Reasoning.

Abstract

Strict theories of reasoning are schoolmarmish in their insistence on rules and structure, but this gives them an advantage when inference is relatively well behaved. In the case of reasoning with deductively valid arguments, Strict theories give a convincing account of the universality of certain inference forms and the productivity of reasoning in comprehension and production. However, Strict views are rather frail, since they have to appeal to nonreasoning processes (memory limitations, comprehension failure, conversational factors) when inferencing breaks down. They seem less suited to inductive and analogical arguments, though they may be helpful in restricted situations where the inference is routine or the domain well understood. By contrast, Loose theories are inarticulate and nerdy. They dispense with formal rules in favor of continuous functions defined over beliefs, and the inferences they describe are modulations of these functions. In some sense, they are more robust than Strict theories, since they apply not only to inductively strong arguments but also to deductively valid ones as a limiting case. In fact, we have seen that they can provide insight into subjects' responses to purportedly deductive problems that are too unruly for Strict theories to handle. They are also extremely literal-minded in refusing to recognize permanent generalizations. Their only generalities are temporary products of their updating schemes. Because of these features, Loose theories have trouble producing new beliefs, explaining or justifying their own inferences, and keeping straight the difference between correlational and causal evidence. As I've described them here, Strict and Loose views are postures, not scientific theories. It's hard to see how either point of view could be exclusively true, but equally hard to combine their insights successfully.