Anti-Exceptionalism About Logic

Logic is the study of the standards of good argumentation. Ideally these are standards for reasoning in all situations, from chess to mathematical proofs, from economic models to political discourse. However, there is disagreement about precisely which standards apply. Traditionally, logic has been described as a foundational discipline. Logic has been thought to form the foundation for all other knowledge, be it in mathematics or the natural sciences. In the history of philosophy, from Kant to Frege, logical knowledge has been given a special status. Since logical knowledge cannot be the result of empirical justification, it must be a priori (i.e. independent of sensory experience), foundational, and self-evident. Briefly: Logical knowledge is exceptional. This exceptionalist thesis has met with some resistance in the more recent literature. The development of rival logical theories in the 20th century has put pressure on the idea that logical truths are self-evident. Criticism of special cognitive powers that could ground the a priori status of logic has become more common, not least because of research in experimental psychology. A common reaction is that logical knowledge must be understood in connection with other sciences, not as an immutable foundation for all knowledge. Anti-exceptionalism is precisely an attempt at developing a theory of logical knowledge and theory-choice that discard the privileged status logic has had. According to this method, logical theories cannot be established prior to other scientific theories, but must instead be developed as a response to practical challenges in the sciences. It follows that in selecting a logical theory we should use the same methods as for non-logical theories. As a result, the traditional view that that logic is an innate rational capacity is challenged. But even if this tendency is not new, little has been said about precisely how we acquire logical knowledge on anti-exceptionalist premises. One type of theory choice of particular interest is abduction, i.e. inference to the best explanation. The project has developed an account of how abductive arguments for logical theories work. This involves a formulation and weighting of selection criteria for comparing logical theories, e.g. fit with the evidence, explanatory power, unification with other theories, simplicity, etc. Such criteria derive from philosophical debates where rival logical theories are introduced (e.g. theories of truth and vague expressions). The criteria must be applicable to determine which theories give the best explanation of various logical phenomena (such as validity and consistency). One of the project?s main points is that when an abductive method is applied to logic, it is far from clear that classical logic - and theories built on classical logic - are the most satisfactory. Rather, the method opens for various forms of logical pluralism. Following the development of abductive selection criteria, the project applies the method to a number of concrete logical theories. Debates about controversial argument typically involve rival theories that claim to best accommodate the evidence. The project studies how selection criteria can be applied in the debate about the semantic paradoxes in order to rank theories. It also investigates how satisfying different selection criteria (e.g. strength and simplicity) can be in conflict, and if there is a consensus for the application of any criteria (e.g. the theories ability to recapture mathematical proofs). Scientific theories typically explain a phenomenon and offer predictions that can be tested. Both explanatory and predictive power are two central components when we choose between rival theories. If theory-choice in logic works similarly to that of the sciences, it is natural to expect similar features. But what is explained by a logical theory, and does a logical theory offer predictions? The project develops an account of non-causal explanation for theories of validity, and shows how successful and unsuccessful predictions inform theory-choice. Theories of validity contain hypotheses about validity in the form of generalizations about truth-preserving arguments, and these generalizations are in turn underwritten by law-like statements in the model- and proof-theory.

Prosjektets resultater danner fundamentet for å diskutere utvelgelse av logiske teorier på forskningsfeltet. Sammenstilling og vurdering av rivaliserende teorier kan foregå i systematiske rammer, og prosjektet peker på hvordan man kan utvikle testbare prediksjoner på bakgrunn av logiske teorier. Prosjektet åpner for en ny forståelse av hva som utgjør progresjon i logikkfaget, for eksempel overgangen fra syllogistiske til klassiske teorier.

Logic isn't special. Its theories are continuous with science; its method continuous with scientific method. Logic isn't a priori, nor are its truths analytic truths. Logical theories are revisable, and if they are revised, they are revised on the same grounds as scientific theories. These are the tenets of anti-exceptionalism about logic. The position was first suggested by W. V. Quine, and has more recently been defended by Timothy Williamson. Their efforts have mainly been to argue against exceptionalism, in particular that logic is a priori. In contrast, this project aims to develop a positive programme for anti-exceptionalism. This involves providing a non-apriori epistemology for logic that is continuous with other sciences. We work on the assumption, therefore, that theory selection in logic will be a matter of abduction, i.e. inference to the best explanation. There are four key challenges to anti-exceptionalism that the project will answer. First, we will articulate precise abductive criteria for theory selection. Second, we will classify and assess logical theories based on the abductive strategy. In particular, we argue that the abductive strategy favours nonclassical logic. Third, we will identify empirical presuppositions of exceptionalism and anti-exceptionalism, in order to see whether they are confirmed by the findings of experimental data in the psychology of reasoning. The hypothesis is that the aprioricity thesis is unsupported by the data. Fourth, we will argue that anti-exceptionalism is compatible with logic being normative.