scientific article; zbMATH DE number 1163371
That a patient has a characteristic symptom is a reason to think he has the condition itself. It is a \textit{prima facie} reason, not a conclusive reason, since it is possible to have the one without the other. Still, where there is no reason to think the patient does not have the condition, it will seem that we can suppose he does. Similarly, that you have promised to meet me is a reason for you to do so. It is a \textit{prima facie} reason since it can happen that, in spite of your promise, you do not have the duty to meet me. Still, where there is no reason to think you do not have this duty, it will seem that you ought to meet me. Asked why he thinks a patient has a condition, a doctor can give his reason: the patient has the symptom. Pushed further he can point to the evidential relation: that the patient has the symptom is a reason to think he has the condition. We seem in some sense to infer the supposition from the reason. And the inference parallels \textit{modus ponens}, with a reason sentence -- in this case a sentence expressing the evidential relation between symptom and condition -- in the role of ``if \(\dots\) then \(\dots\). Here I will follow this parallel between reason sentences and ``if \(\dots\) then \(\dots\), borrowing from studies of conditionals to obtain a formal understanding of reasons. This will lead me in section 2 to paraphrase reason sentences in terms of conditionals. It will lead me in section 3 to present a model-theoretic interpretation for these conditionals. And it will lead me, in section 4, to treat inference from reasons as allowed argument, a species of defeasible inference involving conditionals. Allowed argument will be treated there in purely syntactic terms as allowed consequence; section 5 provides a model-theoretic characterization, allowed entailment. In the illustrations throughout I will continue to slip back and forth between reasons to think something and reasons to do something. Though different in their substance, from a formal perspective these two kinds of reasons are similar. For the entire collection see [Zbl 0885.00033].
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4394897)
