What is really possible?

With this second workshop, we want to tackle logical and philosophical aspects of the notion of real possibility that is central to our research project.

The workshop

What is really possible 2 (WIRP-2): 

Logical and philosophical aspects of real possibility

will take place in Konstanz, Germany, on Friday/Saturday, 21/22 September 2012.

This is a satellite workshop following the conference GAP.8 of the German Gesellschaft für Analytische Philosophie (Society for Analytic Philosophy), which takes place in Konstanz from Monday 17 to Thursday 20 September 2012. See the conference website for more information about the venue. The full program of  GAP.8 is available here in pdf format.

Attendance of the workshop is free, but limited due to space restrictions; if you wish to attend, please register via email to possibilities@phil.uu.nl.

Workshop description

We live in a world of possibilities. Much of our practical life – planning, deciding, hoping and fearing – only makes sense before a background of options to choose from and possibilities for what the future will bring: real possibilities in concrete situations. Such real possibilities are dynamical: they vanish when they are not actualized. Real possibilities are commonly represented within branching frameworks such as provided by the Prior-Thomason account of Branching Time and Belnap’s theory of Branching Space-Times. In those branching frameworks our world is pictured as a tree of histories branching off from a single past course of events into multiple possible futures. Work on real possibility has so far been mainly technical, e.g., in the formal-logical study of the semantics of the future tense and the problem of future contingents.
In this workshop we want to combine logical and philosophical aspects of real possibility. Important logical questions concern the interaction of time, modality and quantification. Connected to these logical issues there are questions of interpretation: How should a branching structure be understood in the first place? Are all branches equally real? What is the difference between branching and divergence? How can individuals be situated within a branching picture? Exploring the relation of real possibility to other notions of possibility is another desideratum. Metaphysical and physical possibilities are prominent in accounts of causality and laws of nature. The role of real possibilities in metaphysics and in the sciences still needs to be assessed.

Speakers

• Michael De (Utrecht University, now Konstanz)
• Christopher Hitchcock (Caltech)
• Alex Malpass (Bristol University)
• Thomas Müller (Utrecht University, now Konstanz)
• Tomasz Placek (Jagiellonian University, Kraków)
• Antje Rumberg (Utrecht University, now Konstanz)
• Niko Strobach (University of Münster)
• Stephan Torre (University of Barcelona)
• Sara Uckelman (Tilburg University, now Durham)
• Jacek Wawer (Jagiellonian University, Kraków)
• Leszek Wroński (Jagiellonian University, Kraków)
• Christian Wüthrich (University of California at San Diego, now Geneva)

Program

The workshop will start on both days at 10.00; coffee will be available from 9:30. Talks are 45 minutes, plus 15 minutes time for discussion. On the second day, the workshop will end with a longish plenary discussion.

Abstracts

Michael De: Partial endurantism

I present a theory of partial endurantism couched in a Lewisian framework according to which moments are mereological sums of suitably related individuals and endurance is had by partial overlap. The theory is defended from various objections to endurantism including the argument from temporary intrinsics which purports to put perdurantism in favourable light. A partial endurer can undergo intrinsic change when intrinsic properties are understood as being had simpliciter rather than only relative to times. As such, the problem of temporary intrinsics favours neither perdurantism nor endurantism.

Partial endurantism gives rise to a natural view of branching within worlds without requiring worlds themselves to overlap. Branching occurs just when two distinct moments partially overlap the same moment. I consider some advantages this view has over other views of branching worlds. Moreover, extending partial endurantism to the modal analog of partial overlap of worlds results in a theory of de re modality that is, I argue, susceptible to neither the problem of accidental intrinsics nor the Humphrey ob jection (nor another little-known problem concerning the duplication account of intrinsicality). I conclude that partial overlap provides a unified solution to a host of temporal and modal problems.

Christopher Hitchcock: The relative a priori and the space of possibilities

In both his historical work on the logical empiricists, and in his philosophy of science, Michael Friedman has developed the idea that a scientific theory delineates a space of possible ways in which it could be false. For example, the framework of Newtonian mechanics allows one to coherently formulate the claim that gravity obeys an inverse cube law, even though the theory itself tells us that this is false. By contrast, the theoretical framework does not even admit the possibility of non-Euclidean space or relativistic spacetimes. One interesting manifestation of this idea is in recent debates over the status of velocity in classical physics.

It is unclear just what the metaphysical status of this kind of theory-relative, non-actual, possibility is. It does not look like ‘real possibility’ in the sense intended in the title of this workshop. On the other hand, I will suggest that Friedman’s idea also finds a place in mini-theories or localized models. For example, structural causal models describe a specific causal structure, but also indicate possible ways in which the structure could be disrupted. These come closer to being ‘real possibilities’. Thinking carefully about what kinds of disruption are possible helps us to address issues concerning the metaphysics of causation and events.

Alex Malpass & Jacek Wawer: A future for the thin red line

The Thin Red Line (TRL) is a theory about the semantics of future-contingents. The central idea is that there is such a thing as the ‘actual future’, even in the presence of indeterminism. It is inspired by a famous solution to the problem of divine foreknowledge associated with William of Ockham. In the modern branching-time setting, the theory of the TRL is widely regarded to suffer from several fundamental problems. In response to one of the arguments formulated by Nuel Belnap and his associates we propose our semantics, which is a cross between the TRL and supervaluationism. We discuss the notions of truth, validity and consequence which result from our final semantics, and demonstrate some of its pleasing results. We focus in particular on objections against supervaluationism raised by Williamson, Tweedale and Fara and investigate if and how they affect our account.

The talk is based on our same titled paper which is to appear in the forthcoming issue of Synthese and which is available online via Springer’s Open Access.

Thomas Müller: Case-intensional branching time: A substantial logic

Modal logic is a powerful tool for formalizing and clarifying conceptual assumptions about modality, and research in modal logic is flourishing. Most recent developments in modal logic are concerned with propositional modal logic. Our practical interests in modality are however concerned with possibilities and necessities for individual things: Is this powder soluble? Could this dog bite? Can I still catch the train? Modal propositions about individual things need to be formalized in a modal predicate logic, or quantified modal logic.

In this paper we will first introduce recent work, conducted jointly with Nuel Belnap, on a specific system of modal predicate logic that originated with Bressan (1972). We start with the flexible notion of possible cases rather than possible worlds, and we apply Carnap’s method of extension and intension for terms and sentences uniformly, to arrive at the system of case-intensional first order logic, CIFOL. The main novel features of CIFOL are that individuals are represented not as extensions, but as intensions, and that principles for tracing individuals across cases can be represented within the system. This provides a foothold for the idea that we should represent an individual substance as an intension falling under a natural trancing principle.

In the second part of the paper, we will describe the application of CIFOL to branching time. In branching time, cases are moment-history pairs. The representation of individuals in this framework poses some challenging interpretational questions. We will discuss these questions and argue that case-intensional branching time is a promising candidate for analyzing the modal and temporal aspects of the notion of an individual substance.

Tomasz Placek: On agents’ choices

However mathematically beautiful the stit framework of Belnap, Perloff, Xu, von Kutschera (and others) is, it fails short of providing an adequate representation of human choices. First, since this theory is based on a possible histories semantics (branching time or branching space-times), it depicts an agent as navigating through  a maze of pre-existing possibilities, leaving no room for the agents’ creation of possibilities. Second, since the stit framework leaves aside information-related aspects of  agency, it makes agents choices similar to dice-tossing. In contrast, agents typically report  being “forced”  to their decisions by the available data. Further, a large share of our choices are automatic, and if they are not so, they are results of deliberation, i.e., of a process of gathering and processing of information.

The aim of this talk is to improve on the mentioned shortcomings of the stit framework by accommodating some insights of the psychological two-systems theory of rationality  of Kahneman, Tversky, Stanovitch, West  (and others). The present attempt relies on three key elements: First, agents are taught to discern decision situations (e.g., decisions between menu items). Second, we have a (programmable) system  responsible for automatic choices (e.g., at a multiple-choice test, pick “C” if you don’t have a clue). Finally, there is a “programmer system” that offers a solution for  decision situations for which the automatized reaction is not (yet) available, where this solution is arrived by Bayesian updating and then passed to the first system, for automatic execution in the future.

Antje Rumberg: Real possibilities in a branching framework

In our everyday lives we constantly encounter real possibilities. These real possibilities are future-directed possibilities in concrete situations, something that is physically possible given the local circumstances. Real possibilities can be represented within branching frameworks, such as Branching Time and Branching Space-Times, which picture our world as a tree of histories branching into multiple possible futures. The branching structures representing our world can be employed in the semantics of languages containing modal and temporal operators if a valuation on the structure is provided that assigns truth values to the sentences of the language relative to some suitable index of evaluation. Each branching structure allows for several models: each combinatorially possible valuation is admitted. If we want our models to capture the notion of real possibility, some restriction is in order. Some relation between the structure and its valuation needs to be established. The space of future possibilities has to be narrowed down to such possibilities that are physically possible given the relevant local circumstances.

Branching structures are often understood as a means to reduce modality. Rather than taking the branching structure as fundamental and reducing modality to it, I will propose a dynamic modal explanation of the branching structure itself. The fact that two histories branch at a certain point will be accounted for by the potentialities of objects and their local arrangement. If the local future possibilities are grounded in the local arrangement of objects and their potentialities, the specification of the arrangement at one point in time suffices to determine the subsequent branching structure as well as the arrangement at any future point in that structure. In this way, we get a dynamic picture of real possibility: some possibilities disappear and new possibilities emerge as time progresses.

Niko Strobach & Martin Pleitz: Really possible family trees: A Belnap-style version of modal bio-logic

Structures which Belnap (1992) intended as branching space-times can be reinterpreted as structures of bio-logic if the accessibility relation is the relation “is a bio-logical ancestor of ”. In a first step, real possibilities vanish from the models. But in a second step they can be reintroduced. We will show how this can be done in a fashion which is very close to Belnap’s models and might help liberalise them on their original interpretation, too. Belnap-style modal bio-logic allows addressing such issues as the modal metaphysics of ancestry and species membership, the status of possible living beings and the limits of their possible coexistence as a task for metaphysically sensible constraints on models.

Stephan Torre: Fission and uncertainty about the future

What should our attitudes about future events be in cases involving fission and branching? Suppose I know that later today I will undergo a Parfittian fission operation where the left half of my brain is removed from my head and placed into a new healthy body. The resulting person, call him ‘Lefty’, wakes up in a blue room. The right half of my brain is removed from my head and placed in a different healthy body. The resulting person, call him ‘Righty’, wakes up in a red room. Given full knowledge of all these facts, what should my attitude be about waking up in the blue room? Does it make sense to claim that I can now be uncertain about whether I will wake up in the blue room or whether I will wake up in the red room?

According to a number of philosophers, cases of personal fission are not confined to philosophy thought-experiments, but instead are commonplace. Many philosophers think that the openness of the future is best explained in terms of branching spacetime. According to the Everett interpretation of quantum mechanics, we inhabit a branching multiverse. On a natural understanding of these branching views, when the world splits into two, we do too.

After outlining some assumptions and commitments, I will consider one attempt at accounting for uncertainty in fission cases that has been recently proposed by Simon Saunders and David Wallace. Saunders and Wallace claim that uncertainty about the future in fission cases is a form of de se uncertainty, uncertainty about who one is. I will raise two objections for the Saunders and Wallace account. In the second half, I will investigate an alternative way of how there can be uncertainty about the future in cases involving fission. The alternative proposal appeals to a framework in which the space of possibilities is given in terms of individuals rather than worlds. On the account I propose, one can have complete knowledge of who one is, yet have uncertainty about who one will be.

Sara Uckelman: Possible impossibilities in medieval logic

The idea of there being “real possibilities” implies that there must also be so-called “unreal possibilities”, for if there were no unreal possibilities then all possibilities would be real. In this talk we consider this issue from a novel perspective, by looking at the role of possible impossibilities: impossible situations or scenarios which are nevertheless in a certain sense possible, and in this particular sense—which must be made explicit—shed interesting light on questions in logic and metaphysics. Our particular focus is on a sub-type of the genre of medieval logic called obligationes, namely, the species positio impossibilis. Ordinary obligational disputations operate on the Aristotelian premise that the impossible does not follow from the possible, and hence if a disputant concedes and denies the same sentence, or concedes both a sentence and its denial, then either he has violated some disputational rule or norm or the original starting point of the disputation was impossible. But some authors writing on obligationes point out that it is possible to reason from seemingly impossible premises without falling prey to logical explosion, and they investigate this in impossible positio. We introduce the genre of obligationes and the particular subspecies of impossible positio, looking at rules for and examples of it from a number of different medieval authors, and consider how this type of disputation can be used to shed light on actual possibilities, metaphysical, theological, and logical.

Leszek Wroński: Substances in branching frameworks

There is more than one way to think of substances in branching frameworks. In my talk I explore – mainly in the context of Belnap’s theory of branching space-times – the option (present in recent papers by Th. Müller, T. Placek and N. Belnap) of viewing substances as sets of transitions as opposed to sets of point events. I also review some open problems regarding the transition structures of BST models.

Christian Wüthrich: Humean possibilities

This talk will contend that all modality is grounded in actuality, and in actuality only. ’Real’ possibility can only ever arise a posteriori out of building blocks we collect in our interactions with the actual world. The theories we thus construct span ranges of possibilities, which do not physically exist but are ’real’ to the extent to which we accept the theory on which they are based. Humans and other epistemically restricted agents do not have access to the entire ’Humean mosaic’ of occurrent facts and are thus compelled to ’theorize’ about the world they inhabit, i.e., they must hypothesize about its regularity structure. From this hypothesizing emerges the modality we ascribe to our world. Modality, on this view, is ’de theoria’ rather than either ’de re’ or ’de dicto’. The view sketched here thus rejects both modal realism, de re essentialism, and necessitarian combinatorialism, as well as the empiricist position which takes necessity to arise from stipulated semantic conventions.