When presented with an argument, we often rush to judgment, evaluating it before we fully appreciate it. One important lesson to take from the preceding sections is that it's best to be sure that you understand the argument before commenting on it. This can be done through careful identification and reconstruction, the units you have just worked through. In this unit, we will discuss how to evaluate argument once we have reconstructed them.
One evaluates arguments by assessing their quality, i.e., how good they are as arguments. They might be eloquent as speeches or spine tingling as theater, but that won't make them good arguments. An argument's purpose is to compel a listener to believe the conclusion on the basis of the reasons given in support. To be a good argument, it must supply agreeable reasons that make the conclusion seem clearly true. Thus, a good argument guides reason, whether or not it appeals to emotion.
In this final section, we consider techniques for determining the quality of an argument. A strong argument will have three things going for it:
A thorough evaluation should subject an argument to scrutiny along each of these dimensions. In what follows, we present each in turn, commenting on strategies that one can use while evaluating arguments. At the end of this unit, you should be able to provide systematic, three-part evaluations of arguments that determine whether the conclusion should be believed.
Consider two buildings built from the same blueprint. When they are under construction, the crews erect two structures consisting of a foundation and criss-crossing I-beams. They look identical in form, even though they obviously contain different specific pieces. The buildings go up, and then the owners decide to fill them with different contents, e.g., different carpets, elevators, windows, etc. Here we have two buildings with the same structure but different contents.
When it comes to structure and content, arguments are analogous to buildings. First, they have structures in addition to contents. The standard form representation of an argument will reveal much about its structure, laying out the sentences that compose it. These sentences have grammatical structures, in addition to being about things (i.e., in addition to having content); further, they are interrelated, sharing terms and phrases that make their truth values depend on each other. Second, these structures, or forms, are repeatable---we can find other arguments that are about different things that have the same structures.
Consider argument (I):
While not the most interesting argument in the world, it is useful for our purposes because its structure is easy to observe. One good way of revealing the structure of an argument is to replace the sentences found in the reasons and conclusion with symbols---this takes your mind off of what the argument is about and forces you to attend to the structures of the sentences and the relationships among them. If we do that here, we get the following form (1):
We replaced "It's hot in Moscow" with "A" and "It's unbearable in Lewiston" with "B". Now that we can see the form of the argument, though, we can use it to build other arguments with the same form. All we do is replace "A" and "B" in our symbolic form with other sentences. For more, see Expanded Notes on Argument Forms.
While the form of arguments isn't everything, it is an essential element nevertheless. As with buildings, if the structure of an argument is weak, the argument will be weak. If the structure is strong, then the argument could be a good one, depending on the other essential element, viz., content. If an argument's form is good, then the reasons will support the truth of the conclusion; otherwise, they will not. Given this, it is important to have a clear sense of argument form, as it is one aspect that you will want to evaluate. For our purposes, it will do to talk about two broad categories of argument forms, the certainty producing and the probability producing. Arguments within these categories conform to different standards, and we will describe these standards briefly. More detailed and systematic study of argument form can be found in courses in formal logic.
Certainty Producing Arguments (CPA). Also known as deductive arguments, these are intended to produce their conclusions with certainty. The rhetoric of the argument context can indicate the presence of CPAs, as can reliance on the mere form of sentences and the roles played by certain terms, e.g., "if ... then", "all", "or", etc. (See Expanded Notes on CPAs for more on this.) In addition, CPAs are often associated with the analysis of concepts, as opposed to the investigation of the world of our experience. Note, though, that these are mere rules of thumb for identifying CPAs.
To say that the conclusions are produced with certainty is to say that if the reasons are true in a CPA, then the conclusion must be true. A CPA that meets this standard is known as valid. Note, though, that the standard of validity is merely hypothetical---it says that IF the reasons are true, the conclusion me be true; thus, if the reasons are false, then all bets are off. This is just more evidence that form is only part of the story---you also need content, and it will be the content that determines truth and falsity of particular reasons, as we will see in the next section. In valid CPAs, we say that the conclusion follows from the reasons.
Good CPAs, then, are valid arguments in which the conclusion follows from the reasons. Bad CPAs do not guarantee their conclusions, and you can detect this through the use of counterexamples. A counterexample is a specific example, rooted in what is possible, that makes the reasons in a CPA true and the conclusion false. What this demonstrates is that the truth of the reasons does not force the conclusion to be true, and so the CPA in question fails to meet the standard of validity and is invalid. A very good technique for determining the quality of a CPA involves attempting to identify a counterexample; if you can, the argument is formally weak, but if you can't (and you've really looked), then the argument may well be quite good. Still, it would help to know a bit more about the details of CPAs, and in particular, what makes a good CPA good, aside from the fact that it doesn't make room for counterexample. For these details, see Expanded Notes on CPAs.
Probability Producing Arguments (PPA). Also known as non-deductive arguments, these are intended to increase the probability that their conclusions are true. That is, if the reasons are true in a PPA, then there will be a high probability that the conclusion is true, if the argument is a good one. Signs that indicate the presence of a PPA include (a) the conclusion speaks of things that go beyond the reasons, (b) the conclusion is more tentative and is not presented as certain, given the reasons, and (c) the argument presents a conclusion supported by scientific reasons. As before, these should be regarded as rules of thumb.
If the the truth of the reasons in a PPA raise to a high probability the truth of the conclusion, we will call the argument strong. Note that the standard of "high probability" is vague---there is no clear way of specifying this in general. What counts as "strong" will vary, then, from argument to argument and context to context. (So if you evaluate an argument as strong, it would be a good idea to have an argument ready to back up your evaluation!) To reflect the difference between CPAs and PPAs, we do not say that the conclusion of a PPA follows from its reasons; rather, we say that the reasons support the conclusion.
PPAs come in a greater variety than CPAs, so it is more difficult to provide a general characterization of what makes one good or bad. But there are standards that apply to these arguments, and it is important to familiarize yourself with them. There are several types of PPAs, and in the Expanded Notes on PPAs, we detail these types, describing them and commenting on the standards that apply to each. However, we can make a couple of general observations at this point. First, a PPA will typically be stronger if it is more detailed; the more vague a PPA is, the less compelling it will be. Second, a good PPA will typically have a conclusion that does not depend on a single, specific type of observation or data point, even if there are a lot of them; the broader the types of observations that support the conclusion, the stronger it generally is. (Think of a table---if it only has one leg, it won't be as stable as if it has several legs in different places.) Third, the stronger the conclusion of a PPA, the weaker the argument. If the conclusion is weak, it will not take as much to support it, and other things being equal, this will mean a stronger argument overall.
Form is very important, but it is hardly the only aspect of an argument that can be evaluated. In fact, form is not typically the obvious thing to evaluate---it is hidden beneath the surface and often difficult to ascertain. The more obvious aspect is that of content, i.e., what the argument is about. After completing Exercise One, continue on to the next section, where we discuss how to evaluate the content that fills in an argument's skeletal form.
Consider argument (II):
Evaluated in terms of its form, this is a good argument---it is a valid CPA of form (1) detailed above. But it is nevertheless a bad argument. After all, each of the sentences is false. Sentences (2) and (3) are clearly false, but so too is (1), since if Idaho were larger than California, it could still be smaller than Alaska, given that Alaska is bigger than California. What this demonstrates is that when it comes to argument quality, form isn't everything. Content matters, and here the content is in bad shape.
The content of an argument is what the argument is about, and this is based on what the sentences that constitute the argument are about. The primary measure of content quality is truth value. If any of the sentences are false, then the argument will be weak. If the sentences are true, then the argument will typically be strong, so long as it has good form. Consider argument (III):
All of the sentences in (III) are true, but it is still a bad argument. The truth values of (1) and (2) don't force (3) to be true because (1) and (2) have nothing to do with (3). Thus, the reasons and conclusion should have content that is related. This is often guaranteed by the form of the argument---in the case of (III), the argument has bad form, and this undermines the connection between the reasons and conclusion. (However, in certain unusual cases, an argument can have good form (logically speaking) and have true sentences and still be a bad argument, as is explained in the Expanded Notes on Content Evaluation.)
Thus, the evaluation of content is a two-part job. First, you need to determine the truth values of the reasons. Assuming that the form is in order, assessment of the reasons as true will generally imply that the argument is a good one. (Although don't judge before evaluating the context, as we will do in the next section.) There is little that can be said here about how you do this, as it depends on the discipline within which you're working. You might be able to tell by looking whether the reasons or true, or it might be a lot tougher than this. [INSTRUCTOR: You might wish to introduce at this point some specific detail about how one inspects substantive claims in your discipline for truth or falsity.] Second, you need to determine the overall thematic coherence of the argument. If the conclusion addresses a topic not mentioned in the reasons, or it goes beyond the reasons in a way that seems like a stretch, then that is a sign that the argument content might not be as coherent as it should be for the argument to be strong.
After finishing Exercise Two, proceed to the final section of this unit, Evaluating the Context.
If you have moved an argument through the preceding two evaluation stages and it seems good, then there is one more stage before it receives your seal of approval. Keep in mind that arguments are not produced in vacuums---they are generally produced for a particular purpose in the context of action. Thus, it is possible to have an argument that is impeccable on its own merits but is nevertheless bad because it doesn't fit the context in which it was introduced. Perhaps no argument was required, or perhaps an argument was the very thing that the other participants in the discussion didn't want to hear. It is important to remember that arguments are given in contexts, and if these don't fit together, then you need to evaluate the argument negatively.
We can go into this in a bit more detail. The first layer of context that we might wish to evaluate is what we can call the intentional context, i.e., does the argument suit the needs of the arguer? Given the goals she has, does the argument serve them? If not, then the argument is a bad one in that case. Second, given that it fits with the intentional context, does it work in the practical context where it was produced? The arguer wanted to present an argument, and this one worked for that purpose, but was an argument really required? Did it do more harm than good? One common kind of practical context is a discourse context, i.e., the discussion, conversation, debate, etc. to which this argument was a contribution. In practical contexts, there will be content constraints, since the argument might not be about the right thing, given the overall nature of the activity in question. (Imagine someone who produced a great argument about the quality of a local lending institution to a bunch of fisherman talking about their favorite spot to fish at the river.) For more about context evaluation, see Expanded Notes on Context Evaluation.
In the following exercise, you will have the chance to test your context evaluation skills on a few argument/context pairs. When you have completed this exercise, you have finished the Critical Thinking Worksite. The only thing left to do is the final project, which is described in the next section.
V. Putting It All Together
Review the arguments in Grendel, Chapter 5, one last time and proceed to the Final Exercise. In the preceding sections, you should have identified and reconstructed an argument from that chapter. It is now time to bring your critical thinking about it to a close.
Let no one tell you that you aren't a critical thinker. When confronted by an argument on the street, we tend to think critically as a matter of course. The skills we have focused on at this worksite are skills you have used before, many times. We often move through the stages discussed very quickly, identifying the fact that it is an argument from the tone or the rhetoric, spotting the conclusion while we are getting a sense for the reasons, and then rushing to evaluate, all in one quick motion. Laying these skills out in full view as we have done here brings out their detailed and subtle character. The work you have done in studying them may have little impact on the critical thinking you do among friends or when little is on the line; however, knowing the details can come in quite handy when you are in a tight spot, or when the decision you make is a very weighty one. When you need to slow things down and proceed carefully, it helps to know how to do this. It is good to think when all else fails (or even before all else fails!), but it is better to think well.