Linked arguments and the validity requirement

In this paper I demonstrate that most textbook accounts of the linked/convergent distinction fail to conform to the widespread intuition that all valid arguments ought to be classified as linked arguments. I also show that standard textbook accounts of linkage and convergence cannot provide a satisfactory treatment of fallacies of irrelevance and, due to their general insensitivity to the epistemic context in which arguments are offered, must be supplemented by subjective accounts of linkage and convergence which appeal exclusively to authorial beliefs and intentions.Drafts of this paper were read at the Ontario Philosophical Society meeting held at Trent University in October 1990 and the Central Division meeting of the American Philosophical Association held in Chicago in April 1991. I thank Trudy Govier, Hans Hansen and an anonymous referee for helpful and encouraging comments on various drafts.