
One important distinction to make when you form an argument is that correlation is not causation. In Gula's book, he identifies three kinds of causes: necessary, sufficient, and contributory (120).
A necessary cause is one that is required to satisfy a particular state of affairs or situation. For example, water is a necessary cause for human life to exist. To take another example, in order to make a three-cheese pizza, one needs to have access to at least three different kinds of cheese.
I cannot exist if I don't drink water. Likewise, I cannot make a three-cheese pizza if I do not have access to at least three different kinds of cheese.
Notice that a necessary cause is something that we need in order to achieve some goal or some state of affairs, but in the examples that I gave above, we still need other causes in order to satisfy our goal or state of affairs. We need three different kinds of cheese in order to make a three cheese pizza, but just because we have three different kinds of cheese does not mean we can make a three-cheese pizza.
A three-cheese pizza requires other ingredients. Pizza usually has some kind of tomato sauce on it. Pizza also usually has some kind of bread or crust that provides the base for the cheese and sauce. In the same way, the existence of human beings requires more than water. The existence of human beings also requires some kind of food and a certain range of body temperature. Thus, a necessary cause is required to achieve some goal or state of affairs, but it is not sufficient.
A sufficient cause is one that is sufficient to cause some state of affairs. In Nonsense, Gula writes: "If Y always occurs when X is present, then X is a sufficient condition of Y" (120).
For example, hitting my shin against a piece of furniture will cause a bruise. In other words, hitting my shin against a piece of furniture is enough by itself to cause my shin to bruise. But notice that there are many other ways to get a bruise; in other words, there are thousands of sufficient causes for a person to be bruised: If a baseball traveling at high velocity hits my shin, it will cause a bruise; if someone of suitable size punches me in the eye with a clenched fist, it will cause a bruise; if I fall off my bike and land on my knee, it will cause bruise. All these examples are sufficient causes for a bruise, but none of them are necessary.
The final kind of cause is a contributory cause. A contributory cause is a cause which is neither a necessary cause nor a sufficient cause. A contributory cause occupies a kind of middle ground between the two other kinds of causes. It contributes to a state of affairs.
For example, eating a lot of saturated fat may contribute to a person having a heart attack. But since some people who eat a lot of saturated fat never have a heart attack, eating a lot of saturated fat is not a sufficient cause for a heart attack. And since some people have heart attacks without eating a lot of saturated fat, eating a lot of saturated fat is not a necessary cause. Instead, eating a lot of saturated fat is one factor that could contribute to a person having a heart attack.
To summarize the distinction between these different kinds of causes, Gula gives the following principles:
1. If X does not happen, then Y will not happen. (X is a necessary cause of Y.)
2. If X happens, then Y will happen. (X is a sufficient cause of Y.)
3. If X happens, then Y may happen. (X is a contributory cause.)
A lot of fallacious reasoning is a result of people misapplying these three different kinds of causes. To illustrate how fallacies are disseminated by misapplying these three different kinds of causes, let's think of the way that politicians speak.
A politician might say "If I am not elected, then the economy will not grow." But growing the economy is a very complicated phenomenon.
The failure to elect a single politician will not in itself inhibit the economy from growing. The more rational formulation would be for the politician to say "If my opponent is elected, then it's possible that the economy will not grow."
This is the more logical formulation because it treats the election of a single politician as a single, contributory cause. Most voters, regardless of their political beliefs, will probably grant the truth of the latter statement. "Sure," we might say, "if I vote for your opponent and he is elected, then it's possible that the economy will not grow. But it's also possible that by electing your opponent the country will experience an economic boom. It's possible that the United States balances its budget and the unemployment rate drops to 4 percent. So what?"
Furthermore, suppose that a politician says, "If I am elected, then the
economy will grow." Of course, once again, we know that growing the
economy is a complicated process. It includes billions of variables--some of which remain unknown until the moment they are instantiated, some of which may remain unknown in perpetuity.
The election of a single politician is
not a sufficient cause to grow the economy. What the politician should
say, if the politician wants to respect the strictures of logic and good
reasoning, is "If I am elected, then it's possible that the economy
Of course, broadcasting such a slogan will not play as well
in the soundbites during the evening news.
When politicians attempt to connect phenomena, it would be logical for them, in most cases, to speak about contributory causes. But campaigns that run on logic and sound reasoning have a tendency to end with concession speeches.
Since most of us are not running for political office, we don't need to worry about making a good soundbite. Instead, let's worry about becoming better writers, better thinkers.
Let's avoid, as best we can, confusing the different kinds of causation.