Daniel Cohen

According to the semantics in Wittgenstein's Tractatus, a picture and what is pictured must have the same logical form. However necessary that may be, it cannot suffice to make one fact a picture of another. The grounds for the pictorial relation, it is argued, must be found in the transcendental will. Following a suggestion by Ramsey, the semantic resources of the Tractatus are used to construct a new interpretation of propositions as equivalence classes of facts. The nature of the involvement of the will helps explain why ethics and logic are alike ineffable.

This paper explores the outlines of a framework for evaluating arguments. Among the factors to take into account are the strength of the arguers' inferences, the level of their engagement with objections raised by other interlocutors, and their effectiveness in rationally persuading their target audiences. Some connections among these can be understood only in the context of meta-argumentation and meta-rationality. The Principle of Meta-Rationality (PMR)--that reasoning rationally includes reasoning about rationality-is used to explain why it can be rational to resist dialectically satisfying arguments or accept logically flawed ones

One result of successful argumentation – able arguers presenting cogent arguments to competent audiences – is a transfer of credibility from premises to conclusions. From a purely logical perspective, neither dubious premises nor fallacious inference should lower the credibility of the target conclusion. Nevertheless, some arguments do backfire this way. Dialectical and rhetorical considerations come into play. Three inter-related conclusions emerge from a catalogue of hapless arguers and backfiring arguments. First, there are advantages to paying attention to arguers and their contexts, rather than focusing narrowly on their arguments, in order to understand what can go wrong in argumentation. Traditional fallacy identification, with its exclusive attention to faulty inferences, is inadequate to explain the full range of argumentative failures. Second, the notion of an Ideal Arguer can be defined by contrast with her less than ideal peers to serve as a useful tool in argument evaluation. And third, not all of the ways that arguers raise doubts about their conclusions are pathological. On the contrary, some ways that doubts are raised concerning our intended conclusions are an integral part of ideal argumentative practice.

A complex network of reciprocal relations connect arguments and stories. Arguments can occur in stories and stories can be parts of arguments. Further, stories can themselves be arguments. Whether a text or exchange serves as an argument partly depe nds on how we read it, i.e., on the story we tell about it and how well we argue for that story, but the circle is not as vicious as it appears. Or at least, that is the story we present and the argument we tell in this dialogue revisiting the ancient ar gument between the poets and the philosophers.

Psychological and neuroscientific data suggest that a great deal, perhaps even most, of our reasoning turns out to be rationalizing. The reasons we give for our positions are seldom either the real reasons or the effective causes of why we have those positions. We are not as rational as we like to think. A second, no less disheartening observation is that while we may be very effective when it comes to giving reasons, we are not that good at getting reasons. We are not as reasons-responsive as we like to think. Reasoning and argumentation are, on this view, charades without effect. This paper begins by identifying a range of theoretical responses to the idea that reasoning and argumentation have little casual role in our thoughts and actions, and, consequently, that humans are not the reasons-giving, reasons-responsive agents that we imagine ourselves to be. The responses fall into three categories: challenging the data and their interpretations; making peace with the loss of autonomy that is implied; and seeking ways to expand the causal footprint of reasoning and argumentation, e.g., by developing argumentative virtues. There are indeed possibilities for becoming more rational and more reasons-responsive, so the reports of our demise as the rational animal are greatly exaggerated.

As a rule, there is nothing in the words themselves to mark off metaphors from literal language. If a boundary could somehow be drawn, it would be in constant need of re‐adjustment as metaphors become entrenched, idiomatic, and finally literal, and literal phrases are put to figurative or hyperbolic, and then metaphorical uses. Further, there is no algorithmic recovery of the intended meaning of a metaphor from the meanings of its components, no function that takes literal meanings as its arguments and has metaphorical meaning as its output. That is, metaphors are syntactically invisible, pragmatically unstable, and semantically under‐determined. Their creation, use, and interpretation are almost impenetrable mysteries. As if to spite semantic theorists, they are also linguistic commonplaces, present in every kind of discourse. The discourse of philosophy is no exception to this rule, availing itself of the various stylistic, heuristic, and explanatory roles metaphors can play. Philosophy also produces metaphors of another sort, metaphors that differ in important ways from the metaphors of art and literature, as well as the metaphors of the natural and social sciences. These “grand“ metaphors are lasting monuments to the triumphs of philosophical discourse, even as they are also testimony to the failures of philosophical theorizing. This is because philosophy’s grand metaphors were never intended that way. They are better read as structural hypotheses that have become metaphors only after failing as literal truths. Ironically, it is precisely as metaphors that these images are integrated into our ways of thinking. In the end, thought itself is re‐structured, and these failures as descriptive truths succeed in becoming “constitutive truths” of a new reality.