The Erotetic Theory of Reasoning: Bridges Between Formal Semantics and the Psychology of Deductive Inference*
Year: 2013
Type: article
Abstract: The intuition behind Part I of the erotetic principle is that the process of interpreting premises largely reduces to the search for answers to questions posed by other premises, regardless of whether those premises superficially look like questions.This approach will be grounded by a fresh look at the linguistic meaning of premise statements.Informally for now, a reasoner will take a disjunctive premise like "John and Bill are in the garden, or else Mary is" to pose the question of which of the disjuncts is the case.If she then accepts as a second premise that "John is in the garden," she will interpret it to be as strong an answer as possible to the question in context.As luck would have it, "John is in the garden" is part of the first possible answer to the question at hand, and not the second, so the reasoner will conclude that the question in context has been answered: "John and Bill are in the garden," deriving a well-known illusory inference (Walsh and Johnson-Laird, 2004).The inference is fallacious in that we are not entitled to make it given the information provided by the premise statements, on a linguistically plausible account of the meaning of those premise statements, an issue to which we shall return.The foregoing example is just one of many intricate data points on reasoning that need to be captured.As noted, we are not only interested in systematically capturing divergences in naïve reasoning from what we are entitled to conclude from given premises; we also want to account for the fact that our reasoning endowment makes correct reasoning possible, as evidenced by the simple fact that it is possible to train oneself to reason in a manner that respects classical validity.This is where Part II of the erotetic principle comes in: although our natural capacity for reasoning is not specifically aiming at classical validity, there exists a systematic strategy of using questions, made available by our natural capacities, which would guarantee that we only make inferences we are in fact entitled to make.Toward the end of the paper, we prove that this reasoning strategy exists.In a certain specific sense we will develop below, questions make us rational.By the erotetic principle, our desire for answers makes our reasoning fallible, but our ability to ask the right questions puts us back on track. The scope of this paperIn this paper, we focus on reasoning with premises involving propositional connectives, such as expressed by the English words 'or', 'and', 'not', and certain interpretations of 'if'.Within the domain of reasoning we have picked out, we address both the problem of failure and the problem of success.In other words, we seek to explain within a single system both how naïve reasoning diverges from what we are entitled to conclude and how it is possible for us to come to reason correctly.For our purposes, solving the problem of success means that we have to show that there exists a reasoning strategy using our naïve reasoning capacities that is classically sound, and, under performance idealizations, classically complete.We want to explain part of the puzzle of how science and philosophy are possible for humans.Since science
Source: Philosophical Perspectives
DOI: 10.1111/phpe.12029
Authors: Philipp Koralus, Salvador Mascarenhas
Institutions: University of Oxford, New York University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1487384579