Other Considerations with Respect to Arguments
Aside from the above fallacies, there are several other things to consider when making and responding to arguments.
Correlation Does Not (Necessarily) Equal Causation—As we saw in the post hoc fallacy, the fact that A happens before B does not mean that A caused B. We should always be careful about making claims that movement in one variable is causing movement in another, or that the presence of one condition concurrently with another condition means that one caused the other.
Variables can be correlated with each other in several ways, only one of which is a causal relationship. Let’s say that variables A and B are correlated. One possibility is that A might in fact be causing B, or vice versa. Another possibility is that a third variable, C, is simultaneously causing A and B, without there being a direct relationship between A and B. Also, maybe A is actually causing C, which is then causing B, in which case C is called an intermediate variable. Finally, A and B may be correlated, but there is actually no relationship between them.
Appealing to Authority— Appealing to authority is another fallacy if the authority in question lacks credibility. We commit this fallacy when we make arguments and use evidence from sources who are not qualified in a given area, are religious texts or authorities who cannot be questioned, or whose motivation comes into question because of partisan interests or financial conflicts of interest. If I make an argument that my state should open more charter schools and I cite a statement from a state legislator who is also on the board of directors of a company that builds charter schools, I would have committed this fallacy. But there are many legitimate uses of appealing to authority. In fact, it is a common and totally valid strategy in making effective arguments, since the authority being cited may be more knowledgeable in a given area. And that’s what I’m going to do now. Robert Fogelin and Walter Sinnott-Armstrong wrote,
“Since some people stand in a better position to know things than others, there is nothing wrong with citing them as authorities. In fact, an appeal to experts and authorities is essential if we are to make up our minds on subjects outside our own range of competence.”[1]
Argument by Analogy—People often use analogies when they make an argument. An analogy is a linking or comparison of two things, typically for the purpose of explanation or clarification. The first President Bush said that Saddam Hussein was “worse than Hitler.” John Dean, President Nixon’s White House counsel, has written a book arguing that the skullduggery in the George W. Bush administration is “worse than Watergate”. Gay rights advocates argue that bans against gay marriage are akin to bans against interracial marriage that existed well into the 1960s.
Are these analogies helpful or not? Using a false or distorted analogy is clearly a fallacy and really should belong in the list of fallacies above. But a well-crafted analogy can be very useful. So the effectiveness of arguing by analogy is a matter of degree. If we want to judge the usefulness of an analogy linking A with B, we need to follow a few simple criteria:
Slippery Slope—As with arguing by analogy, improper use of a slippery slope argument can be fallacious. But slippery slope arguments can be effective if the arguer clearly lays out the progression down the slope. A slippery slope argument occurs when a person suggests that if we take one action it will lead to a chain of events resulting ultimately in disaster. A common one lately is the argument that says if we legalize gay marriage, we’ll have to legalize polygamy, and then pedophilia, incest, and bestiality. During the Cold War, the Domino Theory was a form of slippery slope argument for American involvement in the Vietnam War—If we don’t stop the commies there, then Laos and Cambodia will go red, followed by Thailand, Australia….and then we’ll be fighting the communists on the beaches of California (or at least the communists will be in a position to strangle us without actually invading). Of course, South Vietnam fell in 1975, but the subsequent cascade of dominoes failed to happen. Slippery slopes do happen, so this form of argument can be used effectively. For example, a white Southerner in the 1940s who argued that any cracks in the edifice of segregation will lead to its complete collapse (the end of segregated schools, the end of bans on interracial marriage, the end of employment discrimination, etc.) was obviously predicting exactly what happened.
Generalization vs. Overgeneralization—Perhaps the trickiest part of sophisticated argumentation is the issue of generalization versus overgeneralization. We are often counseled against making generalizations, particularly if they involve stereotyping people based on their race, gender, religion, national origin, and so forth. This is good advice, for such stereotypes are as often wrong as they are accurate, so it’s better to treat people as individuals rather than representatives of some larger group. On the other hand, very little social science analysis could take place without some well-considered and carefully stated generalizations. If we gather and analyze data that falls into certain patterns, we would be remiss as political scientists if we failed to point that out. So I am on safe ground if I make the generalization that older Americans tend to vote at higher rates than do younger Americans, because that’s the story the data tell after every election. Notice that I phrased the generalization in a nuanced way using the word “tend”, instead of distorting reality by saying “Old people vote and young people don’t.”
[1] Robert J. Fogelin and Walter Sinnott-Armstrong, Understanding Arguments. An Introduction to Informal Logic. San Diego: Harcourt-Brace Jovanovich, 1991. Page 117.
[2] Thompson, Habits of the Mind. Page 123.
Correlation Does Not (Necessarily) Equal Causation—As we saw in the post hoc fallacy, the fact that A happens before B does not mean that A caused B. We should always be careful about making claims that movement in one variable is causing movement in another, or that the presence of one condition concurrently with another condition means that one caused the other.
Variables can be correlated with each other in several ways, only one of which is a causal relationship. Let’s say that variables A and B are correlated. One possibility is that A might in fact be causing B, or vice versa. Another possibility is that a third variable, C, is simultaneously causing A and B, without there being a direct relationship between A and B. Also, maybe A is actually causing C, which is then causing B, in which case C is called an intermediate variable. Finally, A and B may be correlated, but there is actually no relationship between them.
Appealing to Authority— Appealing to authority is another fallacy if the authority in question lacks credibility. We commit this fallacy when we make arguments and use evidence from sources who are not qualified in a given area, are religious texts or authorities who cannot be questioned, or whose motivation comes into question because of partisan interests or financial conflicts of interest. If I make an argument that my state should open more charter schools and I cite a statement from a state legislator who is also on the board of directors of a company that builds charter schools, I would have committed this fallacy. But there are many legitimate uses of appealing to authority. In fact, it is a common and totally valid strategy in making effective arguments, since the authority being cited may be more knowledgeable in a given area. And that’s what I’m going to do now. Robert Fogelin and Walter Sinnott-Armstrong wrote,
“Since some people stand in a better position to know things than others, there is nothing wrong with citing them as authorities. In fact, an appeal to experts and authorities is essential if we are to make up our minds on subjects outside our own range of competence.”[1]
Argument by Analogy—People often use analogies when they make an argument. An analogy is a linking or comparison of two things, typically for the purpose of explanation or clarification. The first President Bush said that Saddam Hussein was “worse than Hitler.” John Dean, President Nixon’s White House counsel, has written a book arguing that the skullduggery in the George W. Bush administration is “worse than Watergate”. Gay rights advocates argue that bans against gay marriage are akin to bans against interracial marriage that existed well into the 1960s.
Are these analogies helpful or not? Using a false or distorted analogy is clearly a fallacy and really should belong in the list of fallacies above. But a well-crafted analogy can be very useful. So the effectiveness of arguing by analogy is a matter of degree. If we want to judge the usefulness of an analogy linking A with B, we need to follow a few simple criteria:
- The number of ways A and B are similar.
- The relevance of the similarities between A and B.
- The relevance of the differences between A and B.[2]
Slippery Slope—As with arguing by analogy, improper use of a slippery slope argument can be fallacious. But slippery slope arguments can be effective if the arguer clearly lays out the progression down the slope. A slippery slope argument occurs when a person suggests that if we take one action it will lead to a chain of events resulting ultimately in disaster. A common one lately is the argument that says if we legalize gay marriage, we’ll have to legalize polygamy, and then pedophilia, incest, and bestiality. During the Cold War, the Domino Theory was a form of slippery slope argument for American involvement in the Vietnam War—If we don’t stop the commies there, then Laos and Cambodia will go red, followed by Thailand, Australia….and then we’ll be fighting the communists on the beaches of California (or at least the communists will be in a position to strangle us without actually invading). Of course, South Vietnam fell in 1975, but the subsequent cascade of dominoes failed to happen. Slippery slopes do happen, so this form of argument can be used effectively. For example, a white Southerner in the 1940s who argued that any cracks in the edifice of segregation will lead to its complete collapse (the end of segregated schools, the end of bans on interracial marriage, the end of employment discrimination, etc.) was obviously predicting exactly what happened.
Generalization vs. Overgeneralization—Perhaps the trickiest part of sophisticated argumentation is the issue of generalization versus overgeneralization. We are often counseled against making generalizations, particularly if they involve stereotyping people based on their race, gender, religion, national origin, and so forth. This is good advice, for such stereotypes are as often wrong as they are accurate, so it’s better to treat people as individuals rather than representatives of some larger group. On the other hand, very little social science analysis could take place without some well-considered and carefully stated generalizations. If we gather and analyze data that falls into certain patterns, we would be remiss as political scientists if we failed to point that out. So I am on safe ground if I make the generalization that older Americans tend to vote at higher rates than do younger Americans, because that’s the story the data tell after every election. Notice that I phrased the generalization in a nuanced way using the word “tend”, instead of distorting reality by saying “Old people vote and young people don’t.”
[1] Robert J. Fogelin and Walter Sinnott-Armstrong, Understanding Arguments. An Introduction to Informal Logic. San Diego: Harcourt-Brace Jovanovich, 1991. Page 117.
[2] Thompson, Habits of the Mind. Page 123.