Chapter Five: (Re)constructing Arguments


I.    Introduction

Consider these cases:

  1. You sit down with your copy of Newsweek, a cup of coffee, and kick back for some light reading. Within the pages of the magazine, you find a George Will column. This guy drives you crazy, but your eyes are drawn into the first paragraph and then pulled down like a blind--you always lose this battle of the wills. Somewhere near the third word, you start to grit your teeth. By the third paragraph, you're excoriating him in an audible voice. By the end, you are left shaking your head. "How can he believe that?" you ask yourself. "Smart guy, yes, but man ..." You know you have just wrestled with an argument, and it is one about which you have an opinion. However, it may not be obvious just where in all that prose the argument was. The conclusion might be plain--the fact that they are often featured in the headline doesn't hurt. But what are the reasons? Surely not every thing that is said is a reason--but which sentences deliver reasons and which just create the mood?

  2. Your teenager wants her own car. She points out that she's 15 and she recently started driving, so the time has come for her to have wheels of her own. You disagree. Since your primary reasons are that she is 15 and recently started driving, you know that you will not sway her to your point of view by tacking them on to an "Absolutely not!" But you nevertheless wish her to share your point of view. (As it happens, your point of view has been lonely since she was 12, but such is a joy of parenthood.) It's pretty clear that you need to soften her up, perhaps by making concessions, or by calling her attention to problems that are attendant upon car ownership. But how to do this in a way that isn't so obvious that it undermines your efforts? Whatever tack you take, you must take it soon if you are to have any chance. You have a conclusion and you need an argument, but you don't know what it should look like or how you should set it up.

Argument identification is the crucial first step in critical thinking--you must perceive that beliefs are at stake before you begin deliberating between the options. This is true whether you're on the reception end, as in the first case, or on the production end, as in the second. But the crucial second step is argument (re)construction. You know either that you are confronted by an argument or that you need one, but what is its precise character? Without this knowledge, evaluation of the conclusion in light of its reasons will be pointless.

In what follows, I will develop the craft of argument (re)construction in some detail. I begin with a word or two-hundred about the attitude required to do this job well. I then present two distinctions. The first, between the relevant and the irrelevant, applies to information presented in a conversation or interaction containing an argument. The second, between the explicit and the implicit, applies to steps of the argument that you recover from the conversation or interaction. After describing the standard form in which arguments are represented for the purpose of evaluation, I supply a step-by-step process for constructing or reconstructing arguments.

II. The Right Attitude

Whether constructing or reconstructing an argument, it is important to keep your audience in mind, at least if you are after meaningful dialogue. The business of argumentation is often predicated on disagreement, and disagreement has a way of making dialogue empty and worthless. One way to avoid this is to be sure that you are treating the arguments you reject in as equitable a fashion as possible, consistent with your own beliefs and goals. If you are crafting an argument that you plan to endorse, you should be mindful of the positions of your audience as revealed in previous arguments; further, your argument should engage the strongest versions of those arguments that you can recover. If you are reconstructing an argument, it is important to aim for a reconstruction that captures the intent of the arguer. In the event that the audience has not made themselves perfectly explicit throughout, as is very likely, this will require giving them the benefit of the doubt from time to time.

In other words, the attitude to adopt in preparing an argument intended for consumption by others is one of charity. There are two important reasons for being charitable at this stage of the critical thinking game. First, if you are charitable to a fellow arguer, it is more likely that you will cast their argument in a form that meets with their approval, thereby enabling the discourse to move forward. If a substantive and sustained exchange is the goal, then a charitable approach will likely prove rewarding. Second, if you intend to criticize an argument advanced by another, it is much better to criticize it in its strongest form, as this will put your position in the best possible light. The alternative tack will put one in danger of committing the strawman fallacy--after all, no one is impressed when you knock over a strawman. (Detail about the fallacies is found in Chapter Seven.)

III. The Relevant and the Irrelevant

Once you have recognized that you are in the presence of an argument, it is crucial that you discriminate the relevant from the irrelevant. Much will likely be said in a conversation or a text, but not all of it will be connected to the persuasive effort. Some of what is said may be small talk designed to establish the social setting, some may be rhetorical flourishes designed to enhance the old rather than deliver the new, and some may be tangents intended to inform or amuse. While these contributions have their purpose and place, they do not typically contribute to the construction of an argument, and so do not advance the cause of argument reconstruction. Once again, argument evaluation—the point of critical thinking—requires an argument to evaluate, and the more precisely this argument is presented, the more cogent the evaluation will likely be. Thus, it is crucial that you identify which utterances are relevant to the task of constructing the argument.

This is not typically an easy task. It is easier if you are constructing the argument, rather than reconstructing it, but even then it can get confusing. It is not uncommon for someone to get caught up in the moment and add steps to an argument that they themselves do not appreciate. Also, arguers can be seduced into thinking that everything they say is relevant—how could you dare suggest that I’ve said something irrelevant? Still, in general, the person who is responsible for introducing the argument is the best one to consult for the purpose of discriminating what is truly relevant from what is not.

If you are a listener or reader reconstructing an argument, the task is more difficult. If you are engaged in a conversation, then you may be able to enlist the aid of the arguer in doing this. If the argument is located in a text, the job is more complicated. Among other things, you may have difficulty determining the number of arguments present in the text. If there is more than one argument in a particular argumentative discourse, then a claim that may be irrelevant to the specific argument under consideration may be relevant to a different argument advanced in the text. Thus, claims are often not relevant or irrelevant simpliciter, but rather relative to specific arguments contained in a discourse. Further, you may find it necessary to revisit the text during the process of argument reconstruction and reassess claims. As the argument takes shape, it may turn out that certain claims have an important role even though they appeared initially to be irrelevant.

While there is no precise set of conditions that can be used to separate the relevant from the irrelevant, there are useful rules of thumb that can be used. These correspond to the following questions, asked of specific claims:

  1. Was the claim under consideration intended by the arguer to be a part of the argument?

  2. Would it be uncharitable not to fold the claim into the argument that you are reconstructing?

  3. Is the claim related to other parts of the argument?

  4. Is the claim on the same topic as other parts of the argument?

These rules are not all on the same footing. Affirmative answers to (1) or (2) will typically justify claims as relevant, while negative answers will justify evaluation as irrelevant. Affirmative answers to (3) or (4) supply good evidence that claims are relevant, while negative answers supply good evidence for the opposite conclusion; however, content connections of this sort do not prove relevance or irrelevance and are overridden by answers to (1) and (2).

IV. Explicit and Implicit Steps

When you make decisions about relevance, you work with explicit claims, that is, claims that are introduced into the discourse with the voice or the pen. Once initial (and provisional) judgements of relevance have been made, you will have divided up the claims explicitly made in the discourse into those that will figure into the argument, as you now understand it, and those that will not. However, in general, not all the claims that form a part of the argument will be expressed—some will be left implicit in the discourse. If you’re reconstructing an argument and you’re lucky, the argument you’re after will be announced as such, with ample rhetorical and structural cues serving up a long list of relevant claims. However, in many cases, you will find that the list of relevant claims is quite brief, too brief in fact to yield a charitably complete picture of the argument. The difficulty of argument reconstruction is compounded in cases of the latter sort by the fact that arguers do not always explicitly indicate every step of their arguments; instead, they often ask their audiences to "read between the lines." In such arguments crucial presumptions are left unsaid, either because they are deemed too obvious or because they are underappreciated by the arguer. (This is not to say that the arguer does not recognize them as a part of the argument; rather, it may only be that they fail to appreciate their importance and so fail to express them explicitly.) Distinguishing the explicit claims into a relevant set and an irrelevant set will get you part of the way toward a charitable reconstruction, but ultimate success depends on filling out the set of relevant claims with those that are left implicit by the arguer.

When we argue, we often take logical leaps, either because we don’t notice them or because we don’t think of them as leaps. Thus, it can be the case that we are not sensitive to some of the assumptions that underpin our conclusions. Critical thinking, however, places a high value on taking small steps, and in light of this, critical assessment of an argument may depend on identifying unexpressed assumptions. If you are constructing an argument, it is prudent to take things slow and recognize the steps necessary to convey you from premises to conclusion. If you are reconstructing an argument, then as before, things become trickier. If you are arguing with someone in real time, you can inquire into what they are leaving implicit; otherwise, you must decide on your own whether to credit them with the claim.

Once again, as before, there are rules of thumb that can aid in the identification of implicit argument steps. However, because you are proceeding without the aid of the arguer, identifying implicit argument steps is more of an art than a science. You must be careful to treat all implicit argument steps as even more provisional than the explicit steps deemed relevant. With that in mind, the following rules of thumb can be of use:

  1. Credit the arguer with the step if it is necessary to make the argument go through. (Does the arguer need the step to have an argument at all? If so, it is consistent with the Principle of Charity to credit them with it.)

  2. Credit the arguer with the step if its absence would vitiate the argument. (This differs from (1) in that here there would be an argument without the step, but it would be weaker. The Principle of Charity is also behind this rule.)

Charitable attribution of a step to an argument can be overridden under certain circumstances. Foremost among these is the situation in which the arguer explicitly denies the claim in question. Additionally, one should be leery about attributing a step if it is inconsistent with things said elsewhere by the arguer—here you might choose to be charitable at the level of the overall view rather than at the level of the specific argument. Finally, if the tone of the arguer suggests strongly that they would not endorse the claim, then one should try to reconstruct the argument without it.

V. Standard Form

Once you have identified the relevant explicit claims and have identified claims left implicit, it is useful to organize them in a way that reflects their contributions to the argument as a whole. The first step in this direction involves regimenting the claims in what is known as standard form. To begin, select the claim that is the conclusion and write it below a horizontal line. Above the line, list the rest of the claims, all of which will be regarded as either premises or intermediate conclusions. It is helpful to try and arrange these in order, from those that represent the arguers starting points to those that are asserted immediately prior to derivation of the conclusion. Number each of these consecutively, and then number the conclusion last. You will end up with something like this:

1. Premise/Intermediate conclusion

2. Premise/Intermediate conclusion

3. Premise/Intermediate conclusion


n. Premise/Intermediate conclusion


n+1. Conclusion

This is highly stylized, but it is effective in focusing attention on the argumentative character of the claims involved and the relationships between them.

VI. Putting It All Together

In many ways, constructing or reconstructing an argument is the trickiest and most difficult stage of critical thinking. Argument identification is aided considerably by conventional contexts and by key words, while argument analysis turns on the use of systematic structures and the application of precise standards. Argument (re)construction, by contrast, often involves flying blind, pulling arguments out of undifferentiated tracts of text and, when that isn’t enough, the thin blue. Nevertheless, we can systematize the practice somewhat. In this section, I supply a procedure that one can follow in doing (re)construction work. (‘Text’ refers here to verbal or written collections of claims.)

Reconstruction Procedure

  1. Evaluate text and determine which explicit claims are relevant. Write them down.

  2. Divide these into premises and conclusions.

    1. Conclusions will be supported, premises will provide support.

    2. There may be more than one conclusion, since texts can contain multiple arguments.

    3. A given claim could be both a premise and a conclusion. Arguments can be nested, and when they are, conclusions of sub-arguments serve as premises in larger arguments.

  3. Organize the claims into groups. (Keep in mind that there will be conflicts, and when there are, you should invoke the Principle of Charity; however, it isn’t always the case that there is an easy resolution to such conflicts. Some arguments are just confused, and no amount of massaging can remove the knots.) Organization can proceed using the following defeasible guidelines:

    1. One conclusion per group. As above, a claim can be a conclusion in one group and a premise in another.

    2. If two claims are in the same paragraph (section, etc.) and the argument comprises that paragraph (section, etc.), then put them in the same group.

    3. If two claims focus on the same subject matter, put them in the same group. The subject matter should be as finely individuated as the text in question allows.

    4. Pay attention to rhetorical markers. Markers like "first", "second", etc. are occasionally used to signal moves in an argument, and when they are, they will influence the distribution of claims to groups.

  4. Cast the organized groups of steps in standard form, with the conclusion below the horizontal line.

  5. Arrange the premises by putting those that are topically the most closely related to the conclusion nearest the horizontal bar, and then moving up from the bar in a way that reflects increasing topical distance from the conclusion.

When you reach the end of this procedure, you should have at least one argument in standard form. With this in hand, you are ready to proceed to argument analysis, the stage where arguments are evaluated and critical thinkers judge their merits.