On Classifications of Fallacies

VIII. 2, Spring 1987 Informal Logic On Classifications of Fallacies MICHAEL F. SCHMIDT San Jose State University Research suggests to me the conclu ever can be" [1] sion that three fundamental kinds of C.L. Hamblin writes: "We [in 1970] fallacies are-interrogative, premisory, have no theory of fallacy at all, in the and inferential. However, in this arti sense in which we have theories of cle, the focus will be on premisory fal correct reasoning or inference. Yet we lacies. In particular, the article aims to feel the need to ticket and tabulate show that there are four distinct kinds [fallacies] .... "[2] Gerald J. Massey of premisory fallacies, "begging the argues that an account of fallacies pre question", "inconsistency", "falsity", supposes a theory of invalidity and that and "omission". The article begins "a theory of invalidity has yet to be with a very brief review of some schools developed"[3], while others argue for of thought about classifying fallacies; a "theory of reasoning". One of this then it turns to the task of reviewing journal's referees points out that "no some actual classifications of fallacies a priori scheme of classifications can to see what the tendencies are; five fairly be evaluated until there is some tendencies are found; from them unified theory of argument." emerges the conclusion that "incom And, Peter Ramus (in 1543) wrote: plete information" is distinct from First, should not the overall description "false premise" and from "invalid of [dialectical] vices itself arise from the inference", that "begging the ques direct opposition of virtues so that, tion" is distinct from all three, and that for every kind of virtue there should "inconsistent premises", "false pre be just one kind of vice? And in so far mise", "incomplete information" I and as there are two overall virtues in "begging the question" form a natural dialectic, one of Invention and one of constellation of premisory fallacies, dis Judgment, so there ought to be two tinct from the fallacy of many questions overall vices, one opposed and hostile and from the fallacies of invalid in to true Invention, the other to correct Judgment; .... [4] ference. Appended is a bibliography of recent and contemporary writings Now! let us review sOme important on the topic of fallacies. and interesting classifications of fal lacies-those of Aristotle, Richard Whately, John Stuart Mill, H.W.B. Schools of thought Joseph, W. Ward Fearnside, Howard Kahane, and Ronald Munson-and There flourish several schools of then see what the tendencies are. thought about the classification of falla cies. Let us review their main tenets. It is well known that Augustus De Aristotle Morgan claimed: "There is no such thing as a classification of the ways in Probably the best known classifica which men may arrive at an error: it tion is Aristotle's. His On Sophistical is much to be doubted whether there Refutations begins: 58 Michael F. Schmidt Let us now discuss sophistic refuta Moreover, in his Rhetoric Aristotle tions, i.e. what appear to be refutations lists some spurious "lines of argu but are really fallacies instead .... ment"; there are nine in all; lithe That some reasonings are genuine, first. .. arises from the particular words while others seem to be so but are not, employed"[9]; this first spurious line is evident. This happens with argu ments as also elsewhere, through a of argument has two varieties: (a) certain likeness between the genuine "when, without having gone through and the sham.[5] any reasoning process, we make a final statement as if it were the conclusion According to Aristotle, sometimes of such a process"[10]; and, (b) when reasoning and refutation fail because we use "similar words for different arguments depend upon language: things"[11] (probably a reference to a It is impossible in a discussion to bring number of sorts of ambiguity). Note in the actual things discussed: we use that the former variety is not among their names as symbols instead of them; the fallacies enumerated in On Sophis and therefore we suppose that what tical Refutations. follows in the names, follows in the The remaining eight spurious "lines things as well, ... {6] of argument" are presumably inde Fallacies due to language include: pendent of language; they include some that are enumerated in On Sophistical word ambiguity f grammatical ambi guity (i .e., amphiboly), accent, and Refutations, including the special case three others. of "false cause" now known as "post At other times reasoning and refuta hoc, ergo propter hoc"; and, they also tion fail because of factors that are include some that are not enumerated "independent of language" .[7] Fal in On Sophistical Refutations, including lacies of this sort include: ignoratio "the use of indignant language, elenchi, begging the question, false whether to support your own case or to cause, many questions, and three overthrow your opponent's" .[12] others. Notice that Aristotle discusses Whately these two sorts of fallacies in connec tion with what he calls "contentious arguments" as opposed to "demon In Richard Whately'S Elements of strative arguments" and II dialectical Logic we find a somewhat different arguments"; that is, in On Sophistical classification: Fallacies are of two main Refutations he focuses more on ques sorts: logical and material. First, logical tions about refuting an opponent than fallacies occur "when the fault is on questions about proposing a scienti strictly, in the very process of Rea~ fic explanation or a solution to a prob soning"[13], i.e., I/[w]here the Conclu lem; and, consequently, he focuses sion does not follow from the Pre more on sham refutations than on sham mises."[14] Second, in the case of ma explanations or sham solutions to prob terial fallacies, although the conclusion lems. does follow from the premises, some Notice too that Aristotle acknow thing else goes wrong. In particular, ledges the possibility of classifying fal there are those material fallacies that lacies differently; for example. are due to an "unduly assumed" pre mise and those due to ignoratio elen The right way, then, is either to divide chi .[15] apparent proofs and refutations as above, or else to refer them all to ig Under the heading of "premise un norance of what 'refutation' is [ignora duly assumed", Whately includes tio elenchi], and make that our starting petitio principii (i.e., begging the ques point: for it is possible to analyse all tion) and "false or unsupported pre the aforesaid modes of fallacy into mise"; and, under the heading of breaches of the definition of a refuta ignoratio elenchi I he includes ad homi tion·IS] nem, ad verecundiam, etc., "shifting Classifications of Fallacies 59 ground", and two others,[16] As it is foreign to the purpose of the Notice that Whately acknowledges present treatise to decide between these that the classification of fallacies is, to confl icting [methaphysical] theories, we are precluded from inquiring into a degree, arbitrary and uncertain :[17] the existence, or defining the extent .. " from the elliptical form in which and limits, of knowledge a priori, all reasoning is usually expressed, and and from characterising the kind of the peculiarly involved and oblique form correct assumption which the fallacy in which Fallacy is for the most part of incorrect assumption, now under conveyed, it must of course be often a consideration, simulates.[23] matter of doubt, or rather, of arbitrary Fallacies of Inference comprise (1) choice, not only to which genus each the Inductive Fallacies of Observation, kind of fallacy should be referred, but even to which kind to refer anyone (2) the Inductive Fallacies of Generali individual Fallacy.[18] sation, (3) the Deductive Fallacies of Ratiocination, and (4) Fallacies of Con He does not explain why it is indeter fusion.[24] Fallacies of Confusion are minate "to which genus each kind of "fallacies in which the source of error fallacy should be referred", but does is not so much a false estimate of the explain why it is indeterminate[19] probative force of known evidence, as lito which kind to refer anyone indivi an indistinct, indefinite, and fluctua dual fallacy", He says: ting conception of what the evidence is, "[25] For, Since. in any Argument, one Under the heading of Inductive Fal Premiss is usually suppressed, it fre lacies of Observation, Mill includes quently happens, in the case of a Falla "non-observation" and "mal-observa cy, that the hearers are left to the alter native of supplying either a Premiss tion".(26] Under the heading of Induc which is nol true, or else, one which tive Fallacies of Generalization, one does not prove the Conclusion.[20] might include the two forms of the Fal lacy of Hasty Generalization: (a) He gives the following example: if a generalization from an overly small person argues from the premise that sample and (b) generalization from an 'the country is distressed' to the con unrepresentative sample. Under the clusion that 'the government is tyran heading of Deductive Fallacies of Ratio / nical , "we must suppose him to cination, one might include the Fallacy assume either that 'every distressed of Affirming the Consequent, the Fal country is under a tyranny,' which is lacy of Denying the Antecendent, etc. 'a manifest falsehood, or, merely that And, under the heading of Fallacies 'every country under a tyranny is dis of Confusion, one might include equivo tressed,' which, however true, proves cation, the "black or white" fallacy, nothing".[21] the argument from the heap, and the slippery slope fallacy. Additionally, Mill quotes Whately Mill with approval-concerning the reason why it is determinate to which class of In John Stuart Mill's A System of fallacies to refer any actual error.

Load more