Some norms of reasoning tell us certain inferences are forbidden β for example, Bumbling Bruce goes wrong in reasoning to the conclusion that π when his sole reasons are that he accepts π β π and π. Some tell us that certain inferences are mandatory β for example, Stunned Sharon goes wrong when she remains in suspension about π when she is fully aware that π and π β π are and remain among her reasons, and there are no cognitive or pragmatic reasons that π might not be credible for her. (There are parallel norms that tell us that certain (non-)inferences are permissible.)
Logical Pluralism (at least the Canberra subspecies) is the view that many relations between sets of sentences and sentences satisfy the core platitudes governing βlogical consequenceβ. These platitudes are generally taken to include formality, topic-neutrality, necessary truth-preservingness, and the capacity to play a role in norms of right reasoning. Logic itself is not normative, but it plays a role in regulating the inference of those who recognise its presence β providing reasons to infer. The norms might tell us that it is (intuitionistically) impermissible to reason to π from ¬¬π, when the truth of the latter is one's only reason for accepting π, and that it is classically mandatory to to reason to π from ¬¬π, when one has the cognitive capacity to consider and accept π and one has no reason against accepting it. Given that one needs to regulate one's doxastic behaviour, some way of aggregating these norms seems both mandatory and impossible.
In this talk, I will elaborate on this and related puzzles, and work through some (pluralist and non-pluralist) options in response.
Monday July 6, 2026 4:30pm - 5:25pm AEST Steele-2373 Staff House Rd, St Lucia QLD 4067, Australia