The taxonomy and analysis of fallacies in Aristotle's Sophistical Refutations pre-date the formal logic of his Prior Analytics A4-6. Of the 64 fully described examples of ?sophistical refutations? which are fallacious because they are only apparently valid, 49 have the wrong number of premisses or the wrong form of premiss or conclusion for analysis by the Prior Analytics theory of the categorical syllogism. The rest Aristotle either frames so that they do not look like categorical syllogisms or analyses in a (...) quite different way than by categorical syllogistic. (shrink)

This book brings together in one place David Hitchcock’s most significant published articles on reasoning and argument. In seven new chapters he updates his thinking in the light of subsequent scholarship. Collectively, the papers articulate a distinctive position in the philosophy of argumentation. Among other things, the author:• develops an account of “material consequence” that permits evaluation of inferences without problematic postulation of unstated premises.• updates his recursive definition of argument that accommodates chaining and embedding of arguments and (...) allows any type of illocutionary act to be a conclusion. • advances a general theory of relevance.• provides comprehensive frameworks for evaluating inferences in reasoning by analogy, means-end reasoning, and appeals to considerations or criteria.• argues that none of the forms of arguing ad hominem is a fallacy.• describes proven methods of teaching critical thinking effectively. (shrink)

In many actual arguments, the conclusion seems intuitively to follow from the premisses, even though we cannot show that it follows logically. The traditional approach to evaluating such arguments is to suppose that they have an unstated premiss whose explicit addition will produce an argument where the conclusion does follow logically. But there are good reasons for doubting that people so frequently leave the premisses of their arguments unstated. The inclination to suppose that they do stems from the belief that (...) the only way in which an argument's conclusion can follow definitely from its premisses is to follow logically. I argue that this belief is mistaken. I propose a revision of the current generic conception of logical consequence, and its variant specifications, to avoid the paradoxes of strict implication. The revised conception can then be naturally extended to include also what we might call 'enthymematic consequence'. This concept is a kind of consequence, whose properties merit investigation. (shrink)

In The Uses of Argument, Stephen Toulmin proposed a model for the layout of arguments: claim, data, warrant, qualifier, rebuttal, backing. Since then, Toulmin’s model has been appropriated, adapted and extended by researchers in speech communications, philosophy and artificial intelligence. This book assembles the best contemporary reflection in these fields, extending or challenging Toulmin’s ideas in ways that make fresh contributions to the theory of analysing and evaluating arguments.

I propose some changes to the conceptions of argument and of argumentative discussion in Ralph Johnson's Manifest Rationality (2000). An argument is a discourse whose author seeks to persuade an audience to accept a thesis by producing reasons in support of it and discharging his dialectical obligations. An argumentative discussion (what Johnson calls ‘argumentation’) is a sociocultural activity of constructing, presenting, interpreting, criticizing, and revising arguments for the purpose of reaching a shared rationally supported position on some issue. Johnson's theory (...) of argumentative discussion, with occasional modifications, is derived from this definition as a sequence of 17 theorems. Argumentative discussion is a valuable cultural practice; it is the most secure route to correct views and wise policies. (shrink)

A conclusion follows from given premisses if and only if an acceptable counterfactual-supporting covering generalization of the argument rules out, either definitively or with some modal qualification, simultaneous acceptability of the premisses and non-accepta-bility of the conclusion, even though it does not rule out acceptability of the premisses and does not require acceptability of the conclusion independently of the premisses. Hence the reiterative associated conditional of an argument is true if and only it has such a covering generalization, and a (...) supposed unexpressed premiss supplied to make an argument formally valid should be a covering generalization. (shrink)

Relevance is a triadic relation between an item, an outcome or goal, and a situation. Causal relevance consists in an item's ability to help produce an outcome in a situation. Epistemic relevance, a distinct concept, consists in the ability of a piece of information (or a speech act communicating or requesting a piece of information) to help achieve an epistemic goal in a situation. It has this ability when it can be ineliminably combined with other at least potentially accurate information (...) to achieve the goal. The relevance of a conversational contribution, premiss relevance and conclusion relevance are species of epistemic relevance thus defined. The conception of premiss relevance which results provides a basis for determining when the various ‘arguments ad’ called fallacies of relevance are indeed irrelevant. In particular, an ad verecundiam appeal is irrelevant if the authority cited lacks expertise in a cognitive domain to which the conclusion belongs, the authority does not exercise its expertise in coming to endorse the conclusion, or the conclusion does not belong to a cognitive domain; otherwise the ad verecundiam is relevant. (shrink)

Some solo verbal reasoning serves the function of arriving at a correct answer to a question from information at the reasoner’s disposal. Such reasoning is good if and only if its grounds are justified and adequate, its warrant is justified, and the reasoner is justified in assuming that no defeaters apply. I distinguish seven sources of justified grounds and state the conditions under which each source is trustworthy. Adequate grounds include all good relevant information practically obtainable by the reasoner. The (...) claim must follow from the grounds in accordance with a justified general warrant. If this warrant is not universal, the reasoner must be justified in assuming that no exception-making circumstances hold in the particular case to which it is applied. (shrink)

Wellman’s “conduction” and Govier’s “conductive arguments” are best described as appeals to considerations. The considerations cited are features of a subject of interest, and the conclusion is the attribution to it of a supervenient status like a classification, an evaluation, a prescription or an interpretation. The conclusion may follow either conclusively or non-conclusively or not at all. Weighing the pros and cons is only one way of judging whether the conclusion follows. Further, the move from in-formation about the subject’s cited (...) features to the attribution of a supervenient status is often but one moment in a more complex process. (shrink)

Informal logic is a new sub-discipline of philosophy, roughly definable as the philosophy of argument. Contributors have challenged the traditional concept of an argument as a premiss-conclusion complex, in favour of speech-act, functional and dialogical conceptions; they have identified as additional components warrants, modal qualifiers, rebuttals, and a dialectical tier. They have objected that "soundness" is neither necessary nor sufficient for a good argument. Alternative proposals include acceptability, relevance and sufficiency of the premisses; conformity to a valid argument schema; conformity (...) to rules for discussion aimed at rational resolution of a dispute. Informal logic is a significant part of philosophy. (shrink)

I argue, contrary to a recent assertion by Lilian Bermejo-Luque, that the inference-claim in an argument of the form ‘p, so q’ is not its associated material conditional ‘if p then q’. Rather, it is the claim that the argument has a covering generalization that is non-trivially true. I defend this interpretation against three objections by Bermejo-Luque.

I describe and evaluate Harald Wohlrapp’s proposal in The Concept of Argument that we should see reasonable argumentation as guided by the “principle of transsubjectivity... that, beginning with my subjectivity, I put my actual ego up for consideration as well as heighten and transcend it by seeking to participate in a general human potential, which is only attainable by recognizing the subjectivity of the Other”, and thus as having a quasi-religious meaning.

The epistemologist John Pollock has implemented computationally an architecture for a rational agent which he calls OSCAR. OSCAR models both practical and theoretical (or epistemic) reasoning. I argue that Pollock's model of practical reasoning, which has seven components, is superior not only to the two-component belief-desire model stemming from Aristotle, but also to the three-component belief-desire-intention model developed especially by Michael Bratman. Despite its advantages, Pollock's model of practical reasoning is incomplete in at least three respects: it is solipsistic, it (...) is egoistic and it is unsocial. (shrink)

278 non-freshman university students taking a l2-week critical thinking course in a large single-section class, with computer-assisted guided practice as a replacement for small-group discussion, and all testing in machine-scored multiple-choice format, improved their critical thinking skills, as measured by the California Critical Thinking Skills Test, by half a standard deviation, a moderate improvement. The improvement was more than that reported with a traditional format without computer-assisted instruction, but less than that reported with a format using both computer-assisted instruction and (...) essay-type assignments. Further studies are needed to test hypotheses suggested by these results. (shrink)

This book proposes guidelines for constructing and evaluating definitions of terms, i.e. words or phrases of general application. The guidelines extend to adoption of nomenclature. The book is meant to be a practical guide for people who find themselves in their daily lives or their employment producing or evaluating definitions of terms. It can be consulted rather than being read through. The book’s theoretical framework is a distinction, due to Robert H. Ennis, of three dimensions of definitions: the act of (...) the definer, the content of the definition, and its form. The act of a definer is what the definer does in defining a term; the book distinguishes, following Ennis, three basic acts of defining: reporting, stipulating, and advocating. The content of a definition is in one sense the information that the definition conveys and in another sense the words in its defining part. The form of a definition is the way it is expressed, for example as a definition by genus and differentia. (shrink)

Freeman’s syntactic criterion for linked argument structure is often readily applicable, captures intuitively linked structures, and implies that refuting a single premiss of a linked argument suffices to refute the argument. But one cannot sharply separate analysis from inference evaluation in applying it, whether an argument satisfies it can be uncertain, it under-generates cases where refuting one premiss suffices to refute an argument, some arguments satisfying it can be easily rescued if a single premiss is refuted, and Freeman’s underlying account (...) of probative relevance is dubious. (shrink)

37 arguments were selected by random sampling methods from calls to radio and television phone-in programs. I discuss whether my general theory of inference evaluation applies to them and how frequently they exemplify a recognized argument scheme. I also compare their dependence on context, their complexity and their quality to those features of a previously studied sample of 50 scholarly arguments.

A conclusion is a “material consequence” of reasons if it follows necessarily from them in accordance with a valid form of argument with content. The corresponding universal generalization of the argument’s associated conditional must be true, must be a covering generalization, and must be true of counter-factual instances. But it need not be law-like. Pearl’s structural model semantics is easier to apply to such counter-factual instances than Lewis’s closest-worlds semantics, and gives intuitively correct results.

In his 1933 monograph on the concept of truth, Alfred Tarski claimed that his definition of truth satisfied “the usual conditions of methodological correctness”, which in a 1935 article he identified as consistency and back-translatability. Following the rules of defining for an axiomatized theory was supposed to ensure satisfaction of the two conditions. But Tarski neither explained the two conditions nor supplied rules of defining for any axiomatized theory. We can make explicit what Tarski understood by consistency and back-translatability, with (...) the help of (1) an account by Ajdukiewicz (1936) of the criteria underlying the practice of articulating rules of defining for axiomatized theories and (2) a critique by Frege (1903) of definitions that conjure an object into existence as that which satisfies a specified condition without first proving that exactly one object does so. I show that satisfaction of the conditions of consistency and back-translatability as thus explained is guaranteed by the rules of defining articulated by Leśniewski (1931) for an axiomatized system of propositional logic. I then construct analogous rules of defining for the theory within which Tarski developed his definition of truth. Tarski’s 32 definitions in this theory occasionally violate these rules, but the violations are easily repaired. I argue that the Leśniewski-Ajdukiewicz theory of formal correctness of definitions within which Tarski worked is superior in some respects to the widely accepted analogous theory articulated by Suppes (1957). (shrink)

In Giving Reasons, Bermejo-Luque rightly claims that a normative model of the speech act of argumentation is more defensible if it rests on an internal aim that is constitutive of the act of arguing than if it rests, as she claims existing normative models do, on an aim that one need not pursue when one argues. She rightly identifies arguing with trying to justify something. But it is not so clear that she has correctly identified the internal aim of arguing (...) as showing that a target-claim is correct on the basis that a reason offered in its support is correct. First, if arguing is as she claims an attempt to justify, it is best construed as an attempt to justify the action or emotion expressed in its conclusion. Second, it is doubtful that qualified reasons and conclusions can always be reasonably reconstructed as unqualified claims, and even more doubtful that non-constative reasons and conclusions can always be reasonably reconstructed as indirect claims. Third, she needs to explain and justify her introduction of the concepts of showing and correctness in her analysis of the act of arguing. (shrink)

I resolve an apparently unresolved dispute about how probable uniform experience makes an extrapolation from it, and draw some general lessons about such enumerative induction. Uniform experience does not necessarily confer a high probability on an extrapolation of or generalization from that experience. Rational extrapolation or generalization typically involves a lot of specific background information, though not necessarily a general assumption that nature is uniform or that the future will resemble the past. And new evidence which is highly likely on (...) one hypothesis but highly unlikely on any of its competitors does not necessarily make the former hypothesis highly probable. (shrink)

In his Fallacies, Hamblin (1970) castigated what he called the “standard treatment” of fallacies in introductory textbooks of his day as debased, worn-out, dogmatic, and unconnected to anything else in modern logic. A bit more than 50 years later, I investigate the treatment of fallacies in six English-language introductory textbooks with a section on fallacies that have gone into 10 or more editions, to see whether their treatment of fallacies has taken account of the scholarship on fallacies that Hamblin’s book (...) evoked and is better than the treatment that Hamblin described. The answer is: not much. I conclude by setting out criteria for an adequate treatment of fallacies in an introductory textbook. (shrink)

En Giving Reasons, Bermejo-Luque sostiene correctamente que un modelo normativo del acto de habla de la argumentación es más defendible si se basa en un objetivo interno constitutivo del acto de argumentar, que si se apoya en objetivos que que uno no necesita perseguir cuando argumenta, tal como, en su opinión, sucede con otros modelos normativos. Ella identifica correctamente argumentar con intentar justificar. Pero no está claro que haya acertado al identificar como objetivo interno de la argumentación el mostrar que (...) una afirmación de referencia es correcta sobre la base de que las razones ofrecidas en su apoyo son correctas. En primer lugar, si argumentar, tal como ella defiende, es un intento de justificar, la mejor manera de interpretarlo es como intento de justificar la acción o emoción expresada en la conclusión. En segundo lugar, es dudoso que razones y conclusiones cualificadas puedan siempre ser reconstruidas razonablemente como afirmaciones no cualificadas, y más dudoso aún que razones y conclusiones no-constatativas puedan siempre ser reconstruidas razonablemente como afirmaciones indirectas. Más dudos aún es que podamos reconstruir razones y conclusiones no-constatativas como tesis indirectas. En tercer lugar, necesita explicar y justificar su introducción de los conceptos de mostrar y corrección en su análisis del acto de argumentar. (shrink)

People sometimes argue for questions, as can be verified by a Web search using as search terms phrases consisting of a conclusion indicator and an interrogative particle. These arguments provide a reason for asking the question and thus try to establish that it needs to be answered. Typically, they do so either in order to motivate interest in discovering the answer or in order to challenge addressees or a third party to explain their behaviour. The “inferential erotetic logic” of the (...) Polish logician Andrzej Wiśniewski provides a basis for evaluating the inferences in such arguments. For an inference from one or more statements to a question to be valid, the statements and the context must entail that the question has a true answer without entailing that any particular answer is true. Further, there must be a point to asking the question, such as the addressees’ ignorance of the correct answer to it; this requirement may be a pragmatic rather than a semantic constraint. Wiśniewski’s logic also covers inferences from questions to questions, but humans rarely articulate such inferences; this part of his logic is applicable to problem solving and proof theory rather than to arguments for questions. (shrink)