The theory and practice of argumentation and debate

by Victor Alvin Ketcham

While an opponent may cover up a fallacy with the deliberate intention to deceive, yet the existence of most fallacies is not suspected by those who use them. Therefore the use of fallacious arguments is seldom evidence of dishonesty but is almost always the result of careless reasoning or inability to detect and remedy such errors. To classify fallacies into groups for the purpose of discussion is a most difficult undertaking. Any division that can be made will not prove all inclusive and all exclusive in practical application. Hard and fast divisions are sure to overlap, and a particular fallacy may be treated under one division or another according to the standpoint of the student and the combination of circumstances under which it exists. For the purpose of this discussion we shall divide fallacies according to the kind of argument in which they occur and according to the form in which they are usually found. This method of division will best serve our practical object, which is the detecting and eliminating of fallacies.

I. Fallacies of Induction.

In a perfect induction a fallacy may be detected by scrutinizing the conclusion to make sure that it includes only the specific instances upon which it is based, and then examining each of these specific instances to see that it is true as a matter of fact. If the conclusion includes more than the facts warrant or if the alleged facts are false the perfect induction is fallacious.

In searching for fallacies in an imperfect induction the rules which have already been pointed out as governing the construction of such an inductive argument should be applied. In order to make a systematic search for fallacies in arguments involving this kind of reasoning, the following steps should be taken.

1. The number of specific instances relied upon to support the inductive conclusion should be determined.

It is comparatively easy to determine the number of incidents claimed to support the conclusion, provided they are all stated in the argument. In such a case the searcher for fallacies merely counts these incidents and passes on to the next step in his investigation. Seldom, however, is the task so easy. In most arguments the writer or speaker extends his conclusion far beyond the actual facts offered in its support. Often the speaker states that “hundreds of other cases,” or “incidents too numerous to mention,” or “thousands of similar cases,” etc., can be produced to show the validity of the induction. The debater should never be overawed by such sweeping statements or allow them to cause him to cease his search for fallacies. He must be insistent in his demand that the number of incidents upon which the conclusion is based be exactly stated or at least that the number be shown as large enough to offset the probability of coincidence. The fallacy of the induction can then be shown to exist by pointing out that the number of incidents in support of the induction is not sufficiently great to warrant its acceptance.

2. The class of persons, events, or things about which the induction is made should be scrutinized with a view to determining whether it is homogeneous.

The discussion of this requirement for a valid imperfect induction which has been previously given will make plain the nature of the investigation under it. A fallacy may be exposed in such an argument by showing that the class of persons, events, or things about which the induction is made is not homogeneous in respect to the particular about which the conclusion is stated.

3. Whether or not the specific instances cited in support of the conclusion are fair examples should be determined.

It is usually easier to detect unfair examples in an opponent’s argument than in one of the debater’s own construction. The person who uses an induction is almost always prejudiced in favor of the instances which support it, but to the unprejudiced mind the fairness of a given example is not hard to determine. It is therefore important that the investigator assume an unprejudiced attitude towards the examples offered as representative of the class about which the induction is made. The existence of a fallacy in an argument based upon an imperfect induction may be repealed by showing that the specific instances cited in support of the conclusion are not fair examples.