The straw man fallacy

In this paper, an analysis is given of the straw man fallacy as a misrepresentation of someone's commitments in order to refute that person's argument. With this analysis a distinction can be made between straw man and other closely related fallacies such as ad hominem, secundum quid and ad verecundiam. When alleged cases of the straw man fallacy are evaluated, the speaker's commitment should be conceived normatively in relation to the type of conversation the speaker was supposed to be engaged in. The straw man fallacy appears to be a modern addition to the list of traditional informal fallacies covered in the logic textbooks. No mention of this fallacy as a distinct type of fallacy in the standard treatment, or as a historical item, is made by Hamblin (1970). The first inclusion of it we can find in a textbook as an informal fallacy is in Chase (1956: 40). Aristotle did not include the straw man fallacy in his list of sophistical refutations, although he does indicate, in several passages, an awareness of something very close to it. Evans (1977: 81) mentions that in Aristotelian dialectical refutation, where the dialectician refutes another party's views by deducing adoxa (implausible propositions, generally held to be false) from them: "Aristotle requires of the serious dialectician ... fidelity [according with the real or expressed views of the other party] in representing the views of others ..." Aristotle indicates in several places (Topics 105 b 6; On Sophistical Refutations 174 b 21) how this principle of fidelity for genuine refutation could be exploited in sophistical refutation, by only giving the appearance of the real view of the other party as the basis for your refutation.1 This comes fairly close to a recognition of what would nowadays be called the straw man fallacy. 1 In Topics (105 b 6), Aristotle writes of a useful method for forming propositions to refute an opponent: `choosing not only opinions actually received but also opinions which resemble these ..." In On Sophistical Refutations (174 b 21), Aristotle writes of the tactic of looking for contradictions between "the answerer's views and either his own statements or the views of those whose words and actions he admits to be right ..." This tactic sounds more like what we would call a form of the circumstantial ad hominem attack (see section 5, below). But it also has elements of awareness of the straw man tactic, as well. Further (174 b 34), Aristotle suggests, "One should also sometimes attack points other than the one mentioned, excluding it if one can make no attack on the position laid down..." This tactic might nowadays be classified under ignoratio elenchi (wrong conclusion), or it could also be a reference to the straw man fallacy. 116 The straw man fallacy DeMorgan (1847: 281) also indicated an awareness of the kinds of faulty inferences associated with misrepresenting another party's views in argumentation. But he, like Aristotle, did not use the term `straw man fallacy,' or some comparable expression, to classify a single category of error of this type. Hence the historical question of how straw man first entered the logic curriculum, as a distinctive fallacy, remains open. But as shown in this paper below, it is now in (at least a few) leading textbooks, and is definitely a very important fallacy in its own right, in the logic curriculum. In this paper, the goal is to give a practically useful analysis of the straw man fallacy that can be applied to real cases in everyday argumentation, and a theoretically clear and exact enough analysis that is adequate to distinguishing between straw man and several closely related neighboring fallacies. 1. Initial account of the fallacy Johnson and Blair (1983: 71) define the straw man fallacy as committed "... when you misrepresent your opponent's position, attribute to that person a point of view with a set-up implausibility that you can easily demolish, then proceed to argue against the set-up version as though it were your opponent's." They cite the following three conditions, for a pair of arguers M and N, and a pair of positions, Q and R: (1) M attributes to N the view or position, Q; (2) N's position is not Q, but a different one, R; and (3) M criticizes Q as though it were the view or position actually held by N. According to their analysis, the straw man fallacy can be defined, in general, by the meeting of these three characteristic conditions (1983: 74). The framework here, as Johnson and Blair put it (1983: 70), is one of an adversary context where two participants in dialogue, M and N, are arguing with each other. That is, one is attacking the other (has the aim of refuting or criticizing the other), and each is trying to defend his or her own position from the attacks of the other. This is a very clear account of the logical structure of the straw man fallacy. But how does one define the variable Q, representing the arguer's position? The way advocated in this paper is to define it as the total commitment set of a participant in a dialogue. This way of defining an arguer's position utilizes the device of a commitment set (Hamblin 1970: 264), a set of propositions listed, e.g. on a sheet of paper, or in a computer data base, representing what an arguer in a dialogue has committed herself to, as a result of moves (like asking questions, or making assertions) she has made during the course of that dialogue. But even if we can define `position' normatively and abstractly, in general, it is another question to determine what it amounts to in a specific case. According to Govier (1992: 157), the straw man fallacy is committed "when a person misrepresents an argument, theory, or claim, and then, on the basis of that misrepresentation, claims to have refuted the position that he has misinterpreted." Govier brings out some of the main practical difficulties in dealing with the problem posed by the straw man fallacy