1. Introduction^[1]

This chapter reviews what is involved in identifying, interpreting, displaying, evaluating and responding to arguments. The term ‘judging’ encompasses all these activities, since all of them call for judgment. Most critical thinking textbooks focus on arguments that are reason-claim complexes. These are arguments about what is claimed to be true or reasonable to believe or what to do. Their conclusions are either propositions—sentences with truth values (i.e., are true or false, probable or improbable, plausible or implausible)—or else they are prescriptions—recommendations or imperatives, (which are not true or false, but rather are sensible or ill-advised, wise or foolish, virtuous or vicious, and so on). It might turn out that different kinds of judgment are appropriate for these two kinds of aims of argument.

[Other chapters in this book go into these matters in greater detail. In Chapter 10, pioneers and experts in diagramming arguments, Martin Davies, Ashley Barnett & Tim van Gelder, explain how to use computer programs to teach students how to map the structure of arguments. In Chapter 14, Christopher Tindale offers a theoretically up-to-date introduction to using fallacy analysis in evaluating arguments.]

Here “judging arguments” includes: (1) judging that a particular text of discourse is or contains an argument, (2) judging just what the contents and structure of each such arguments consist of, (3) judging the strengths and weaknesses of each such argument, and (4) judging how best to respond to the argument(s). Correspondingly, judging arguments in this sense will have these four components, in this order: (1) argument identification, (2) argument analysis or mapping, (3) argument evaluation, and (4) responding to an argument.

The order matters, because each successive judgment presupposes the successful completion of its predecessor(s). Also, the judging can break off at any point. Obviously if it’s determined that there is no argument, the exercise ends at (1). But sometimes although it is clear that someone is trying to argue, their communication can be so unclear that it is impossible to decide just what the argument is supposed to be, so the exercise ends at (2). Furthermore, it’s often the case that there is no response to make beyond registering and defending one’s assessment of the argument, so the exercise ends at (3).

Finally, while these four judgments can be distinguished, in practice there is usually interaction among them. For instance the presence of illatives such as ‘since’ and ‘therefore’ can signal (among other things) both (1) the presence and (2) the structure of an argument. Also, often the analyst’s attribution of (2) one structure instead of another can depend on which of the two renders (3) the more plausible or stronger argument.^[2] And (4) what response to make to an arguer obviously depends partly on (3) how one assesses the merits of the argument, and perhaps even on (2) how one has analyzed the map of the argument.

 

2. Identifying arguments

Sometimes a writer will identify his or her contribution to a discourse as an argument:

“My argument for this claim is this:….”

“There are several reasons for adopting this view. First, ….”

“Not everyone agrees with this claim. Here is why I think it is true. ….”

And so on. (Such self identifications will be mistaken if what the speaker or writer has produced is not an argument, but, for example, merely an assertion, or an explanation.)

Critical thinking textbooks almost invariably point out that people use semantic cues to make clear that they intend to be communicating an argument. Words such as ‘therefore,’ ‘since,’ ‘because’ and ‘so’ can mark the illative relation of support or consequence, and thereby be signs of the presence of argument. Unfortunately, they are not foolproof signs, for all of the so-called illatives can be and often are used to perform other speech functions besides signaling the premise-conclusion relation indicative of an argument. So additional cues need to be considered.

Arguments can be anticipated in a variety of contexts. Disagreements typically give rise to arguments, so discourse in a context of disagreement can be expected to be argumentative. Some venues are institutionalized to require arguments. In criminal and civil trials, lawyers are expected to argue, and judges are expected to give arguments for their rulings. Even when doubt has not been expressed, there can be informal customs or institutionalized norms that place the burden of proof on anyone who would assert a claim, as in the case of scientific reports, scholarly articles in the humanities, or the aforementioned judges’ rulings. The so-called “editorial” pages of newspapers, which contain editorials by the paper’s editors, opinion pieces by regular or occasional columnists, and letters to the editor, are places where controversies are addressed (or initiated), and so where any “side” needs to be backed up by the reasons that its advocate thinks support it and that might persuade others to accept it.

Experienced speakers or authors write and speak with the audience they are addressing or hope to reach in mind. If you can anticipate the kinds of doubts or questions the likely intended audience will harbor about the claims asserted, you can identify where the writer or speaker needs to be providing arguments. This can help to identify their presence in parts of the text that otherwise are hard to account for.

Yet another indication of the presence of an argument in a stretch of discourse is that some of the assertions “make sense” as support for one of the others. That is, even if there are no cues, if some of the assertions were to be taken as support for another one and that would be a plausible argument, then the discourse may reasonably be taken as containing a plausible argument, if there are no contrary indications.

 

3. The structure of arguments: Argument mapping

Since the reason a critical thinker attends to arguments is to judge whether they provide the support for their claims that they’re alleged to, the critical appraiser needs to understand accurately just what that support is supposed to be. That means getting a clear picture of the route that the author’s reasons take in providing that support. Maps or diagrams showing the structure of an argument’s reasoning can be drawn by hand; they can also be displayed on a computer’s monitor. Drawing a diagram, or fashioning one on a computer monitor, that displays the route(s) the reasoning taken is called “argument mapping”. Maps or diagrams showing the structure of an argument’s reasoning can be drawn by hand; they can also be displayed on a computer’s monitor. In either case, sets of conventions have to be adopted (and learned by the reader or viewer). Thus any argument-mapping program will need what old-fashioned geographical maps used to call a “Legend”. [Argument mapping, including computer-assisted argument mapping, is discussed in detail in Chapter 10.]

The analyst faces several choices when dealing with with arguments “on the hoof”—i.e., “in their native habitats” or “as found”. Below are described various aspects of an argument’s structure. It should not be overlooked, however, that these structural properties are determined by functional properties of the discourse. For instance, what counts as a premise or a conclusion is determined by the functions of the sentences in the discourse. Does p serve as support for q? If so, then p is a premise and q is a conclusion; or does q serve as support for p? If so, then vice versa.

3.1 Ordering

There is no convention in English composition for the order in which premises and conclusions appear in a text containing an argument, and in practice any order may be found. Thus it’s to be emphasized that ‘conclusion’, when used to refer to the claim being defended in an argument, has a different meaning than when it is used to refer to the final section of a text.

The analyst will have to identify which assertion states the conclusion, and decide on a convention for organizing the premises spatially on the page.

3.2 Deletion

In addition to the sentences expressing the premises and conclusion of an argument, typically there are parts of the discourse that are not elements of the argument. They might be performing other functions, such as clarifying, or explaining, or they might be simply irrelevant to the argument. Such material can be set aside, not to appear on the map of the argument.

3.3 Multiple-arguments

Frequently single arguments are combined in various ways. There can be a single main conclusion, but one or more of the premises directly supporting it are themselves the conclusions of other arguments. Or there can be more than one line of reasoning supporting the same conclusion. Some theorists hold that meta-arguments are supplied because arguments are always occasioned by doubts, disagreements, questions or challenges, either from an actual interlocutor or from one imagined or anticipated by the arguer. If so, the arguer who inserts a meta-argument must have anticipated some challenge. Other theorists, while they grant that this is often or even usually the case, hold that arguers can offer additional arguments to cement their case, whether that’s needed to answer doubts or not.^[3] Yet others hold that, at least in some special fields if not generally, there are conventions that require every assertion to be defended unless one of the standard exemptions applies (e.g., it is self-evident, or in some sense primitive, or defended elsewhere) even if there are no actual or anticipated doubts.

Here is a simple example, in this case an argument about what to do, taking 1 to be meant to convey that the addressee should take her raincoat.

1 You’ll be wanting to take your raincoat. 2 It might rain this afternoon. 3 At least, that was the forecast on this morning’s news. 4 You won’t want to get your new dress soaked.

Here, 2 and 4 combine to support 1, and 3 supports 2.

3.4 Addition

Consider simple arguments like these. “You should take your raincoat. It’s going to rain.” Or “He’ll be late for dinner. His train’s been delayed.” In each case, how does the reason offered provide grounds for the conclusion? In each case there’s an understood connection. If you’ll want to have your raincoat if it rains, then if it’s going to rain, you should take your raincoat. In the other case, if his not arriving at the regularly scheduled time will cause him to be late for dinner and if the train’s being delayed means he’ll arrive later than scheduled, then he will be late for dinner. By adding an obviously true piece of unexpressed information, the analyst can make the reasoning of the argument more perspicuous.

How does one know what to add? One approach is to add the associated conditional (AC) of the argument as a new premise. The associated conditional is the conditional statement former by taking the conjunction of the stated premises as its antecedent and the conclusion as its consequent. For the above examples, this advice would yield these reconstructions:

A.
1. It’s going to rain.
2. If it’s going to rain, you should take your raincoat. (AC)
3. You should take your raincoat.

B.
4. His train has been delayed.
5. If his train has been delayed, he will be late for dinner. (AC)
6. He’ll be late for dinner.

Adding the associated conditional renders the argument deductively valid (by modus ponens), but it does so often at the cost of adding a questionable new premise. In arguments about what to do, like A, it overlooks the possibility of contrary considerations. If you are going to be indoors the whole time and/or if your raincoat would be an unwanted encumbrance, 2. is ill-suited advice. In arguments about what to believe like B, it overlooks possible conditions of rebuttal. If dinner will be delayed too, for some other reason, 5. might be false; or if the delay is too long he will miss dinner altogether, and 5. would be false. If any of such arguments’ premises is false, the conclusion is not supported. This result has led some theorists to call for a modification of the associated conditional, and sometimes of the conclusions too, to make the argument plausible, consistent with the known commitments of the arguer. Thus the first argument might be restated as:

C.
7. It’s going to rain.
8. You’ll want your raincoat if it rains.
9. You should take your raincoat.

And the second might be restated as:

D.
10. His train has been delayed.
11. The train delay will probably delay his arrival until after dinner begins.
12. He’ll probably be late for dinner.

As Hitchcock (2017, p. 60) has noted, the traditional treatment of such arguments has variants:

Ennis (1982), for example, distinguished two types of what he calls ‘gap fillers’: used assumptions and needed assumptions. Govier (1987, 1992) favours a policy of ‘no supplementation without justification’ (1992. p. 50), which leads her to focus on the first of these types, what the arguer implicitly accepts or can reasonably be assumed to accept; she does however allow that one may supply a ‘missing premise’ which is implied by ‘the direction of the reasoning’, a notion for which she gives no theoretical analysis. Van Eemeren and Grootendost (1984, 1992) regard the unexpressed premiss as implicit in the argument, thus focusing on the second of Ennis’s two types. Anderson and Belnap (1961, p. 719) treat the task of evaluating the inference of an enthymeme neither as one of discovering some unstated claim which the arguer accepts nor as one of discovering some further premiss which is implicit in the argument, but rather as one of discovering an additional true sentence from which in combination with the stated premise the conclusion follows logically.

Hitchcock himself objects to the notion that it is necessary to supply unexpressed premises. He argues that every argument assumes a general conditional that its proponent thinks warrants the inference from the stated premises to the conclusion. Some such conditional inference license is presupposed by every argument. It is not another premise, just unexpressed, for if it were, the argument would require yet another inference license warranting the new set of premises as adequate support for the conclusion. But then the new inference license would have to be treated as another unexpressed premise to be added to the premise set, giving rise to the need for yet another inference license, and so on ad infinitum.

3.5 Restatement

Most theorists of argument analysis allow for restating the arguments in a text in order to make the reasoning of the arguments more perspicuous. For instance arguers will use pronouns to refer to people and things in their arguments, and when arranging the premises in an order that exposes their steps towards the conclusion there can be a number of ‘he’s and ‘she’s and ‘it’s that are confusing. It doesn’t change the sense of the argument if these are replaced by their referents. Arguers employ synonyms for stylistic variation, but it can be clearer if the same term is used for a given referent throughout the reconstructed argument.

However, there are limits to restatement. Often the associations of given words lend force to the argument, and if a particular term used by the arguer is substituted by the analyst when reconstructing the argument, the argument can lose some of its persuasive power. Also, if question-begging value laden terms are replaced during reconstruction, possible illicit devices will be cleansed and the argument can be made to appear less specious than it is.

On balance, probably a good rule of thumb is as far as possible to leave the language of the original argument untouched, and make changes only if the argument is otherwise unintelligible.

 

4 Evaluating arguments

This topic turns out to be more complicated than many textbooks seem to assume. Many suggest that to evaluate an argument one needs to determine whether its premises are true, and if so, whether they support its conclusion. An argument is thus to be accepted or rejected, on these grounds. But think of D.J. O’Keefe’s distinction between an argument that someone makes (argument₁), and an argument that two (or more) people have (argument₂). An argument that people are having can possess a number or virtues, or their corresponding vices. It can be judged to be friendly or unfriendly, sympathetic or hostile, constructive or destructive, pig-headed or accommodating, personal or detached, and so on. These and similar qualities do not apply to an argument that someone makes. Arguments of the latter sort are more or less convincing, provide strong or weak support, and so on.

Arguments about what to do introduce further complications. For while the audience might agree with the values and goals and the means-ends relations appealed to in the argument in support of the prescription, and thus grant that they constitute good reasons for it, its members might at the same time espouse other values and goals that weigh against the prescription and that in their judgment carry more weight. Thus, while granting the merits of the argument as far as it goes, they do not agree that it provides adequate support for its conclusion. To be decisive, such arguments need to include a premise to the effect that the audience will find no overriding contra-considerations, and such a premise will be difficult to defend.

Scholars grounded in different fields tend to work with different conceptions of argument, and as a result, to evaluate arguments differently. Communication theorists tend to focus on arguing—the characteristics of the communicative exchange in various kinds of arguments that that people have with one another. Theorists from linguistics backgrounds tend to focus on the pragmatics of arguing—the different uses to which arguments are put, the way language works in arguings, and the practical functions of arguing. Philosophers tend to be interested in the logic of the arguments used in arguing and the conditions under which such arguments contribute to justified belief and knowledge acquisition. Rhetoricians tend to focus on arguments about what to do (Kock 2017).

4.1 Soundness

There have been various approaches to determining what makes for a good argument—“good” in the sense that its audience should either be convinced by it or in the sense that it should influence the audience to be inclined towards accepting its conclusion.

As ‘sound’ is often used in logic, a “sound” argument has true premises and they either logically entail the conclusion or they provide strong inductive grounds for accepting the conclusion. Some would both strengthen and weaken the truth condition here, so that a “sound” argument’s premises are either known (by arguer and target audience) to be true, or else are reasonably believed (by them) to be true.

Some regard these two criteria—combinations of premise truth or reasonable believability with inferential deductive validity or with inductive strength—as exhaustive. “A good argument is either deductively valid or inductively strong”, it is said. That view is true by definition of you define ‘inductively strong’ to cover any argument that has a good inference that is not deductively valid. “Inductive strength” thereby becomes a catch-all, to denote a property of any good argument that is not deductively valid. Others recommend distinguishing a wider variety of ways non-deductively valid arguments can be good arguments nonetheless.

Many theorists today classify logically good arguments as either deductively valid or defeasible. An argument that is rationally compelling but not deductively valid is said to be defeasible. Thus a strong inductive argument is defeasible, but so are some arguments that aren’t typical inductive arguments. An appeal to authority would be an example. Moral arguments that appeal to rules that can have exceptions are defeasible. The premises of a strong but defeasible argument provide good reasons for accepting the conclusion, even though a situation is conceivable in which some additional statement that is consistent with the premises but inconsistent with the conclusion is also true.

Two approaches to determining whether an argument is good side-step the dispute over whether “deductively valid and inductively strong” exhaust the class of rationally compelling arguments. One approach is to check to see whether the argument scheme employed is appropriate and correctly used; the other approach is to employ the criteria of acceptability, relevance and sufficiency.

4.2 Argument scheme assessment

Here is how Walton, Reed and Macagno (2008, pp. 1-2), leading theorists of this approach, introduce the idea of argument schemes:

Argumentation schemes are forms of argument (structures of reasoning) that represent common types of arguments used in everyday discourse, as well as in special contexts like those of legal argumentation and scientific argumentation. They include the deductive and inductive forms of argument that we are already so familiar with in logic. However, they also represent forms of argument that are neither deductive nor inductive, but that fall into a third category sometimes called defeasible, presumptive, or abductive. Such an argument can rightly carry weight, or be a plausible basis for acceptance.

Walton et al. include a compendium of 60 schemes representing “the most commonly used forms of argument” (2008, p. 308; see pp. 308-346).^[4] To each scheme is attached a set of “critical questions” (ibid., p. 3). These are questions that test an argument instantiating such a scheme in any circumstance and which, if answered satisfactorily in any particular case, authorize the argument as holding, at least tentatively.

Walton, who is recognized world-wide as an authority on argument scheme theory, introduces it thoroughly in Chapter 11, above. The reader who wants to know about it can turn there for a detailed account.

Using argument schemes to assess arguments does not require having on hand a list of schemes with their accompanying critical questions and looking these up, like checking in a book of recipes. Instead, the assessor can formulate the pattern of reasoning that the argument in question exhibits and then sketch the conditions that would have to be satisfied for any argument of this pattern to be plausible, thereby creating the pertinent critical questions. If the use of the argument in question on this occasion satisfies those conditions, it can be judged to be plausible; otherwise, not.

4.3 Checking Acceptability, Relevance and Sufficiency

Another general method of assessment is to use the criteria triad of Acceptability, Relevance and Sufficiency—ARS. The contention is that an argument is good if, but only if, its grounds are acceptable, relevant and sufficient. Govier (e.g., 1992) uses “adequacy of grounds” in place of “sufficiency”. What constitutes each of these elements needs to be specified.^[5]

Here is a sketch of an analysis of the acceptability of premises as a criterion of argument cogency. Being acceptable is a property of a sentence, p, in relation to a person, S. In this method of argument assessment, however, acceptability is the special case of a criterion for the adequacy of unsupported premises in an argument, A. A premise of an argument, A, is unsupported if no reasons are offered as part of A in its support. Such premise acceptability is relative to any person, S, who is a member of, or joins, the audience to which A is addressed. A premise is acceptable to S if S is entitled to accept it.

S is entitled to accept p when p is an unsupported premise of an argument, A, addressed to S or to an audience of which S is a member, or that S entertains, and:

1. S knows that p, or S believes that p and is entitled to believe that p, or;
2. p is known to be true or reasonable to believe in S’s cognitive environment, or;
3. p follows from assertions S is entitled to accept.

The idea is that an arguer does not have to defend premises that the interlocutor and members of the audience know to be true, or that they believe on good grounds, or that are matters of common knowledge that the arguer can reasonably take them to know or reasonably believe, or that are implied by any of these.

For S to accept a premise, p, of A is for S to act in assessing A as if p is true. S can be entitled to accept a premise but not accept it. That is, a person might fail or refuse to accept a premise he or she ought to accept. And S can accept a premise that S is not entitled to accept.

Relevance and sufficiency are criteria for the adequacy of the link between the acceptability of the premises and the acceptability of the conclusion. The reasons offered must be probatively relevant to the acceptability of the conclusion for S. They have a bearing on the acceptability of the conclusion for the interlocutor or audience. Their truth would, in the absence of any other grounds, make it more likely or more plausible that the conclusion is true than would be the case if they were not true. (A reason, as the term is used here, is not identical to a premise, although a single premise can be a reason. Usually reasons consist of sets of two or more premises that are only jointly relevant.) Note that this “bearing” concept does not allow for degrees of relevance. Offered grounds either have a bearing, or they don’t. There is another concept of relevance according to which evidence can be more or less relevant. According to this latter, “weight”, concept, relevance signifies strength of support, and using the distinctions I am making here, this latter kind of relevance bears on the sufficiency of the grounds.

It has been argued that relevance is redundant, since sufficiency already presupposes it. You can’t have enough evidence unless what you count as evidence is already relevant. That is true. However, people’s arguments sometimes include irrelevant reasons. Those have to be identified and set aside before judging the sufficiency of the relevant ones that remain.

Sufficiency is the requirement that the relevant reasons offered supply enough of the right kinds of evidence to entitle the interlocutor or members of the audience to accept the conclusion as it is qualified. In many cases one example constitutes anecdotal evidence, and bears hardly any weight in supporting a conclusion that generalizes to all or even most members of a class. In others, one example can suffice as proof, e.g., that something is possible. In many cases, the findings from a well-drawn stratified random sample of 2000 people can justify a probabilistic generalization applying to 350 million people (see Chapter 18, on generalizing). So both the quantity and the quality of the evidence are important for assessing its sufficiency as support for a conclusion. Whether the reasons in support of a claim count as sufficient also depends on whether alleged reasons for not accepting the claim, or alleged reasons for rejecting any of those arguments, have been successfully refuted. What counts as “enough” will vary with the precision and the generality of the standpoint. If no direct evidence for the standpoint at issue is given, it can be independently supported by arguments for rejecting alternatives to it.

These sketches of the modified ARS criteria need to be filled in, but the assumption is that they too are general in the respect that deductively valid and inductively strong reasoning and arguments, as well as those with other kinds of good consequence relations, all will pass their test.

The argument scheme approach and the ARS approach both assess the argument as a whole. A couple of other well-known approaches assess just the adequacy of the inference from the premises to the conclusion, and come into play independently of whether the premises are known to be true, or are reasonable to accept.

4.4 Testing by possible counter-examples

Testing by possible counter-examples is a way of assessing the strength of the link between reasons and conclusion. The reasons have to be appraised separately. The method is this: Step 1: think of considerations that are consistent with the given reasons but inconsistent with the claim being argued for (i.e., think of counter-examples). Step 2: Decide how likely or plausible are the possible counter-examples. Step 3: Draw the appropriate conclusion about the strength of the inference in the argument. Depending on whether any such counter-examples are conceivable, and if so, either probable or plausible to some extent, the reasoning can be determined to be deductively valid, or invalid but with some degree of inductive strength, or invalid but more or less reasonable.

norm	nature of counter-examples	premise-conclusion link
Deductive valldity	no counter-examples can be conceived	possibly deductively valid
Inductive strength	the few counter-examples conceivable are unlikely	inductively strong
	the many counter-examples conceivable are likely	inductively weak
Plausibility	the few counter-examples conceivable are implausible	highly plausible
	the many counter-examples conceivable are plausible	highly implausible

Table 1: Testing by counter-example

Table 1 does not depict it, but likelihood (and unlikelihood), and plausibility (and implausibility) here name continuums, ranging from probability or plausibility that run so high as to be treated as certainties, at one extreme, down to, at the other extreme, improbability or implausibility that run so high as to be unquestionably false, utterly improbable, or wildly implausible. (See Pinto, Blair & Parr 1993, Ch. 6.)

 

4.5 Assessing the warrants of inferences in arguments

Under the influence of Stephen Toulmin’s (1956) suggestion that arguments are more like legal briefs than mathematical proofs, some theorists hold that in arguments the inferential step is authorized by a presumed warrant. This warrant is not another premise; it is an assumption that is supposed to entitle one to draw the conclusion in question from the premises claimed to support it. Here is a statement of this view by David Hitchcock, its originator and most prominent advocate:

A conclusion follows from given premisses if and only if an acceptable counterfactual-supporting generalization rules out, either definitively or with some modal qualification, simultaneous acceptability of the premisses and non-acceptability of the conclusion, even though it does not rule out acceptability of the premisses and does not require acceptability of the conclusion independently of the premisses. … An inference claim is thus the claim that a counterfactually-supporting covering generalization is non-trivially acceptable. (Hitchcock 2017, p.180 quoting Hitchcock 2011, p. 209)

Applying this warrant-defining statement to an example, we get:

1^st premise: My coach says I have great promise to make the national team but I need more coaching and training, which a particular summer soccer school would provide.

and

2^nd premise: My coach is in a position to know such things [e.g., she has the experience and the expertise to be a reliable judge of soccer ability, national team standards, and so on]

So

Conclusion: I have a chance to make the national team if I get the coaching and training this summer soccer school would provide.

Warrant: If someone in a position to know such things were to say of a young soccer player that she has great promise to make the national team if but only if she obtains more coaching and training, then if that young soccer player were to get more coaching and training, she probably would have a reasonable chance to make the national team.

Given the premises, the conclusion follows if, but only if, the warrant is “non-trivially acceptable”. If the warrant is acceptable and the premises are true or otherwise acceptable, it is a good argument: one is entitled to accept the conclusion.

4.6 Evaluating practical arguments—(about what to do)

The conclusion of a practical argument is not that some proposition is true or probable or otherwise acceptable, but that some action or policy should be chosen, or that some decision should be made. Its conclusion will be supported by appeals to ideals, or goals or other values and by claims about how the recommended action will bring about or reflect these goals or ideals. Any disagreement might be about what values are relevant, or, when there is agreement about what values are applicable, the disagreement might be about how to weigh them. For instance many Americans might agree that peace, order and good government are important values, but many might weigh life, liberty and the pursuit of happiness more heavily if the latter values came into conflict with the former. To be sure, one can argue for the greater importance of, say, liberty over order in a given case, or vice versa, but this argument too will turn on values or goals. At certain point, disputants might have to agree to disagree.

In light of this kind of disagreement, two kinds of critical questions can be raised in assessing practical arguments. One is whether all the values or ideals bearing on the disagreement have been mentioned. It might turn out that some overlooked value will change the argument. Another question is whether the means-ends arguments being used are correct. Is the action being argued for really required to reach that objective? In the end, however, the critic’s objections will often come down to a difference in values deemed relevant, or to a difference in the weighting of the values agreed to be relevant. When there has been a thorough and open-minded discussion of the issues, the verdict might have to be that the disputants simply weighted the values differently.

Notice that when as decision in the end comes down to subjective preferences and these clash, “sufficiency” can apply only as a criterion of whether the best case for each side has been made. By hypothesis, once the best case for each side has been argued and these have been understood and appreciated by each side, the final decision is not the outcome of further argument.

4.7 Checking for fallacies

Many textbooks include lists of fallacies judged to trip up arguers or to be used by unscrupulous arguers to disguise poor arguments or shore up a weak case. Presumably textbook publishers include this material to meet a demand.

The current scholarly literature on fallacies, however, is not always reflected in these textbook treatments. For instance, it used to be held that a fallacy was a misleading pattern of argument, one designed to look valid and sound sound but that in fact was neither. Fallacies were thus viewed as counterfeit arguments as Fearnside and Holther’s (1959) titled asserted: Fallacy, The Counterfeit of Argument. Today, the dominant view in the philosophical literature, following Walton (1995), is that fallacies are mis-deployments of otherwise innocent and useful argument schemes. In the speech communication literature, many follow van Eemeren and Grootendorst (1984) and van Eemeren (2010), who argue that fallacies are best conceived as violations of the rules for reasonable discussions designed to resolve a difference of opinion, occurring when a desire to win the argument overrides the commitment to reasonableness that engaging in argumentation presupposes. It’s not clear that a fallacy-free argument is thereby a good argument, but a fallacious argument is flawed. Hence, checking for fallacies might be treated as a preliminary assessment—rooting out the worst arguments.

In light of the continuing interest in fallacies as a tool for argument appraisal, and with a view to offering an up-to-date picture of fallacy theory to instructors, this book includes a separate chapter, Chapter 14, devoted to the topic, “Introduction to the study of fallaciousness”, by Christopher Tindale, taken from his book Fallacies and Argument Appraisal (2007).

 

5. Responding to arguments

What is the appropriate way to respond to an argument that one has assessed and made a considered judgment of its strength? It is useful at this point to keep in mind that our evaluations are judgment calls. Not that they are subjective intuitions (although we do form quick initial opinions). There are criteria for logically good, rhetorically good, and dialectically good arguments. But there can be and often are reasonable disagreements over whether a criterion has been met and over whether the standards being invoked for each criterion are appropriate in kind and rigor. Argument assessments may have to be defended by further arguments and are always subject to reappraisal. Hence it is wise to express one’s verdict about an argument with a healthy dose of humility. The arguer, or other critical reviewers of the argument, might well have good rejoinders to your critique—points that had not occurred to you and that on reflection have both merit and implications requiring you to modify your initial judgment.

It’s often a good idea to distinguish between one’s judgment as to how good the argument is and one’s judgment as to how best to respond to the arguer, if you are conversation partners. Suppose you find the evidence offered to be woefully weak. It might be more productive not to say that, but instead to suggest that the argument would be a lot stronger if evidence such as X, or Y, could be added. The point is that how you communicate your judgment of the argument should depend on what you hope to achieve by what you say. Do you want to help the arguer produce a stronger argument? If so, obviously avoid insulting criticism of the argument as it stands. Do you want to convince the arguer that the argument is fatally flawed (e.g., if it is circular or question-begging)? In that case, there’s no avoiding pointing out how it is, though it might be possible to do so without using offensive labels. For instance, instead of labeling (e.g. “You’ve set up a Straw Man!”), being conciliatory (e.g., “That would be a good objection if that were my view, but my view is different, namely ….”). In short, think of the effect of your critique on the author and think of your objective in communicating with the author, and make your comments serve your objective.

 

6 Summary

This chapter has introduced four aspects of judging arguments in texts of discourse: determining whether there is any argument in the discourse, analyzing the structure of the argument, different ways of evaluating the logic of the argument, and various ways of responding to it. Soundness, appropriate use of an argument scheme and the ARS approach were reviewed as methods of overall argument appraisal. Testing for counter-examples and assessing the argument’s warrant were discussed as ways of assessing the merits of the reasons-conclusion link in an argument. Fallacy identification was listed separately because some fallacies involve problems with the reasons and others problems with the reasoning. The possibility was raised that arguments about actions or policies (what to do) are to be assessed differently from arguments about propositons (what to believe).

 

References

Anderson, Alan Ross & Belnap, Nuel D. Jr. (1961). Enthymemes. Journal of Philosophy 58 (23): 713-723.

van Eemeren, Frans H. & Grootendorst, Rob. (1984). Speech Acts in Argumentative Discussions. Dosrdrecht: Foris.

van Eemeren, Frans H. & Grootendorst, Rob. (1992). Argumentation, Communication and Fallacies. Hillsdale, NJ: Lawrence Erlbaum.

van Eemeren, Frans H. & Grootendorst, Rob. (2004). A Systematic Theory of Argumentation, The Pragma-Dialectical Approach. Cambridge: Cambridge University Press.

van Eemeren, Frans H. (2010). Strategic Maneuvering in Argumentative Discourse. Amsterda: John Benjamins.

Ennis, Robert H. (1996). Critical Thinking. Upper Saddle Rivere, NJ: Prentice Hall.

Fearnside, W. Ward & Holther, William B. (1959). Fallacy, The Counterfeit of Argument. Englewood Cliffs, NJ: Prentice-Hall.

Hitchcock, David. (1987). Enthymematic arguments. In Argumentation, Across the Lines of Discipline. Proceedings of the Conference on Argumentation 1986, Frans H. van Eemeren, Rob Grootendorst, J.

Anthony Blair & Charles A. Willard (Eds.,), pp. 289-298. Dordrecht: Foris.

Hitchcock, David. (2011). Instrumental rationality. Argumentation in Multi-Agent Systems. 7^th International Workshop, argMAS 2010 Revised Selected and Invited Papers, Peter McBurney, Iyad Rahwan & Simon Paarsons (Eds.), pp. 1-11. Heidelberg: Springer.

Hitchcock, David. (2017). On Reasoning and Argument, Essays in Informal Logic and on Critical Thinking. Cham, Switzerland: Springer.

Govier, Trudy. (1987). Problems in Argument Analysis and Evaluation. Dordrecht: Foris.

Govier, Trudy. (1992). A Practical Study of Argument, 3^rd ed. Belmont, CA: Wadsworth.

Johnson, Ralph H. & Blair, J. Anthony. (2006). Logical Self-Defense. New York: IDEA Press.

Kock, Christian. (2017). Deliberative Rhetoric, Arguing About Doing. Windsor, ON: Windsor Studies in Argumentation.

O’Keefe, Daniel J. (1977). Two concepts of argument. Journal of the American Forensic Association 13 (3): 121-128.

O’Keefe, Daniel J. (1982). The concepts of argument and arguing. In J. R.Cox & C.A. Willard (Eds.), Advances in Argumentation Theory and Research. Carbondale, IL:Southern Illinois University Press.

Pinto, Robert C., Blair, J. Anthony & Parr, Katharine E. (1996). Reasoning, A Practical Guide for Canadian Students. Scarborouogh,ON: Prentice-Hall Cana

Sperber, Dan & Wilson, Diedre. (1986). Relevance: Communication and Cognition. Cambridge, MA: Harvard University Press.

Tindale, Christopher W. (2007). Fallacies and Argument Appraisal. Cambridge: Cambridge University Press.

Tindale, Christopher W. (2015). The Philosophy of Argument and Audience Reception. Cambridge: Cambridge University Press.

Toulmin, Stephen, N. (1956). The Uses of Argument. Cambridge: Cambridge University Press.

Walton, Douglas. (1995). A Pragmatic Theory of Fallacy. Tuscaloosa: University of Alabama Press.

Walton, Douglas, Reed, Chris & Macagno, Fabrizio. (2008). Argumentation Schemes. Cambridge: Cambridge University Press.