Fallacy

8 Faulty Classification
The violation of each of the four rules gives a different fallacy. Some of the rules, however, are more complicated than others and will be treated here at greater length. There are many ways in which a construction can be vague (Rule 4), and it will be useful to distinguish several of them in the discussion. Similarly, the third rule is related to some misconceptions about the nature of classification, and we shall also wish to discuss these at length. The fallacies of non-exhaustive or non-exclusive classifications should be relatively easy to understand, and one or two examples will probably suffice. So under the present heading, though all violations of the rules are properly regarded as "faulty classifications," we shall take up only the failures to be exhaustive and exclusive.
(a) Non-exhaustive Classification (Rule 1)
To criticize a classification for being non-exhaustive is to assert that it does not take account of the whole situation. It is always possible to divide, that is to classify, the situation so that everything is regarded as one thing or another without remainder: beautiful or non-beautiful, true or non-true. Such divisions are model classifications as to being exhaustive, but they may be far from adequate to a given purpose (see #6). The difficulty arises when the classification is supposed to distinguish important characteristics in the situation.
In argument, where the classifications are often more complicated than cases like the marbles, one way to test a position depending on them is to search for elements in the situation which have got left out. If one finds such elements, it by no means follows that the classification is deliberately selective (see #1) or deceptive. It may just be careless. It is often possible to correct classifications that leave something out by a more careful attention to the basis of division.
EXAMPLE COMMENT
Traveling Salesman Peter has submitted his expense account. A few days pass, and the boss calls him in: "I understand the $421 for plane fare, the $128 for hotel bills, the $84 for dining out, and the $37 for entertaining. But what is this $202 for miscellaneous?" Peter shrugs, "Oh, that's for things I couldn't fit in." Possibly Peter will satisfy his boss, but if he uses the same principles of accounting in his income tax returns, he can expect a visit from the treasury agents. This example is not an argument. It merely illustrates the frequent use of the feeble "miscellaneous" heading to take care of classifications which are not exhaustive.
Overheard on a bus: "She never says a kind word, and when she does, she doesn't mean it." The speaker here has made what is called an Irish bull, that is, a statement that on the face of it is contradictory nonsense: "I'll never forgive you until the day you die, if you should live so long." The Irish may use the bull for humorous effect, but the speaker in the example presumably recognized that the distinction "kind or unkind" didn't cover all the evidence, and that a third division "only apparently kind" was needed.
A critic writes, "Ivy Compton Bennet and Paul Goodman, to name two, are lively writers pointing to new directions. Yet important modern novels fall into three classes, all unfortunate for contemporary letters. There is the historical novel, more or less carefully reconstructing a period or a situation of the past and exploiting it to give us an understanding of our heritage. This is the proper function of the historian or biographer, not of the novelist. There is, secondly, the psychological study, which purports to convey by literary means insights into personal relations. This seemed a promising field initially, but all the novels of this school are failures, and now it is evident that failure is inherent in the method. Finally, there is the novel of manners, in the tradition of Jane Austen and Henry James. But this tradition is exhausted, and what we get today are problem pieces and regional studies, like those of the southern school." Even if it can be assumed that the critic's elliptical arguments convey sound judgments -- that, in fact, each of the three classifications mentioned has failed and is now creatively useless, the unspoken conclusion that the modern novel is in a bad way as a literary form follows only if it is true that there are no other varieties of novels than those mentioned. There seem to be many interesting novels that do not fit the categories of the critic, some of them also very different from each other, for example the novels of Ivy Compton Burnett and Paul Goodman, whom the critic himself mentions.

(b) Non-exclusive Classification (Rule 2)
A classification is non-exclusive when elements in the situation get put under more than one co-ordinate head. The trouble here can sometimes be corrected by choosing a different basis of division. A simple artificial example would be for someone to attempt to divide animals as "cold-blooded" or "vertebrates." Since reptiles, fish, etc., are both, this classification does not operate for these important sections of the animal world. The classification is easily corrected by distinguishing first among the vertebrates and invertebrates, then perhaps among the cold and warm-blooded animals. The shark then is properly classified as species vertebrate, subspecies cold-blooded; elephants as vertebrates, warm-blooded; crabs as invertebrates, cold-blooded. This is not very profound, but it is at least applicable to the animal world.
One way to examine an argument dependent on a classification is to test whether or not the classification is exclusive. If it is not for any one case at all, then there is a possibility that the classification is not applicable to the case or cases at issue in the argument. If the classification is faulty in general, then it may be faulty in important particulars. Again it should be observed that the fault may not be deliberate. It is very hard to make adequate classifications, and the classifier may simply not have noticed that his divisions are not applicable to certain elements of the situation. The thing to do is to examine the fundamentum divisionis, the basis of the division, and see if the classification can be repaired.
EXAMPLE COMMENT
Peter argues, "You either save your money or you invest it. If you save it, you get caught in the inflation, since money in the stocking will buy less tomorrow than it will if you spend it today. If you invest it, you run the risk of losing it. They've got you going and coming." Many forms of savings are relatively free of risk -- deposits in banks, government bonds, and some types of investment trusts. The accumulated interest is normally larger than the rate of inflation. In the case of sound stocks and some real estate, the value inflates with the rest of the economy. Peter's argument is a false dilemma.

All arguments pose a dilemma insofar as they rest on classification. The dilemma becomes a false dilemma when the classification is faulty. False dilemmas turn on a non-exclusive classification whenever one of the alternatives is not as represented, just as it will be non-exhaustive whenever there is a third alternative. In this connection it will be instructive to review #7 and see if you can decide which of the dilemmas cited there get their plausibility from non-exhaustive classifications, which from non-exclusive, that is, which violate Rule 1 and which Rule 2.

EXAMPLE	COMMENT
Traveling Salesman Peter has submitted his expense account. A few days pass, and the boss calls him in: "I understand the $421 for plane fare, the $128 for hotel bills, the $84 for dining out, and the $37 for entertaining. But what is this $202 for miscellaneous?" Peter shrugs, "Oh, that's for things I couldn't fit in."	Possibly Peter will satisfy his boss, but if he uses the same principles of accounting in his income tax returns, he can expect a visit from the treasury agents. This example is not an argument. It merely illustrates the frequent use of the feeble "miscellaneous" heading to take care of classifications which are not exhaustive.
Overheard on a bus: "She never says a kind word, and when she does, she doesn't mean it."	The speaker here has made what is called an Irish bull, that is, a statement that on the face of it is contradictory nonsense: "I'll never forgive you until the day you die, if you should live so long." The Irish may use the bull for humorous effect, but the speaker in the example presumably recognized that the distinction "kind or unkind" didn't cover all the evidence, and that a third division "only apparently kind" was needed.
A critic writes, "Ivy Compton Bennet and Paul Goodman, to name two, are lively writers pointing to new directions. Yet important modern novels fall into three classes, all unfortunate for contemporary letters. There is the historical novel, more or less carefully reconstructing a period or a situation of the past and exploiting it to give us an understanding of our heritage. This is the proper function of the historian or biographer, not of the novelist. There is, secondly, the psychological study, which purports to convey by literary means insights into personal relations. This seemed a promising field initially, but all the novels of this school are failures, and now it is evident that failure is inherent in the method. Finally, there is the novel of manners, in the tradition of Jane Austen and Henry James. But this tradition is exhausted, and what we get today are problem pieces and regional studies, like those of the southern school."	Even if it can be assumed that the critic's elliptical arguments convey sound judgments -- that, in fact, each of the three classifications mentioned has failed and is now creatively useless, the unspoken conclusion that the modern novel is in a bad way as a literary form follows only if it is true that there are no other varieties of novels than those mentioned. There seem to be many interesting novels that do not fit the categories of the critic, some of them also very different from each other, for example the novels of Ivy Compton Burnett and Paul Goodman, whom the critic himself mentions.

EXAMPLE	COMMENT
Peter argues, "You either save your money or you invest it. If you save it, you get caught in the inflation, since money in the stocking will buy less tomorrow than it will if you spend it today. If you invest it, you run the risk of losing it. They've got you going and coming."	Many forms of savings are relatively free of risk -- deposits in banks, government bonds, and some types of investment trusts. The accumulated interest is normally larger than the rate of inflation. In the case of sound stocks and some real estate, the value inflates with the rest of the economy. Peter's argument is a false dilemma.