Chapter16Fallacies of Expertise

16.1 Genuine Appeal to Authority

An appeal to authority is often not only legitimate but completely necessary. Without our ability to trust the testimony of others on authority, human society in any form would be impossible. In complex societies where much knowledge is socially distributed in positions requiring study and specialized knowledge or skill, our dependence on authority takes on an edge: We need to be able to rely on the specialized knowledge of others—people for the most part who we don’t even know. But people are not always knowledgeable, reliable, or honest, and authority is not always legitimate. We need therefore to be able to distinguish genuine authority from its mere appearance.

If we are to appeal to an authority, that appeal must genuinely support reasonable belief. As a result, a genuine appeal to authority must meet several conditions:

Criterion 1 might seem silly to have to point out—that if a person claims expertise, they actually have it. But there are many cases of people claiming they have degrees that they don’t. In fact, CBC’s Marketplace was able to purchase 3 PhDs2 online without doing any work at all. But this problem has predated the internet. This is why academic integrity is so important—we need standards to verify that our credentials mean something. So if a person is claiming expertise, it can be helpful to verify that expertise. We (Eric and Kristin) assure you that our PhDs in philosophy are real!

According to criterion 2, if a person actually has expertise, they need to stay within the realm of that specific expertise. The further people go in their education, their expertise usually becomes narrower and more specialized. Thus it is important to make sure not just that a person has expertise but that they are speaking within their expertise. Dr. Oz is a cardiac surgeon, but he regularly dispenses medical advice about all areas of health and medicine. For example, Dr. Oz has claimed that apple juice has unsafe levels of arsenic3 in it. While you do want your surgeon to actually have a medical degree, specializing in surgery, you wouldn’t want that same doctor to dispense public health advice about infectious diseases. They are different areas of expertise.

Criterion 3 tells us that we need to investigate authorities for taint. For the most part, the only way we can meet our basic needs is to work for money. This is true of most everyone, including most experts who have authority on specific areas of knowledge. Universities pay researchers and professors to create new knowledge and teach, but does the money they receive mean we can’t trust what they say? In order for people to have the time and ability to develop expertise, they need money to meet their basic needs while studying, so the mere presence of money doesn’t undermine credibility. But if an expert (who may be recognized and speaking in their area) is paid specifically to say that something is a fact, then that is considered taint. Being paid to specifically assert a claim undermines authority. For example, Dr. Oz has been paid to promote specific therapies.4 This alone undermines credibility.

Criterion 4 might require some time to discuss. Not all domains of expertise are created equally. Firstly, the subject matter must be one in which expertise actually produces agreement in judgement because there is an independent matter of fact to investigate and acquire knowledge about. This can be a difficult matter to determine. It is difficult to figure out, for example, whether an oracle is an expert, and this is so precisely because there is no agreement on whether there is an independent way to determine whether the oracle gets things right. Most areas of expertise have wide swaths of agreement on matters of fact and some dispute or disagreement on less established areas of inquiry.

Following a model of scientific inquiry, if areas of knowledge are tested and retested, then claims become more established. But there are areas of inquiry where being an expert does not necessarily give the expert’s claims authority. Agreement in the field might not be possible in principle because of the kinds of claims the field makes. The Long Island Medium, Theresa Caputo, claims to be an expert on talking to the dead. So does John Edward, who claims to be a psychic medium. These two “experts” might disagree on what a dead person’s message is to their loved ones (imagine if we set up an experiment where they are to talk to the same person separately), but does their disagreement really further the inquiry toward a fact of the matter? No, because their disagreement is not the biggest problem. The problem is that the domain that they claim expertise in cannot in principle give authority to their claims.

Psychic or medium work of “talking to the dead” uses age-old techniques of deception, such as cold reading and shotgunning. Same goes for astrology, religion, palmists, extrasensory perception, creationism, faith healers, alternative medicine, homeopathy, and the list goes on. The point here is not to trash people who believe in these phenomena but to remind us that when we use a claim from an authority in our reasoning, it needs to be dialectically acceptable. It needs to withstand rational scrutiny. These claims cannot be used in arguments without meeting various criteria.

Criterion 5 is really the last step in validating an authority. If all other criteria are in place, then what do we do if there is disagreement? We have to rely on consensus among the right experts but realize that some claims will not (initially) generate consensus. The claim that “some claims will not generate consensus” needs to be clarified. Some claims have not yet generated consensus—more investigation needs to occur. But if consensus among experts is in principle not possible, then we are back at criterion 4. Imagine, however, a case where there is consensus among thousands of experts, but there is one outlier who doesn’t agree. Does that undermine consensus? A couple of points here: First, political punditry often exaggerates the level of disagreement when an issue is “hot button.” So it is always important to find out exactly who is disagreeing in this case. If it is one person who is paid (criterion 3) to produce the appearance of a disagreement, then the disagreement is not genuine. But if there are genuine disagreements among reputable experts, then we have to say that when using claims. Criterion 5 rules out basing claims only on an outlier’s view when genuine authorities have reached a consensus.

16.2 Fallacious Appeal to Authority

The fallacy of appeal to authority is committed when at least one of the necessary conditions of genuine authority is not met. You will notice that this means the fallacy of appeal to authority is a kind of a negative definition.

Here are nine criteria of the fallacious appeal to authority:

Example 1 should be clear that the speaker’s medical opinion is not credible because he is merely an actor who plays a doctor on TV rather than an actual doctor. Jennifer Love Hewitt might be a car enthusiast and perhaps quite knowledgeable, but she is not a recognized car expert. And example 3 should alert us to at least two things: Serena Williams is not a recognized nutritional expert, and she is a paid sponsor, so we know that she is being paid to promote Cheerios.

16.3 Fallacy of Snob Appeal

Another way we might inappropriately appeal to an authority is using snob appeal. Social authority is conferred not just through expertise and degrees but also often through one’s perceived prestige or membership in exclusive spaces or places. Snob appeal can be very insidious because it uses social pressure to try to motivate belief.

Snob appeal is one of the worst kinds of appeals to authority because it uses as an authority something completely inappropriate. Social superiority has no bearing on the truth of one’s claim. This needs to be pointed out so much more in everyday contexts where people who have social clout can define the truth of situations.

The fallacy of snob appeal is trying to play on vanity and special interest to motivate belief. It is no surprise that these techniques are often used by politicians and advertisers to motivate belief in claims. Donald Trump has said many things that amount to saying, “I’m rich, so believe what I’m saying.” You would think we don’t need to point this out, but his view is that he has been successful (we can question his business record), and that means he is an expert on the economy, a good way of life, the political structures we should adopt, and so on.

16.4 Appeal to Tradition

Tradition is often appealed to in everyday contexts, keeping social norms and customs in place. Sometimes this is fine—for example, we have always put brown sugar in our spaghetti sauce, therefore we should do it that way. Nothing really hangs on this recipe in terms of truth. It is more about a way of life that one wants to emulate and continue. But sometimes tradition is used as a form of adjudicating truth about the world or truth about right and wrong, both of which require justification and argument. Another way to think of this is, Why should we infer that something is right just because it has always been done that way? Tradition is fine in certain domains, but it isn’t by itself (ipso facto) a reason to believe or to do something.

When we appeal to tradition, we use only the fact that something is a tradition as a reason to support a conclusion. This starts to look like cultural relativism, which we discussed in connection with biases. Consider these two different examples that both appeal to tradition:

1. 1. We’ve always had an abortion ban. Therefore, abortion should be banned (moral).
2. 2. We always put brown sugar in the spaghetti sauce, therefore it should continue (non-moral).

Example 2 gives us a reason to put the brown sugar in, but only if we want the sauce to taste the same. It isn’t a moral argument like example 1, which is about abortion. Example 2 depends on what you already desire—how you want your sauce to taste. You might want it to taste traditional. But, when it comes to the morality of abortion, this is beyond individual preferences and needs a larger argument than the way things have always been done.

All three of these examples share the same structure, the idea being that the very fact that something is a tradition (voting democratic, dropping pledges in mud, having zoos) at one time is used by itself as a reason to support a claim. Tradition is fine. It isn’t by itself a reason to believe or to do something.

16.5 Appeal to Nature

The appeal to nature is perhaps a more narrow version of the appeal to tradition, but it deserves a moment of consideration. What is nature? We don’t ask that with our tongues in our cheeks. When we think of nature, we might think of spaces untouched by human interference, but aren’t humans natural? So then, is nature just all that there is in the world? Is nature good or bad? Marketers use “all natural” to market nature to us, but what does it mean for something to come from a “natural source”? These questions are underneath the appeal to nature, but so is the question of whether nature is good or bad. Usually nature is used to mean something is good—you don’t want to go against nature (or do we?). We discussed how “nature” was used in a personified way with hypostatization. Here, nature is presumed to be an authority to justify claims.

All three examples highlight how “nature” is selectively used to justify something that needs reason. Nature cannot be appealed to without additional rational support. Remember that important concepts in our arguments need to be defined. This fallacy does not rule out that nature can be an authority, but this is just the start of a dialogue. It cannot be appealed to without independent reasons and an explanation of one’s interpretation of specific natural phenomena.

16.6 Appeal to Anonymous Authority

Sometimes expertise is appealed to and the subject of that expertise is unnamed or doesn’t exist. It may be some kind of amalgam of public opinion, but really it is made up. To say that “some people are saying” a certain claim is fallacious. There is no authority to that claim whatsoever. Some people could be saying anything at any time. Why is it relevant? Why is it appropriate? What is the justification for this appeal? Appeals to authority appeal to the testimony of an expert according to specific conditions. Appeals to an anonymous authority appeal to an unnamed or unnameable authority using words such as “some people are saying.”

Another way to put this is to say it’s an appeal to rumour. We shouldn’t accept rumour in everyday contexts, and we certainly shouldn’t accept it in arguments. It is not dialectically acceptable. The anonymous authority doesn’t necessarily say X is true; it says X is a valid claim worthy of consideration. Anyone can make up things.

Examples 1 and 2 both name the anonymous authority. “They” or “some people” are the subjects of the sentences, presuming there actually is a “they” or a “some people.” Example 3 is even more shifty. It fails to offer a subject and just says that there is demanding going on but doesn’t even tell us what or whom is doing the demanding. This is why when we appeal to an authority, we must actually cite the authority. We have to point to an actual existing expert making the relevant claim.

16.7 The Appeal to Ignorance

This named fallacy might seem strange. How could ignorance ever make an argument anyway? Why would anyone argue this way? The appeal to ignorance has quite a bit of appeal in everyday rhetoric. Surely you’ve heard an exchange such as the one in figure 16.2.

Who has a better argument here? We have to think about the overall claim at issue. The dispute is about whether God exists. The arguers are having a dispute about proving God exists. What counts as proof? Can it count as proof that it can’t be disproven? This is the appeal to ignorance. The burden of proof in an argument must always be on the person making the positive claim. Arguer B is the one holding the claim “God exists” because they are saying that A’s inability to disprove that God doesn’t exist is proof God exists. What’s to stop anyone from assuming whatever they want and just demanding the rest of us disprove it?

The appeal to ignorance is often undetected, and pointing out its regular use can be very effective. Have you ever heard someone say you can’t use a negative to prove a positive? It is likely they were pointing out an appeal to ignorance.

Let’s take another tack. This really has to do with what is considered adequate evidence. Let’s leave aside the question of what would count as adequate evidence of God. Let’s just think about evidence for a cure for AIDS. Consider the following claims:

Is the fact that AIDS does not currently have a cure by itself reason to believe it has no cure whatsoever? No, if something doesn’t have a cure, this does not give us reason to believe a claim about it having or not having a cure. You certainly cannot demand that someone give up their belief that AIDS could be curable by the very fact that there is yet no cure.

The fallacy apparently originates with John Locke, who saw the argument in a weaker light as an attempt to establish or shift the burden of proof (see Douglas N. Walton, “The Appeal to Ignorance, or Argumentum Ad Ignorantiam,” Argumentation 13 [1999]: 367–77). The speaker asserts a proposition that the listener must accept as proposed or offer an argument against. This suggests that there can be different grades of the fallacy.

Let’s look at each of the above claims more closely. Claim 1 is the simplest version:

1. 1. A cure for AIDS hasn’t been found therefore AIDS has no cure.

Claim 2 takes us into different territory. Technically then, the appeal to ignorance fallacy is an instance of irrelevant thesis (as discussed in Chapter 17), as the inability of a person to prove something is not relevant to the correctness of the thesis.

1. 2. You cannot prove that AIDS has a cure, therefore AIDS has no cure.

Claims 3 and 4 we can put together, since they take the failure to disprove a claim as evidence that it is possibly true.

1. 3. You claim that AIDS has a cure, but you cannot prove that AIDS has a cure, therefore you must give up your claim that AIDS has a cure.
2. 4. You claim that AIDS has a cure, and I claim that it doesn’t, but you cannot prove that AIDS has a cure, therefore you must give up your claim that AIDS has a cure and acknowledge that I have a right to believe that it doesn’t.

What do 1–4 have in common? They are all part of a strategy to circumvent the demand that one must possess evidence for the hypothesis they are putting forward. Because it comes in different strengths, in the fallacy of appeal to ignorance, one takes the failure to disprove a claim as evidence that:

1. 1. it is possibly true,
2. 2. it is reasonable to think that it might be true
3. 3. one has a right to believe it without further ado
4. 4. one ought to take it as proof of the conclusion’s correctness.

We might argue that you have no right to criticize our belief if you cannot show that it is wrong; after all, you think it is wrong and we think it is right. So what should we do in the absence of evidence either way? Conclude that we are both equally authorized to believe what we do? No. Concluding that we can believe our favoured claim in the absence of evidence is still epistemically irresponsible. It abandons our serious commitment to seeking the truth, and it is a fallacy. To see this, look at version 4 of the fallacy again. The claim here is that you can’t prove your point, so you have to give up your point and acknowledge that we have a right to ours. But of course, since there is no proof either way, you can offer the very same argument back to us. Version 4 is unstable because we can both offer it, and yet the conclusion is that we have a right (the right to believe) that you lack. The correct end point would be to acknowledge that neither party has sufficient evidence for belief and that therefore belief is unjustified for both parties.

The argument from ignorance has a special place in the arsenal of conspiracy theorists and bigots. Consider racist disputes or disputes about the Holocaust—where the racist or the Holocaust denier is really quite immune to counting anything as proof against their view—or cases where conclusive evidence is not ever likely to be available, as in the case of disagreement between a theist and an agnostic. The fallacy is best revealed by a consideration of more mundane examples that highlight the comparative unfoundedness of conspiracy theories.

To begin with, consider the question of whether there are exactly 271 Russian wolfhounds in Winnipeg. Wolfhounds are rather uncommon dogs; it is unlikely that you could find out how many there are, but Winnipeg is a pretty big city, and there are probably a fair number of wolfhounds in the city. So it is reasonable to believe that there are some Russian wolfhounds in Winnipeg but not too many; so how about 271? The evidence is so circumstantial and vague that we don’t know what to say. There could be 271, but in the absence of any adequate evidence, it is much more likely that it is false than that it is true. Why is this? If it is true, there must be exactly 271 wolfhounds in Winnipeg (so not 272 or 270 or 356 and so on). The point is that there are a very large number of ways it can be false and only one that it can be true. So my failure to prove that there are not 271 is no reason at all to believe that it is true.

Or consider this simpler example. If you were to toss a penny twenty times in a row, you might get heads twenty times. It could happen, so there cannot be proof that it won’t. Still, you would be very wrongheaded to believe it. Since we multiply the odds of the individual tosses (1/2), the odds against it are 1/220, which is less than 1 chance in a million (it is 1/1,048,576)! Many things that could be true (and for which there would be a perfectly acceptable explanation if it were true) are nevertheless hugely unlikely and thus not belief-worthy at all. So let us bring this lesson to consider conspiracy theories.

Consider conspiracy theories such as claims that the earth is flat, the Apollo moon landing was a hoax, 9/11 was secretly planned by President Bush, or the Holocaust never happened. There are such conspiracy theories, and they are believed by a good number of people, but all of them require a large number of unproved and individually extremely unlikely assertions to be true simultaneously. For example, if the Apollo mission was a hoax, a very large number of people had to be in on the hoax, the video would have been shot somewhere and kept secret then and thereafter, and so on. It is not reasonable to believe, given the many opportunities there are for such hoaxes to be made public, that the conspiracy would be effectively managed by secret agencies all these years. Typically, conspiracy theorists attempt to isolate their views from the reach of counter-argument by concocting elaborate secret conspiracies for which there is no evidence but that are such that if those conspiracies did exist, their views would be “reasonable” since, of course, you can’t disprove their secret conspiracies. Arguments from ignorance constitute only one of a variety of strategies that “true” believers use to put their beliefs beyond the reach of criticism. You may remember the movie The Matrix. If the hypothesis of that movie were actually true, then there would be no way to test or evaluate any belief, but of course, the hypothesis of the movie is preposterous; there is no reason to believe it to be true at all, and it is just one of a very, very large number of equally implausible hypotheses, all the others of which would have to be false if that one were true.

The important thing about the fallacy of appeal to ignorance is that using the opponent’s inability to disprove a conclusion as proof of the conclusion’s correctness is a transparently irresponsible piece of reasoning. In taking a belief to be true, one needs sufficient evidence to make the claim belief-worthy; it is not enough just to have an argument against the claim of one’s opponent.

Here are some examples of the fallacy in its strong and more easily identifiable form:

Example 1 is compelling—it has that semblance of correctness. But it goes too far. “No one has been able to show that there isn’t” is only a reason for us to say we don’t know either way. It is not evidence that there is life on other planets. This takes the negative evidence to prove a positive.

Example 2 is similar. The inability to prove is not a reason to conclude against; it is only a reason to keep investigating or be sceptical either way.

Versions of example 3 are pervasive. We’ve even heard philosophers say, “I haven’t heard a good argument for X, therefore it shouldn’t be considered.” This is a problem because the truth is not dependent on what a person has or has not been exposed to. There should be significant humility in each of us to know that what we have and haven’t heard can’t determine what is possible. There is a lot out there that we just don’t know and never will know. This means we keep an open mind, not that we can prove a negative.

Example 4 is not only an appeal to ignorance, but it is question-begging epithets, a fallacy discussed in Chapter 19. The use of “alarmist” and “reputable scientist” render us virtually unable to respond to the argument without being labelled as alarmist and disreputable. The fact that something hasn’t been proved is not reason to believe there positively is no danger, just that we don’t know.

There are some specific and well-defined situations where an appeal to ignorance is legitimate. In a Canadian court of law, a person is innocent if not proven guilty. Also, lawyers often use the term “negative evidence” to refer to the idea that there isn’t supporting evidence for a claim put forth by the prosecution. For example, if the prosecution is claiming that someone was shot at a particular location at a particular time and there’s a lack of appropriate blood splatter, that is considered negative evidence. But note that the claim being put forth needs justification (that someone was shot at a particular location and at a particular time). Here the claim lacks the appropriate physical evidence, thus the claim becomes less likely.