Argumentation studies are related to two Aristotelian sources, on the one hand, the rhetorical and dialectical theories of the Rhetoric and the Topics, and on the other hand, the critical analysis of fallacious sequences (fallacies, apparent enthymemes) in the Prior Analytics, the Rhetoric and mainly in the Sophistical Refutations (Woods 2014). This last line is the basis of the “standard treatment of the fallacies” as reconstructed by Hamblin (Fallacies, 1970).
The definitions from the Sophistical Refutations are taken up in all works dealing with fallacious arguments. The title, Sophistical Refutations, is ambiguous. Firstly, according to the classic joke, it is not ‘an adequate description of the contents of the book’, that is to say, a set of refutations (concerning well defined theses) which would be sophistical, but a refutation of the Sophists’ arguments. The book analyses and rejects the refutations as practiced by the sophists, or “how the sophists refute”.
Aristotle draws a broad distinction between two sets of paralogisms. He defines, on the one hand, paralogisms that “depend on the language used”, and on the other, paralogisms which are “independent of language” (SR, 4). The “language” referred to is the language used in a dialogue, as practiced by the dialecticians or the sophists.
The Rhetoric lists ten “lines of argument that form the spurious enthymemes” (Rhet, ii, 24, 1400b35-01a5, RR 379), clearly related to language. Note that this parallelism enthymeme / spurious enthymeme may lead us to believe that the preceding enthymemes, as enumerated in Rhet., II, 23 are valid, which is not the case. S. Collections (2); S. Expression.
1. The fallacies in the Sophistical Refutations
The six Aristotelian linguistic fallacies are listed in the first column of the following table:
|
Six fallacies “dependent on language” or “verbal fallacies” (lat. in dictione ) |
|
| 1. Homonymy | Lat. æquivocatio; ambiguity, equivocation — S. Ambiguity |
| 2. Amphiboly | Gr. [amphibolia] — S..Ambiguity |
| 3. Combination | lat. fallacia compositionis, composition of words — S. Composition and Division |
| 4. Division of words | lat. fallacia divisionis, S. Composition and Division |
| 5. Accent | lat. fallacia accentis ; wrong accent — S. Ambiguity |
| 6. Form of expression | lat. fallacia figuræ dictionis, misleading expression — S. Expression |
This terminology may seem obscure, but its purpose is perfectly clear; it serves to establish, through a critique of language and discourse, the basic principles of a “logical grammar for argumentation”, supporting the production of reasoned texts and speeches anticipating their criticism.
The seven fallacies considered “independent from language”, are listed in the first column of the following table
|
Seven fallacies “independent of language”, RS 4 (=166b-168b) |
|
| 1. “Accident” | Lat. fallacia accidentis — S. Accident; Definition; Categorization. |
| 2. “The use of an expression absolutely or not absolutely but with some qualification of respect or place, or time, or relation” | Lat. a dicto secundum quid ad dictum simpliciter — S. Circumstances; Distinguo. |
| 3. “That which depends upon ignorance of what ‘refutation’ is” | Lat. ignoratio elenchi; misconception of refutation; evading the question — S. Question; Relevance; Resumption of speech |
| 4. “That which depends upon the consequent” |
Lat. fallacia consequentis — — S. Implication; Causality. |
| 5. “That which depends upon assuming the original conclusion” | Lat. petitio principii ; assumption of the original point; begging the question — S. Vicious Circle |
| 6. “Stating as cause what is not the cause” | Lat. non causa pro causa, non cause as cause — S. Cause – Effect |
| 7. “The making of more than one question into one” | Lat. fallacia quæstionis multiplicis, many questions; complex question — S. Many questions |
These fallacies are actually methodological mistakes.
2. Fallacies, inferences and dialectical games
In contemporary terminology, an invalid inference is often referred to as a fallacy. According to Hintikka, the Aristotelian concept of fallacy refers to something invalid, but not to an invalid inference:
The error in thinking that the traditional fallacies are faulty inferences is what I propose to dub “the fallacy of fallacies”. It is the fallacy whose recognition will, I hope, put a stop to the traditional literature on so-called fallacies. (1987, p. 211)
In other words, a fallacy cannot be simply defined as, “a fallacious argument”; just some, but not all fallacies can be “thought of as mistaken logical or conceptual inferences” (ibid.). Hintikka considers that a fallacy is essentially a move which transgresses a rule in a dialectical game, dialectical games being defined as “information-seeking questioning processes (interrogative games)” (ibid.). It is in this sense that the concept of fallacy has been taken up in the pragma-dialectical theory.
Linguistic fallacies examine the conditions a proposition must fulfill in order to qualify as a premise in a correct syllogistic inference. The fallacy of accident is the consequence of an error in the methodology of definition. Misconception of refutation reflects a poor understanding of the issues involved in the discussion and the problem. Many questions is also a forbidden move in dialectical games, where problems must be serialized to avoid implicit agreements. These different cases clearly demonstrate the non-inferential nature of fallacies, and, for the latter two, their links to rules-based dialogue games.