Please visit
Philosophical Practice
— the rudiments of my dissertation.

Arguments: Part One:
What Is An Argument?

In the new age of blogging and pseudo-journalism, it is important to emphasize basic critical thinking skills. This is part one of what will be series on the concept of argument. The idea is to present, in a simple way, some tips and tricks that can be used to identify, for one thing, if an argument is happening, and more subtly, is an argument is good or bad.

Most blogs are really quite embarrassing. This is understandable enough as the "articles" are one-off, self-published and unreviewed (— with the notable exception of blogcritics.org, which is a peer-review forum). It's quite a bit worse to witness the decay of reasoning skills in what television, newspapers and magazines pander as "journalism" these days. It seems to me that fewer and fewer writers have a fundamental grasp of the ingredients to a good argument, and that damages politics, policy and democracy.

First off, an argument is not one statement. For example, it is misleading (or just plain wrong) to say something like "some believe the argument that education needs more funding". "That education needs more funding" is not an argument — it might be a conclusion to an argument; it might be a premise in an argument. But on its own, it is not an argument. More dangerous is that, when calling a conclusion an argument, the asserter suggests that there is only one path to that conclusion. In the above case, there are many arguments whose conclusion is that "education needs more funding". To call the conclusion or a premise "the argument" vandalizes that fact.

That's the first big point: arguments are more than single statements. More precisely: arguments have structure. This is important because our evaluation of an argument depends on our being able to identify its structure.

The standard example of a syllogism is as follows:

All men are mortal.
Socrates is a man.
Therefore Socrates is mortal.

Let's not dwell on the dull that it is — the point is that an argument has parts. The classic syllogism has two premises and a conclusion. The premises and conclusion, taken together, make up the argument.

It turns out that the argument above is "valid". Arguments can have many other structurally valid forms, such as "modus ponens" or "modus tolens". "Valid", however, doesn't mean that an argument is true. Consider this:

All happy people are made of cheese.
Tanya is a happy person.
Therefore, Tanya is made of cheese.

This is a valid argument. Why? Because, and tattoo this into your brain: validity is structural; it has a valid structure. The further step is to determine whether the argument is "sound".

An argument is "sound" when all of the premises are true, or colloquially, make some sense. The first example about Socrates: it has the structure of a valid argument, and all of the premises are true, so we can call it "sound". The second example: it has the structure of a valid argument, but not all of the premises are true, so we cannot call it "sound". That's where journalism often fails: often, journalists do not investigate how reasonable the premises are, and so only recognize validity (at best).

That's one notion of argument. You might have guessed that there are lots of way to think of what counts as an argument. In the discussion above, you have to know what counts as a valid structure, and you have to be able to identify which statements play which roles in that structure. That's pretty subtle stuff.

Another way to think about arguments is "Argument to the best explanation".

The rough idea is that you're presented with some data that stands in need of explanation — let's assume that you're able to evaluate what that data is. For example, you're standing in front of a burning building, you smell gasoline, and you see someone running away with a gas can in his hand. Your task is to explain this data. Your explanation will count as an argument, because your explanation plays the role of a "conclusion" to the "premises" that are the data. That's a helpful way to start thinking of the structure of an argument to the best explanation.

To help you get to the explanation, it's easy to formulate a question that will guide the possible answers. In the example, you might ask "How did the building catch on fire?". Or you might ask "Why is the man with the gas can running?". Or you might ask "Why do I smell gasoline?" Which question you pick out from the pack depends on what you want to know. For example, if you're with the fire department, you probably want to know how the building caught on fire. If you're the guy with gasoline's buddy, you might want to know why he's running (from you). The question that interests you will help you sort out the data that stands in need of explanation.

Importantly, the question that you choose should not be a "yes/no" question, generally speaking. This is because the "best explanation" would then be limited to two possibilities. You want lots of possibilities at your disposal so that you can evaluate which explanation is best.

In the burning building example, let's say that we decide that the best question to pursue is "how did the building catch on fire?" Well, we've got a smoldering building, the smell of gasoline and a guy running. Certainly the building and the smell stand in need of explanation, and the most obvious explanation is that the fire started with gasoline. But we might want to be more specific. The guy running suggests that maybe "the fire was started deliberately with gasoline." But it could just as easily be that "the fire started accidentally with gasoline (and the guy just happened to be running at the time)." Or it could be that there was an electrical fire and there happened to be gasoline in the building.

See, it's complicated. But thinking like this opens up lots of possibilities. What counts as an argument, here, is the whole ball of wax: the data and the explanation.

Interestingly, more data might be discovered. That data might also stand in need of explanation. Or, that data might make another piece of data no longer need to be explained. For example, what if we learned that some film students were shooting a movie called "Man Running With Gasoline" nearby. Then both the smell and the guy running could be explained away effortlessly. In turn, this makes our preferred explanation look a little less likely.

All of this underwrites the importance of investigation. And also important: charity. We should always try to be charitable when we approach arguments in this way.

These are two approaches to structuring good arguments. It's important also to be able to identify when things aren't good arguments.

In much popular, mainstream media these days, there are two favorite bad arguments: attacking the person, and false dilemma.

The case of attacking the person is fairly easy to spot. For example: "George can't be right: after all, he's a Republican." That sort of attack has absolutely nothing to do with whatever argument that George has advanced.

The false dilemma can be a bit tougher, for it almost always puts one on the defensive. For example: "you're either against abortion or you're a baby killer." Obviously, there are plenty of pro-choice folks who don't actually kill babies. Indeed, it's entirely possible and reasonable to be pro-choice but to oppose actually having an abortion. One way to attack a false dilemma is with a "counter-example". The counter example would show how, roughly speaking, it's not "either/or".

These two structures are "fallacies". There are others (such as the "straw man" or "begging the question"), but if you arm yourself with an ability to identify these, you'll be well on your way to rooting out bad arguments.

Next time: more about how to structure "arguments to the best explanation."

© 2005 Sorrell
August