INTERMISSION
From Why to How
5
WHY TO LEARN HOW TO ARGUE
MANY PEOPLE BELIEVE THAT they already know how to argue: They simply proclaim a reason for their position. They also believe that they are good at it: The reasons that they give seem strong to them. And they believe that they can tell a bad argument from a good one: Just think about it.
If arguing and assessing arguments were really this easy, there would be no need for the rest of this book. You would not need to learn how to argue. You would already know how.
Arguing well is not that simple. Indeed, most people are pretty bad at arguing in many circumstances. They make the same mistakes over and over again. These tendencies do not result from ignorance or lack of intelligence. Even smart people endorse and get fooled by bad arguments if they have not been trained properly. That is why we all need to work hard at learning how to argue.
DO YOU WANT TO MAKE A DEAL?
Paradoxes show how much we have to learn. This became evident when Marilyn vos Savant, a famous mathematician, challenged her readers to solve the Monty Hall problem (named after the host of the American game show Let’s Make a Deal; also known as the Three-Door problem):
Suppose you’re on a game show, and you’re given the choice of three doors: Behind one door is a car; behind the others, goats. You pick, say No. 1, and the host, who knows what’s behind the doors, opens another door, say No. 3, which has a goat. He then says to you, “Do you want to pick door No. 2?” Is it to your advantage to switch your choice?1
Most readers, including several mathematics professors, answered that there is no advantage in switching. This reply seems correct because only two doors (No. 1 and No. 2) remain closed, you know that one hides a goat and the other hides a car, and you seem to have no reason to think that one door is more likely than the other to hide the car.
This appearance is misleading. To see why, recall that there are only three possible arrangements behind the three doors in order: car-goat-goat, goat-car-goat, goat-goat-car. If you pick door No. 1 initially, and then Monty Hall reveals a goat behind one of the other doors, you will win a car 2 times out of 3 by switching. You lose only in the first order (car-goat-goat), but you win in both of the other two orders (goat-car-goat and goat-goat-car).
Experts now agree on this solution (that switching is best), but not everybody is convinced. That is exactly the point here. We are not as good at reasoning as we would like to think. We need to learn how to do better.
WILL YOUR WISHES COME TRUE?
Psychological studies also show us why we need to work on our skills. In some of these experiments, the question is whether an argument is valid in the sense that it is not possible for its premises to be true when its conclusion is false. The results reveal how many people assess an argument as valid because they want its conclusion to be true.2 Consider this argument: “If the referees are unfair, then Manchester United will lose. The referees will be fair. So Manchester United will win.” Many Manchester United fans will probably believe that this argument is valid. This belief is incorrect, however, because its premises are true but its conclusion is false if the referees will be fair but Manchester United will lose anyway. It is possible that Manchester will lose regardless of whether or not the referees are fair. The fans’ mistake results from their reluctance to imagine the possibility of their team losing, which they want to avoid. That is why fans of rivals of Manchester United make this mistake less often. They are happy to admit the possibility of Manchester United losing regardless of the referees. Of course, that does not mean that they are smarter or more logical than Manchester United fans, because they will make the same mistake about their own favorite team. Both sides engage in wishful thinking.
A related weakness is desirability bias, which is the tendency to seek information that supports positions you want to be true.3 Recall the last time you stepped on a scale to see how much you weighed. Studies show that if the scale shows a weight that you like, then you will be more likely to believe it; but if the scale shows a weight that you do not like, then you are more likely to step off and step back on the scale in the hope that it will show a better weight the second time. We all do something like this.
CAN YOU TRUST REPRESENTATIVES?
Our reasoning and arguments are also led astray by heuristics. Daniel Kahneman, a Princeton University professor who won the Nobel Prize in Economic Sciences, called one classic heuristic representativeness. Kahneman and his collaborators gave participants this description of a graduate student:
Tom W. is of high intelligence, although lacking in true creativity. He has a need for order and clarity and for neat and tidy systems in which every detail finds its appropriate place. His writing is rather dull and mechanical, occasionally enlivened by somewhat corny puns and by flashes of imagination of the sci-fi type. He has a strong drive for competence. He seems to have little feel and little sympathy for other people and does not enjoy interacting with others. Self-centered, he nonetheless has a deep moral sense.4
Participants were given a list of nine fields of graduate study. One group of participants were asked to rank those fields by the degree to which Tom “resembles a typical graduate student” in each field. Another group were asked to rank the fields by the likelihood that Tom is in each field. Both groups were also asked to estimate the percentage of graduate students in each of the nine fields. These estimates varied from 3% to 20%, and Tom’s description reflected the stereotype of the smaller fields, such as library science. Nonetheless, participants’ percentage estimates had almost no effect on their probability rankings. Instead, the answers to the questions about representativeness and probability were almost perfectly correlated. This suggests that these subjects neglected the baseline percentage and based their probability estimates almost totally on their judgments of representativeness. They ignored crucial information that should have altered their reasoning.
SHOULD YOU TURN OVER A NEW LEAF?
Another common error arises in the Wason selection task. Participants see four cards with a letter on one side and a number on the other side and one side facing up:
B L 2 9
Then participants are told a rule:
If a card has B on one side, then it has 2 on the other side.
The task is to turn over the minimum cards needed to determine whether the rule is true. The correct answer is to turn over the cards that show B and 9, because the rule is false if the B card does not have a 2 on its reverse or if the 9 card has a B on its reverse. Unfortunately, studies consistently find that most university students (as high as 90%) do not turn over the B and 9 cards. Most turn over either the B card alone or the B and 2 cards. However, there is no need to turn over the 2 card, because the rule will not be falsified whether or not there is a B on the other side. After all, the rule says only what is on the other side of cards that do have a B on one side. It does not say what is on the reverse of cards that do not have a B on one side.
Fortunately, this mistake becomes much less common when the task is transferred to a practical context. Suppose the cards look like this:
Beer Water 15 25
Then participants are told that each card has the customer’s age on one side and what that customer drank on the other side, and the law is:
If customers are less than 21 years old, they are not allowed to drink beer.
The task here is to turn over the minimum cards needed to determine which customers are breaking the law. Participants do much better on this more practical task. Some researchers explain this success by our evolutionary history. We evolved to determine when social rules (such as laws) are violated but not to test pointless generalizations (such as whether cards with B on one side have 2 on the other side).5
CAN WE GET BETTER?
These experiments (and many more) show that we are far from perfect reasoners. Duh! We already knew that. They also specify parti
cular ways in which many people often go astray. That is interesting, and it helps us know when we need to be careful.
The fact that we often get misled does not show that we cannot ever reason properly. Tricky psychologists set up special circumstances in order to get participants to make mistakes. Nonetheless, the Wason selection task shows that we can do better in certain circumstances (practical ones) than in others (abstract ones). Moreover, we can recognize when we made mistakes. After people give the wrong answer on the Wason selection task, it is easy to show them why their answer is wrong. They rarely stick to their original answer. That shows that we can learn and that we can distinguish good reasoning from bad reasoning in suitable circumstances.
Other psychologists have found that different situations are more conducive to proper reasoning. Despite their failures while alone, participants in groups gave around 80% correct answers in the Wason Selection Task; and more generally “Contrary to common bleak assessments of human reasoning abilities, people are quite capable of reasoning in an unbiased manner, at least when they are evaluating arguments rather than producing them, and when they are after the truth instead of trying to win a debate.”6 In addition, institutions (such as science) can be structured so as to maximize the chances that errors will be discovered and rejected, so that they will not be led astray in the long run.7 Thus, we can improve reasoning and argument not only by training but also by instilling a desire for truth and understanding as well as by creating institutions that correct mistakes. Those circumstances are more likely in a culture that understands reasons and arguments.
Our skills at reasoning and argument are both prone to error and correctable. The glass is not only half full or only half empty—it is both. It takes hard and careful work as well as patience and tenacity to get better at arguing and reasoning. Although difficult and not always successful,8 training and practice in argument and reasoning can help people recognize their mistakes, and they can also help people avoid mistakes in reasoning.9 That is why we all need to work hard at learning how to argue.
PART II
HOW TO ARGUE
6
HOW TO SPOT ARGUMENTS
WE SEEM TO ARGUE ALL THE TIME. People disagree on many issues and let each other know it, often at high volume. On the other hand, people too rarely give reasons for their positions. In that sense, arguments are not very common and not common enough. So, are arguments numerous or rare? That depends on what counts as an argument. This chapter will explore that question.
HOW MUCH WOULD YOU PAY FOR AN ARGUMENT?
To understand what arguments are, we need to begin by asking what arguments are not. Some of the main contrasts are illustrated by an insightful troupe of philosophers named Monty Python in their famous skit, The Argument Clinic. If you have not seen it or do not remember it, you should watch it.1 It is a gem.
The skit begins with a customer walking up to the receptionist in the clinic and saying, “I’d like to have an argument, please.” The Receptionist replies, “It’s one pound for a five-minute argument, but only eight pounds for a course of ten.” Despite the savings in bulk, the customer decides to purchase only one five-minute argument.
The Receptionist then needs to find an employee in the clinic to argue with the customer. She looks at the schedule and says, “Mr. DeBakey’s free, but he’s a little bit conciliatory.” What’s wrong with being conciliatory—that is, likely to give in easily? Anyway, the Receptionist instead directs the customer to Mr. Barnard in room 12.
The customer walks down the hall and enters the first room to find Mr. Barnard seated behind a desk. He aggressively yells, “WHAT DO YOU WANT?” then calls the customer a “snotty-faced heap of parrot droppings” and a “vacuous, coffee-nosed, malodorous, pervert.” Annoyed, the customer explains that he came for an argument. Mr. Barnard nicely replies, “Oh. I’m sorry. This is abuse . . . . You want room 12A, just along the corridor.”
This silliness introduces our first contrast with arguments. Abuse is not an argument. I cannot argue for my position or against your position simply by calling you a “pervert.” Why not? Presumably because calling you a pervert does not give any reason against your position, much less any reason for my own position. It is surprising how often people forget this simple point.2
Skipping ahead in the skit, the customer enters a different room, and Spreaders hits him on the head. When the customer reacts, he is told, “No, no, no. Hold your head like this, then go Waaah.” Then Spreaders hits him again. It turns out that this room is for “being-hit-on-the-head lessons.” This concept is absurd, but it reveals a second contrast with arguments. Arguments are not physical fights—or verbal fights. The goal of an argument is not to make an opponent’s head hurt (either by hitting him hard or by making him think hard).
When the customer finally reaches the correct room, a professional arguer named Mr. Vibrating is sitting behind a desk. The customer asks, “Is this the right room for an argument?” The clinician calmly replies, “I’ve told you once.” The heat rises from there: “No, you haven’t,” “Yes, I have,” “I’m telling you I did,” “You most certainly did not,” “Look, let’s get this thing clear; I quite definitely told you,” “No you did not.” The repetition is finally broken when the clinician asks, “Is this a five-minute argument or the full half-hour?” Then the customer realizes what is going on: He is already arguing. Or is he? The customer and clinician continue to say Yes-No-Yes-No-Yes-No until the customer bursts out, “Oh, look. This isn’t an argument . . . . It’s just contradiction . . . . An argument isn’t just contradiction.”
Now we have a third contrast with arguments. Contradiction here means denial, so the lesson is that arguments are not mere denials. If you make a claim, I cannot argue against your claim simply by saying, “No.” It is again unfortunate how many people forget this simple lesson. They think that they can refute someone merely by denying what they say. They can’t.
Why not? What is missing from a bare denial that is present in an argument? The customer tells us, “Argument is an intellectual process. Contradiction is just the automatic gainsaying of any statement the other person makes.” It is not clear what makes something intellectual, but one interpretation is that an argument needs to present some kind of evidence or reason, whereas a bare denial does not present any evidence or reason against the claim that is denied. To say merely that some claim is false is not to give any evidence against it or any reason why it is false.
This point then leads to the customer’s definition: “An argument is a connected series of statements intended to establish a proposition.” This reference to establishing a proposition is a great start, but it is still not quite right. The first problem is that to establish something is to put it on a firm basis. However, some arguments are not firm or even intended to be firm. For example, if we are deciding whether to go to a park or to a museum, I might say, “We went to the park last week, so maybe we ought to go to the museum today. What do you think?” I intend to give some reason for the proposition that we ought to go to the park, but I need not claim that it is strong enough to establish that conclusion. Some arguments are too weak to establish anything, but they still give some reasons.
Another problem is that you cannot establish what was already established in advance. To establish a country is to create one that did not exist before. Analogously, to establish a conclusion is presumably also to bring the audience to believe what they did not believe firmly before. However, we often argue for conclusions that everybody already strongly believed in advance. Just imagine that one mathematician had already proven the Pythagorean theorem (the square of the hypotenuse in a right triangle is equal to the sum of the squares of the other two sides). Then another mathematician comes up with a new proof that is shorter and makes fewer assumptions. Both proofs are arguments, but the purpose of proving the theorem the second time is not to convince people who did not believe the theorem. Everyone already believed it. Yet mathematicians still might
want to prove it in fewer steps with fewer assumptions in order to determine why it is true and which axioms or premises its truth depends on. Their proof aims to explain the theorem but not to establish it. In this respect, Monty Python’s definition is not quite right.
WHAT IS AN ARGUMENT?
One small change is enough to solve these problems with Monty Python’s definition. We just need to replace “establish” with “present a reason for.” Then an argument can be defined as “An argument is a connected series of statements intended to present a reason for a proposition.”3 Reasons do not need to be strong or firm and can support what we already believed, so this change allows weak reasons as well as proofs of the Pythagorean theorem to count as arguments.
The statements that present a reason are called premises. The proposition that they are supposed to be a reason for is called a conclusion. Hence, we can say that an argument is a connected series of premises intended to present a reason for a conclusion.4
This definition tells us a lot about arguments. It specifies the material that arguments are made of (language, though not necessarily writing or speech), what form they take (premises and conclusions, so declarative sentences that can be true or false), and what purposes they serve (to present reasons of some kind). This definition thus covers the aspects—material, form, purpose, and cause—that Aristotle required for complete explanation.5
Think Again: How to Reason and Argue Page 7