François Recanati Philosophie du langage et de l'esprit Collège de France Année 2023-2024 Colloque - Indexical Dynamics : Indexical Dynamics and Composite Modes of Presentation Colloque organisé par François Recanati, Professeur du Collège de France, chaire Philosophie du langage et de l'esprit Intervenant(s) Víctor M. Verdejo, Pompeu Fabra University With roots in Frege's famous remarks (1956, 296), reflection on Rip van Winkle's fantastic story has played a key role in the philosophical study of indexical dynamics (Kaplan 1989, Perry 1997, Branquinho 2008, Ludlow 2019). Consider now the Reverse van Winkle case: Rip van Winkle falls asleep at time t and, while he feels like it's been a very long slumber, only a few seconds have actually passed when he wakes up at time t'. Suppose Rip van Winkle utters "Today is fine" both at t and t' but, while he fully accepts the associated thought at t, he hesitates at t'. The Reverse van Winkle case shows that, if we accept the 'Intuitive Criterion of Difference' (Evans 1982), a particular understanding of the relation between indexicality and thought is wrongheaded. According to this 'linguistic' view, indexicality is a property of linguistic terms only and these terms express thoughts relative to a particular context. If this view were correct, sameness of context of utterance, indexical expression and reference should guarantee sameness of thought. However, the target case shows that the same (day-based) indexical term – "today" – and the same relevant context to refer to the same day may involve conflicting rational attitudes, and hence different thoughts. The case can be raised even if one doesn't accept (contra Perry 1997: 35-38 or Ludlow 2019: 72-75) that the first "today"-thought at t is remembered at t'. One only requires that Rip, at t', doesn't change his mind with respect to the thought expressed at t (cf. Kaplan 1989: 537-538). We should not however haste to embrace the view that indexicality is an essential aspect of thought. If this were so, it should be possible for thoughts to be indexically individuated. Yet sometimes, as the (Reverse) van Winkle case illustrates, thoughts expressed with (same or different) indexicals change with contexts and sometimes they don't. What should be done? To analyse the target case, I will invoke a "composite mode of presentation", i.e. "a mode of presentation that, although 'static', i.e. deployed at a given time in thinking of the object, rests on distinct simultaneous relations to the object, and on distinct ways of gaining information (distinct information channels) based on these relations" (Recanati, forthcoming; see also Dickie & Rattan 2010, Recanati 2016). Thus, in the Reverse van Winkle scenario, at t', Rip takes recourse to two different modes of presentation (MOPs) of a particular day, one based on memory or awareness of it before falling asleep, and one based on his direct awareness of the day in question. While one would typically merge these MOPs into one composite MOP to think and reason, indexically, about a day, Rip van Winkle fails to do so because of his especial predicament. Rip van Winkle has different thoughts, based on different MOPs. But these MOPs would typically constitute one and the same composite MOP in normal circumstances. More needs to be said, however, to fully characterize the cases in which composite MOPs based on different primitive MOPs of a referent are indeed available. My proposal is that this happens when the thinker is aware of the co-referentiality of primitive MOPs. Such awareness – which can be spelled out in a number of ways – may link very different indexical and demonstrative MOPs (perceptual, testimonial, memory-based...). However, composite MOPs need not be restricted to thought expressible with indexicals or demonstratives, and may carry over to any co-referential singular and general terms. This suggests a view in which the MOPs associated with indexicals are cor