On Practical Reasoning under Ignorance

The purpose of this paper is to present an account of practical reasoning under ignorance—i.e., reasoning under conditions where the available information is so uninformative that we cannot assign probabilities to the outcomes of our options. The account shows that such reasoning need not rely on implausible principles , but can nevertheless be logically valid. Put differently, I attempt to show that we can reason in a logically correct manner even if we do not know what the outcomes of our options are or how likely these outcomes are. The proposed approach is applicable to unidimensional and multidimensionalpractical reasoning, and it is therefore useful for analysing real-life decision problems found in a wide variety of choice situations. Its application requires only that an agent has some basic knowledge of propositional logic. To achieve the aim of the article, I first outline when practical reasoning can be said to be logically valid. Section 2 applies the approach to unidimensional reasoning and Section 3 shows how an agent can build up n-dimensional reasoning under ignorance in a logically correct way.