Dialectical and heuristic arguments: presumptions and burden of proof

In law, as in everyday conversation, presumptive reasoning is one of the most common forms of drawing conclusions from a set of premises. On Walton’s view (Walton 1996b: 13), whereas in deduction conclusions are necessarily true if the premises are true, the conclusion of a presumptive reasoning is a simple presumption, that is, it holds in conditions of incomplete knowledge and is subject to retraction should these conditions change. These arguments are grounded on generalizations such as, ‘Birds fly’ or ‘Americans love cars’, leading to conclusions of the kind: ’Plausibly, this bird flies’ or ‘Bob, who is from Michigan, probably loves cars’. However, these arguments cannot be evaluated in light of standards accepted in logic (Walton 1993: 3), and their function is not to provide that premises necessarily imply a conclusion or to provide evidence based on statistical results. Everyday conversation is characterized by incomplete knowledge, in which only few propositions can be considered necessarily true, and only a limited number of data is collected. What is the role of plausible or presumptive arguments, if not leading to truth or statistically backed generalizations? Moreover, how can they be assessed, if the logical standards do not apply? The answer can be found shifting the paradigm from the logical concept of truth of a conclusion to the pragmatic notion of dialectical effects of a conclusion. Presumptive arguments are not true or false; they produce some effects on the dialectical setting of the interaction between interlocutors. The function of presumptive arguments is to shift the burden of proof (Walton 1989: 16; Walton 1988), that is, if the interlocutor is committed to the premises, he should be committed to the conclusion as well, or he has to show why the conclusion is not acceptable. The dialectical nature of presumptive reasoning can be analyzed by inquiring into what a presumption is. We can analyze the following arguments (see Walton 1996b: 17):