Did you see the Trump Axios interview? I mean did you SEE it? The academic term for that is “bonkerballs.” Honestly, I have never seen such a wild display of a combination of incompetence and complete denial of reality in such a short, compact amount of time. Jonathan Swan deserves an award for not allowing his eyes to pop out of his head in shock at what he was hearing. But I don’t know. Maybe he’s used to it by now.
There were a lot of crazy moments. The John Lewis exchange was certainly shocking. The conversation about how he has done more for the Black community than Lyndon Baines Johnson was certainly illuminating. But the segment that captured my attention the most was the absolutely wild back-and-forth between Swan and Trump about COVID stats when Trump was trying to convince Swan things weren’t that bad with a bunch of very simple, color-coded graphs.
First, can we talk about the nature of these graphs?
Trump was dealing with some extremely complicated information. The stats he was dealing with were complex and required in-depth analysis and interpretation. But the graphs he provided were as simple as graphs get. That bar graph he showed Swan looked like what you use to explain the concept of “bar graphs” to second graders when you are teaching them what they are. There is no way these graphs could have contained the information concerning COVID statistics. That is complex data and those were childish representations of simplified numbers.
That’s a basic fallacy. Oversimplification is at the heart of any number of common fallacies – straw man, false dilemma, hasty generalization, and others. Oversimplifying the argument renders it useless.
It also says something about what people tend to assume about evidence. If you put it on a chart, it has to be true, right? If you can quantify if, it must be convincing? But this is the perfect example of how that is absolutely not the case. Just because you can convert something to numbers doesn’t mean it is good evidence. Just because you can put something on a chart doesn’t mean it supports your argument. You have to know what good evidence is, how and why it supports your argument, and when it is appropriate to use it. “Evidence” is not a foregone conclusion. It has to be vetted and used appropriately.
So today we’re going to be talking about what goes into a good argument – we’ll be borrowing from Aristotle, Stephen Toulmin, and Edward S. Inch and Kristen Tudor.
First let me talk about Aristotle. Aristotle provided us with the most basic form of logic for argumentation there is that I have mentioned before but haven’t really defined – the syllogism.
The syllogism is basically a set of three statements – a major premise a minor premise, and a conclusion. Think of them as the Venn diagram of arguments. The major and minor premises are like categories – those are your circles. And where they overlap is the conclusion. So you make your first two statements and what you can logically get from that is the conclusion. It is the simplest form of logic there is.
It is so simple, in fact, that you don’t even need words to do it. You can just use variables to represent the parts of a syllogism and follow the logic. The content almost doesn’t matter – it’s a form and function kind of thing.
Now, just because this is foundational and simple, doesn’t mean it is easy. A LOT of people really struggle with this kind of thinking
Think of it this way – if the ONLY two things you knew IN THE ENTIRE WORLD were the major premise and the minor premise, then the conclusion is what you could also know. That means you can’t make an inference; you can’t bring new information to bear on the situation. This is not a train of thought – this is simply seeing how two categories combine. And many people struggle with the formulaic, structured nature of that.
Let me give you some examples:
The most famous example is Socrates was a man. All men are mortal. Socrates was mortal.
See how simple that is. You have to categorical statements: Socrates fits into this category: “man.” All men fit into the category of “mortal.” So Socrates fits into the category of “mortal.”
It also matters whether your observations and categories are accurate. If I said, “Socrates was a field mouse. All field mice have four legs. Socrates had four legs.” I would have made a structurally correct syllogism. But it still wouldn’t have been useful because the information in it is untrue. The whole thing is unsound. A syllogism can be structurally valid but still unsound because the information in it is untrue.
Some other forms of syllogism are Either A, or B. Not A. Therefore B.
If A then B. A did not happen. Therefor B will not happen.
We are seeing a lot of this in public discourse right now.
Masks help stop the spread of disease. You should do everything you can to stop the spread of disease. You should wear a mask.
It’s simple, it’s clear, and it is just a combination of categories. A does this thing. You should make A happen. You should do A. It’s basically A causes B. B is desirable. You should A.
We’re also seeing the bad end of this, too. For example, it is my constitutional right not to wear a mask. I’m an American citizen. I’m not going to wear a mask.
There are a lot of problems with this argument. First, the first category is not true. In 1905 the Supreme Court held that vaccines could be compulsory, so it doesn’t seem like masks are going to be a real deal-breaker, here.
Secondly, there is a lot of assumption in this syllogism, and syllogisms don’t operate on assumption. Right now this is just a string of loosely connected statements, not a syllogism. I am assuming you mean that as an American the constitution gives you certain rights, but that’s not being stated.
What it SHOULD be is “As an American the constitution grants me certain rights. One of those is the right not to wear a mask. I’m not going to wear a mask.” Then it would only be untrue, not just invalid.
We’re also getting some variations on “Everything the media says is a hoax. I’m hearing about COVID from the media. COVID is a hoax.” We can see how dangerous this is automatically. When the first premise is untrue (and that’s why people warn you about categoricals so often), then the entire syllogism becomes weaponized. This is particularly pernicious because it SOUNDS logical. Structurally, the syllogism is valid. But it’s not true.
And this has real-world consequences. Because the argument SOUNDS logical it spreads. But it is spreading a lie. The irony is that the argument is a lie that is claiming others are lying. In such a convoluted circumstance logic seems to be of little importance.
So what is the point of evidence? What is the point of reasoning?
Next, I want to talk to you about a guy named Stephen Toulmin. Toulmin came along a few decades ago and said, “I think there is more to arguments than just syllogisms.” When we make arguments, we don’t do it in these kinds of stilted, three term plodding things that just go back and forth in these logical tit-for-tats.
Toulmin said there’s actually a bunch of parts to most arguments – at least three, and up to six. He described three essential parts to any argument, the claim, the data, and the warrant, and then other, optional but helpful elements, the backing, the qualifier, and the rebuttal.
The claim is what you think it would be – it’s what you are trying to convince somebody of. Claim literally means the claim I am making. The thing I want you to believe at the end. Data is the evidence I use to support the claim. The thing about data is it comes in a variety of forms. People think “data” and they think numerical evidence, but evidence can be any number of things – beliefs, commonly held opinions, ideals, numerical data, narrative, expert opinion, statistics – it is really as much as matter of how you use it as what is IS. Data us used to support the evidence. And data can be good, or data can be bad. The warrant is essential, but kind of hard to understand. The warrant is the reasoning that connects the data to the claim. Just because you have a whole bunch of data doesn’t mean it necessarily supports your claim. WHY does it support your claim? The warrant articulates that. The warrant can be spoken or assumed.
ALL arguments have to have claim, data, and warrant to be an argument. It’s not an argument if it doesn’t have those three things. For example, if I say, “Dogs make great pets; they are easily trained,” that may make perfect sense. But does it?
In reality those are two completely unconnected statements. I have the claim: Dogs make great pets. And I have the data: They are easily trained. But they have to be connected by a warrant, which you, as a listener, have filled in without thinking about it because it is assumed. But in order for that argument to make sense those ideas have to be connected by the idea that “and easily trained animal makes a great pet.” Otherwise they are just two tangential statements.
The other parts of an argument are helpful, but not essential. The backing is a set of reasoning or data that supports the warrant. So if your warrant needs support you would provide information or reasoning to strengthen your argument there (could you provide evidence for the idea that an easily trained animal makes a good pet?) The qualifier is just a word of phrase that does exactly what it says – it qualifies the claim. It is the “probably” or “most likely” or “almost always” or “90% of the time” part of a claim. It just indicates whether the claim is categorical or if there are times when the claim might not be the best possible solution. The rebuttal is the acknowledgement of those situations in which the claim is not the best solution. It is the “except when.” It just acknowledged that there may be circumstances during when your initial claim may not make the most sense, so you acknowledge that in your argument.
For an example of how this works you can look at some of the arguments made at www.thehealthyamerican.com. For example, they argue that businesses cannot force you to wear masks because they cannot make policies that are discriminatory.
The claim here is that businesses can’t make mask policies because mask policies are discriminatory to those who don’t want to wear masks. But this argument falls apart because it has not valid warrant. The warrant would have to describe how mask policies are discriminatory, and the group cannot adequately do that. There is no group, no protected class, that is being discriminated against by being asked to wear a mask. This argument lacks a sound and cohesive warrant. Any backing for that warrant could easily be blown out of the water just by defining the words “discriminatory.” If they tried to claim people with health problems can’t wear masks that’s fine – that would be the rebuttal of a claim that stores can mandate masks. It can’t stand as a claim on its own. The argument lacks structural integrity.
And if those with health problems are used as the rebuttal for a claim that stores can mandate masks then it is specifically NOT discriminatory because it is allowing for their special case – and those who do not have that special case have to follow store policy just as they would a “no shirt, no shoes, no service” policy.
Just as it is not discriminatory to say, “You can’t wear clothes with profanity” on them in my store, it is not discriminatory to say, “you have to wear a mask in my store.” Because discrimination is more than doing something somebody doesn’t like – it is the unjust or prejudicial treatment of different categories of people or things, especially on the grounds of race, age, or sex.
So this argument becomes nonsensical. Claim: Stores can’t make mask policies. Data: Mask policies are discriminatory. Warrant: …they don’t have one. Because the common understanding of discrimination does not connect the data to the claim. This argument does not hold. They would have to completely re-define discrimination, which would be an argument in and of itself.
Finally I’d like to talk to you about evidence. The organization and structure of arguments is difficult and can make understanding what a good argument is (or isn’t) really challenging. But sometimes it’s not that complicated. Sometimes it comes down to the quality of the evidence presented. Evidence for any argument should be vetted. And there’s no black and white, yes or no, answers for what is good evidence. It is a balancing act, but it’s also a relatively clear one if you approach it honestly. So let’s talk about how you vet evidence in an argument.
First, evidence comes in a variety of forms. Some of the most common are statistics, narrative, and expert testimony. The tendency of many people is to automatically trust statistics the most. But that can be tricky. Because the truth is that ALL of these have pitfalls and can be manipulated.
Narrative we know can be used in treacherous ways. But it can also be the most effective. With narrative you have to ask yourself how is this appealing to me? Why? Does this jive with OTHER evidence I have been presented with? Is this JUST appealing to my emotions or does it have value outside of that? Narrative can be useful – it can provide context, detail, and the human part of evidence that other types of evidence leave out. It can provide nuance and depth that is required for understanding all of the evidence. But it can also be manipulative.
Here’s something important to understand – statistics are NOT objective. They can be manipulated just as easily as narrative but slap some numbers on there and suddenly everybody trusts them. Go back to the example we were talking about in the very beginning – those charts and graphs Trump showed Swan in the Axios interview. Those were obviously all nonsense. But he got those numbers from SOMEWHERE. They probably weren’t just made up. You have to know the context of your numbers. You have to understand the questions that were asked to get your numbers. You have to know WHO was asking the questions and who was ASKED to get these numbers, if it is a survey or a poll. These all make a difference when it comes to statistics and will color your numbers.
Expert testimony may be the most trustworthy of the three, but even then, you have to ask yourself some questions. Is this person an expert in the actual field they are being asked about? Economists aren’t any help when talking about epidemiology. Does this expert seem to agree with other experts in the field? Occasionally you have somebody who has great vision and can see something other people can’t and is right when everyone else is wrong, but honestly that is rare. That’s really the stuff of movies. If the expert is an outlier it is more than likely that expert is wrong – or is maybe not the expert they are claiming to be. Consistency matters. So if the entire scientific and medical community is telling you to take something seriously and behave in certain ways to avoid tragedy, and two or three doctors on YouTube are saying the opposite – ignore the scientific and medical community at your own peril.
Second, there are ways you test these different forms of evidence.
Look for a source’s reliability. Does the source of your evidence have a track-record of being trustworthy and accurate? In the age of COVID that means you look for whether the source of your information has consistently provided sound argument and scientifically accurate information. If your source doesn’t have a track record of good arguments and good evidence, look around – there are a lot that do.
Second, look for expertise. Does your source have extensive background knowledge? Has your source been vetted? Is your source’s background relevant and relative? There are a lot of people out there right now who are telling you to do your own research, sheeple. But if your research consists of some YouTube videos and a random Twitter feed then you aren’t doing yourself any favors. Because THAT’S not research. Instead, depend on the people who have actually DONE the research and listen to what they have to say. There are people out there whose job it is to go through either historical databases or the scientific method to come up with relevant data for your benefit. They are called experts. You spending a few hours down a YouTube rabbit hole and arguing with somebody on Twitter does not match their carefully honed, structured, reviewed, and vetted work. Science, history, sociology, technology, and the like are not democracies. Not everybody’s opinion is equally valuable. The people who have done the work have more say than a blowhard on Facebook.
Look for consistency. External consistency is a tough one – does the evidence you are using generally agree with other credible sources? This may seem confusing because part of the problem is conflicting data. This is why there is a litany of tests. If a thing is consistent with plenty of other sources, but all of those lack recency, objectivity, and reliability, it’s probably no good. If it is brand new information that isn’t consistent with anything but is from a relevant, objective, reliable source that has lots of experience and has very rarely been wrong – it is worth considering. So if you are reading an article and you’re like, hey, I like what this has to say – ask yourself if it is in line with what other reliable sources have to say? You have to be honest about whether you are looking for reliable evidence or evidence that confirms your biases.
Recency! Yes, recency counts. It counts MORE depending on the issue, but more recent information is, as a rule, better. Remain current. Historical information is useful, but believe it or not, recent historical research is better than historical research from 100 years ago, because WE KNOW MORE STUFF NOW. In science, if your information is more than a few years old, it’s pretty useless. If you are using research from the humanities, you have a bit more flexibility, but newer is better. In philosophy and ethics, you can go as far back as you want, but know that you can be challenged with anything new because “old and established” doesn’t mean “better and accepted.” This is really important to understand in the age of COVID because the information is constantly changing. That doesn’t mean the experts are wrong or they don’t know what they are talking about. It means in the light of new evidence they have re-considered their options and come up with a better idea. That’s the way science works. We’re just not used to seeing science in real-time and turns out it is pretty jarring for some people.
Always consider relevance. Always focus on the question/issue at hand. Don’t wander from the topic. No red herrings, no straw men. Keep your evidence actually relevant – this is where syllogisms and Toulmin (probably syllogisms more so) can be really helpful. They can keep your arguments efficient and structured. Does what your talking about actually support your argument? If the discussion is about the safety of opening schools or not and you bring up the constitutionality of mask mandates in a Wal-Mart, you are absolutely not helping.
The thing about making good arguments is that it is hard. Most people think they have a good grasp of it because they hear trash on TV and online, but that’s just the thing – they hear trash. For an argument to be good it needs to be observably true, structurally sound, and have good, vetted evidence. That is actually much harder to come by than one might think.
Some of this is because we don’t really teach argumentation in schools. We teach that you should have reasons for your opinions, but that’s about where it stops. We don’t teach logic, we don’t teach much about vetting sources or research, and we don’t teach the process of argumentation. I see the results of this when my students get to my upper-level rhetoric classes and they are presented with this kind of information and it is mind-blowing. They really struggle with it. And it’s not a political thing. It’s the process of argumentation. It’s just not something we value that much anymore.
But that’s a problem, because we are absolutely influenced by public discourse. We engage in public speech and argumentation all the time, and that is becoming MORE the norm, not less, as more and more people find their voice on the internet. In a world where SO MANY people have the opportunity to be influential, if even to a small group of people, we are massively ignoring what it means to influence and the ethics and process of that. The result is not JUST that there are a lot of bad arguments out there, but that people are finding them convincing.
The coronavirus crisis has shown us just how dangerous this is. Arguments are powerful, whether they are good or bad. Perhaps we need a more concerted effort to teach argumentation at all levels of education.
But there I go again with my liberal indoctrination. I guess I just can’t help myself.
Music in this episode is “Fearless First” by Kevin MacLeod at https://incompetech.filmmusic.io/song/3742-fearless-first.