Chapter15Fallacies of Emotional Bias

15.1 Fallacy of Personal Attack (Ad Hominem)

Our first fallacy follows up on our discussion of bias and emotions. What do we do about the arguments of people we don’t like or have a bias against? Often, speakers go on the attack rather than carefully considering the arguments of others. In this fallacy, you can see how a bias or dislike of the features of a particular person can obscure one’s ability to think clearly about the arguments the person is making. And because of that, one responds inappropriately in the dialogue. This fallacy, also known as the ad hominem argument, which is Latin for “against the man,” indicates that the attack is directed against the speaker or arguer rather than their argument (fig. 15.1).

Personal attacks violate relevance conditions, since features of the speaker (the features that are under attack) are not relevant to the argument. Personal attacks are evidence that a view is not being carefully considered, and they are an expression of hostility against the speaker. To return to Douglas N. Walton’s fifth condition for a fallacy (see Chapter 13), the ad hominem poses serious challenges for the realization of the goal of the dialogue because it derails the dialogue.

In short ad hominem arguments are forms of abuse. Here we look at three versions in more detail; abuse, poisoning the well, and tu quoque (“Look who’s talking”). The illustration above (fig. 15.1) should demonstrate what the three forms of ad hominem arguments share in common. Arguer A is offering an argument, and arguer B is responding and offering additional argumentation. The green arrows between their arguments demonstrate that what they are each saying is relevant to each other’s points. They are thinking critically about claims in an effort toward better reasoning and truth.

Arguer C is using an ad hominem because they direct their statements toward arguer A’s person, not their argument. For abuse, name-calling hurts, and in addition, it doesn’t have a relevant, truth-seeking connection to arguer A’s argument. Tu quoque directs statements to past behaviour and essentially calls the person a hypocrite. And poisoning the well attacks the person’s motivations by bringing up their position or identity factors. We will deal with all three of these so you can tell them apart.

15.2 Abuse

To return to Walton’s second condition, “falls short of some standard of correctness,” here abuse falls short because it just isn’t an argument. Here we simply insult our opponent.

These examples demonstrate responses to arguments that may or may not be good. All we know is that the response is aimed at the person, not the claims at hand.

Example 1 is using a term that is not inherently abusive (one can even describe themselves as a Marxist), but the issue is the role of the claim. The arguer is viewing the identity of the person as inherently negatively affecting their argument. It is being used as an insult. Example 2 does something similar. It is not nice to be called a narcissist. And example 3 is pure mudslinging.

It is important to point out abuse where we see it, not just because it is bad arguing, but because it harms people and degrades the level of discourse we should be aiming for. In this case, one’s response to an argument or a claim is irrelevant because it shifts the conversation to the character or identity of the person, violating an important condition of rational dialogue.

In this case, while we can say this is a fallacy, it goes beyond a bad argument where we can offer ways to improve the argument. There’s no improving abuse—it needs to stop. It doesn’t even approach Walton’s fourth condition—it doesn’t have a semblance of correctness—or at least it shouldn’t! We can tell the person they are distracting from the issue at hand by mudslinging, but it might be too late in that instance to correct their reasoning. They have already demonstrated they will name call.

15.3 Poisoning the Well

Sometimes a form of irrelevance can be to put our opponent into a position where they cannot reply because their legitimacy has been undermined. Whereas abuse is name-calling and mudslinging about the person, poisoning the well specifically directs the ad hominem to the person’s motivations.

This is another way of saying that the person’s argument amounts to mere bias, so we don’t have to listen to them. “Lame stream media” is abusive, but it is also poisoning the well (fig. 15.2). It rules out having to listen to anything they say by saying their motivations for what they say are so overwhelmingly biased, anything they say should be disregarded.

Notice how both 1 and 2 absolutely rule out anything the person says. They’re considered tainted and anything that flows from them is false (water from a poisoned well). Examples 3 and 4 are similar because they confuse interest with bias. Military leadership has an interest in their budget, but saying their opinions are worthless because of that is to attribute to them an unbridled desire for money and war (which you need evidence for beyond their position in the military). Doctors also could be trying to make more money, but maybe not—we don’t know. But the argument in 4 states that by the very fact that doctors make money per appointment they cannot be trusted when they say you need to come back for another appointment. Poisoning the well directs the argument to the person (ad hominem) but specifically it attributes motivations that it then blows so out of proportion they are used to entirely dismiss what the person is saying. So just because someone has an interest doesn’t mean we dismiss everything they have to say.

15.4 Tu Quoque

In Latin, tu quoque approximately means “Look who’s talking.” It is a kind of ad hominem because it directs the dialogue toward the person’s actions rather than their argument. Like other ad hominems, it fails to achieve the rational goal of the dialogue (Walton’s condition 5).

Unlike abuse and poisoning the well, perhaps those who argue using tu quoque have a point? To revisit Walton’s condition 4, perhaps tu quoque has more than a semblance of correctness? The thrust of the tu quoque fallacy is that the speaker fails to follow their own advice. Shouldn’t their actions match their argument? Here is a good place to spend some time clarifying the goals of learning fallacies for critical thinking. We are identifying techniques for analyzing whether an argument is cogent by looking for and weeding out irrelevant information. So we ask, Is a person’s behaviour relevant to the cogency of their argument? When we were studying modus ponens, did it matter who uttered the argument? Another way of asking this is whether it is required to know their behaviour to evaluate the argument? In terms of the goals of argument analysis, no. We adopt a technique of divorcing the speaker from their claims in order to analyze the claims directly (Are the premises true? Is the argument valid? Are there unstated assumptions? etc.).

This is not appropriate in all contexts. There may be significant cultural, linguistic, or spiritual reasons to keep the speaker and their behaviour connected. Also, consider a court of law. In a court of law, we rely on the testimony of others as support for an argument. Consider eyewitness testimony. In this case, their credibility is very important—their credibility is support for the truth of their claims (we have to “take their word for it” that something is true). But in critical thinking, we are evaluating an argument that’s supported with reasons. Those reasons must be evaluated. One’s behaviour is beyond the scope of evaluating reasons for argument analysis.

There is something about these arguments (1–4) that might touch us. If you do think smoking is so bad, why do you smoke? But not withstanding this sympathy we may feel, it is invalid. Whether the speaker smokes or not has nothing to do with the quality of the arguments they might make that smoking is bad for you. It is pure distraction.

In examples 2 and 3, you see that someone has made prior claims, and the speaker’s response to those claims is to change the topic to their behaviour. Here you need to imagine what the original speaker would have been suggesting. In example 2, you could imagine the person said, “We should live in communes (conclusion) because we are social creatures and communal ownership encourages better land stewardship.” Would it be an adequate response to say, “If you think living in a commune is so great, why aren’t you living in one?” We can imagine any number of reasons why the person doesn’t or can’t live on a commune, but those are irrelevant, and the speaker need not defend their actions at this time. The question for argument analysis ought to be “Are our social nature and land stewardship adequate reasons to support the conclusion that we ought to live in communes?” And in example 4, eating a donut is not related to one’s credentials as a dietitian.

Putting all this together, let’s look at an example of an ad hominem from a news article. In a discussion of why some people are not getting vaccinated for COVID-19, a person is quoted as saying, “I mean, they’re mainstream, . . . They’re just going to say what the government wants them to say. I’m not an idiot” (John Burnett, “The Number of Americans Who Say They Won’t Get a COVID Shot Hasn’t Budged in a Year,”2 NPR, May 10, 2022).

Note how the person says, “I’m not an idiot.” This is essentially saying anyone who disagrees is an idiot, which is abuse. They are also poisoning the well by saying that anything the mainstream media says is just what the government wants them to say, thus undermining their very ability to speak on the issue. There’s also a slanted definition here. What does it mean to be “mainstream media”? This really sets up a dichotomy between mainstream and everything else. We always try to ask ourselves whether the people being described would agree with how they are being portrayed. Would the reporter say, “I’m part of the mainstream media,” or would they say something more complex, such as “I’m a journalist with journalistic integrity, and I have an employer who expects specific outputs at given times”? What is true is often more subtle and complex. One way we think of terms like “mainstream media” is that they are terms of the prosecution and not the defence. They are defined in an adversarial way.

15.5 Mob Appeal (Argumentum Ad Populum)

Previously we saw that ad hominem is a larger term that incorporates at least three forms (abuse, poisoning the well, and tu quoque). Mob appeal, often called argumentum ad populum (Latin for “argument of the people”), is also a broader category that incorporates ways of arguing aimed at our emotions, desires, and identities.

One way of thinking about mob appeal is to literally think of a mob: you have a group using scare tactics or other emotional appeals trying to get you to conform. It can be plainly stated as well that mob appeal tries to use the beliefs or feelings of the majority or a group to make a claim to truth.

For example, in mid-2022, one in six Americans are saying they will absolutely not get vaccinated for COVID-19 (Grace Sparks, Ashley Kirzinger, Liz Hamel, Melissa Stokes, Alex Montero, and Mollean Brodie, “KFF COVID-19 Vaccine Monitor: February 2022,”4 KFF, March 1, 2022). While that is not the majority, it could be said to be a popular (enough) opinion, especially since it is about forty-three million adults. Can forty-three million people be wrong?

Or, to put it another way, just because forty-three million people believe something, does that make it true? In this one story, a person against vaccinations says they only trust information from people who feel the same way they do (Burnett, “Number of Americans”).7 This is a form of mob appeal (and cherry-picking, and confirmation bias, etc.). Using a group-based feeling or identity as a sole reason for a belief is misguided.

Why is this a fallacy? Because the speaker is using something other than reason and evidence to try to convince someone that their argument is cogent. It violates a condition of critical thinking that arguments are evaluated for reasonableness, not for their mere ability to sway. Usually, examples of mob appeal are longer, but these shorter examples demonstrate the scope of the fallacy.

For example 1, imagine that there’s a city council budget meeting. A citizen is commenting on the proposed tax hikes. They say that taxes will be raised so much so that they will not be able to afford their monthly power bill. Summarizing the argument, you might say that the person is saying that the tax hike is too large because people who are just making ends meet won’t be able to pay for other important services. This is potentially convincing. Imagine that in response, the city councillor says, “I’m a blue-collar worker myself, and I know how hard it is to pay bills when costs are rising.” They have not addressed the issue of the cost of living; they have just tried to locate themselves in a group-based feeling (the group is “blue-collar workers,” and the feeling is the struggle to make ends meet). In a way, we don’t even need to consult with the question of not being able to afford important services, since the arguer hasn’t addressed it. Here, we would just need to say that the arguer is trying to put themselves in the same group as their opponent in an attempt to define and dismiss their struggle. Just because the person knows what it is like to struggle to make ends meet (assuming this is true!) doesn’t mean that the tax hike is a good idea. The two are not related.

Example 2 does something similar. It puts the emphasis on the audience, attempting to flatter them (fig. 15.4). It essentially says, “We are all smart, so we will agree.” This is fallacious because attempting to sway the audience with flattery is not a reason. Remember, it sets up the dialogue in a way that if you don’t agree, you are not a properly educated and serious group member. It shifts the focus to who you are rather than what we have good reason to believe. Whatever the speaker is trying to get the audience to agree with, they should do so on the basis of the cogency of the argument, not group membership. We haven’t been given reasons; we’ve been appealed to based on flattery and group interests.

Hopefully 3 will be a bit clearer by now. By saying that no one wants to deny a child a decent education, the speaker is creating a group of people imagined to agree on something, then leverages that group to prioritize “our school” over others. It also sets up a dichotomy between “those of us who care” and “anyone who would disagree”—isn’t it possible to care and to disagree? As a reader, we might feel backed into a corner with this example: We care about our school, and we don’t want to deny children a decent education. But we are being flattered, since we’re being told we have all the right values and care about the right things, and we are being told what follows from these shared values.

Mob appeal is essentially flattery of a group or an appeal to special interests. As a consequence, it is almost always in the service of greed and ignorance. If an arguer has good reasons for their position, why use mob appeal?

15.6 Appeal to Pity (Argumentum Ad Misericordiam)

The appeal to pity is really a special form of mob appeal. Fundamentally, it exploits a single emotion, sympathy (fig. 15.5).

It is important to note that the appeal to pity is inappropriate, but it is also unlikely to be effective in a dialogue. Different things make different people feel certain ways. For various reasons, some people have to shut down emotions, and some people need a very drastic situation to be swayed by pity. When you rely on pity alone, you add in a very unruly variable.

At the same time, just because we are improving our critical thinking doesn’t mean that emotions are irrelevant all the time. In fact, pity, like anger or any emotion, is appropriate in many circumstances. Emotions are important guides for how we can live well, but they need to be cultivated and developed. For example, anger is sometimes directed appropriately and sometimes not; it is also sometimes excessive or insufficient even if directed correctly. The issue with the appeal to pity is that the arguer is using pity as a reason for the audience to assent to the claim.

Let’s consider two examples:

In example 1, the question is whether Mr. Cratchit merits a raise, not whether he has the need for more money. We can certainly establish that he needs more money (just ask Tiny Tim). But this is not by itself a reason for a raise.

Usually, a raise is determined by merit, job description, responsibility, and so on. Now you might reply and say that there are unjust social conditions that Mrs. Cratchit is suffering under and that the suffering is a good reason for the employer to raise the standard of living for all the people working for Mr. Scrooge. In that case, collective reasons for changes in the standard of living go well beyond Tiny Tim’s need for an operation—and we’ve just given reasons for our claim rather than relying purely on pity.

Example 2 is a textbook appeal to pity. One’s future education and career are highly charged with feelings of all sorts. The idea that a grade can make a difference in one’s life is certainly worthy of sympathy. In this case, if grades mean anything at all, they have to be connected to the work on one’s assignment, not on the sympathy of the professor. Essentially, one’s potential misery is not relevant to the assessment of their work.

15.7 Appeal to Force or Fear (Argumentum Ad Baculum)

This fallacy, known as the argumentum ad baculum (literally “the argument of the stick”) uses blatant or subtle threats to force the audience to agree with their claims. The idea is that if you don’t agree, bad things will happen (fig. 15.6).

Perceptions of force and the use of fear can vary widely by context; the important point is that the person hearing the argument recognizes him or herself to be threatened. Much like the appeal to pity, the appeal to force or fear is unruly because different people will feel threatened in different amounts or not at all by the same threat. Threats are also not rationally connected to the cogency of an argument. They also shut down any possibility of considering other views.

To return to Walton’s five parts of a fallacy, the appeal to force or fear has a semblance of correctness (part 4). Isn’t it rational to avoid harm to oneself? For the most part, yes! But remember that we are deciphering standards of correctness in reasoning and critical thinking. Is a threat of injury, reputational damage, hardship, or otherwise itself rational grounds for accepting a conclusion? No, and the threats themselves should not be made in the first place.

The threat need not be of force on the part of the speaker. An attorney may commit the fallacy of the appeal to force by telling the jury, “If you don’t convict this murderer, you may be his next victim.” Here the fear of being the murderer’s next victim is being used as a reason to send him to jail, but presumably, the reason for him to go to jail in this instance has to do with crimes already committed and the evidence to that effect.

Example 1 is not only condescending but threatens the person’s livelihood. Example 2 is similar. It is using threats to stop teachers from unionizing. And in example 3, the arguer is not giving good reasons for their son’s actions not being news. They remind the audience of their advertising dollars to covertly threaten to rescind them.

The fallacy of appeal to force or fear has a particularly ugly ring to it. This is largely due to the fact that the other fallacies we have discussed only work if you don’t notice that they are fallacious, but this one only works if you notice its particular fallacious character. The appeal to force or fear might bring up the idea of police officers or mobsters, but people in any group can make threats that don’t have to do with physical harm at all. They could be for reputational damage, withholding goods that one deserves, or exposing personal information.

It is worth mentioning, however, that there are certain classes of disputes that are not so much disputes about the facts as disputes about who gets what piece of the pie. Here you might have two or more groups that both want the same thing, and no amount of arguing will help them find a solution. This is why many disputes escalate to appeals to force. Where there are substantially opposed interests and prima facie good claims on either side, a rational resolution of the dispute may be impossible to work out. There are many television shows that almost entirely demonstrate the use of threats to make claims to power and goods (Game of Thrones, Damages, House of Cards, and Succession, to name only a few). This of course does nothing to make such cases legitimate arguments. Threats of force are never okay.

15.8 Two Wrongs Make a Right

Another way that people will use emotions and personal attack is to suggest what a person would do as a reason to undercut their position. Certainly you’ve heard the phrase “two wrongs don’t make a right,” and yet, people argue this way often. Here, the arguer isn’t addressing the claim under consideration, but rather, they are pointing the finger at an imagined version of the other person. Consider the example of cheating in school. Given the opportunity to do so, one might justify this to themselves, saying “Other people are doing it or would do it.” Does this make it right to cheat?

This fallacy uses a kind of false agreement. It says that you’d do the same, but even if that were true, is that a good reason for belief? The idea here is that the reasons for belief are independent of whether the other person would also act the way you have acted. The other person’s behaviour isn’t relevant. So two wrongs make a right, like most of the fallacies of emotional bias, violates an important relevance condition by not addressing the issue at hand.

In example 1, the fact that other countries also don’t have guaranteed access is not going to move the conversation forward. The arguer is not addressing the issue, which is the benefits or drawbacks of abortion access. Pointing to what is done elsewhere distracts and undercuts the dialogue.

Example 2 doesn’t address the rightness or wrongness of releasing tax returns; it changes the conversation to what the other person would do. This example appears to border on tu quoque except the person hasn’t actually done anything; it is just using the hypothetical fact that a person would do a wrong thing to mean that it is okay for them to do a wrong thing.

Example 3 basically says oil spills are OK because there are other oil spills.