John Divers

We propose a novel interpretation of Lewis on the analysis of modality that is constructed from primary sources, comprehensive and unprecedented. Our guiding precepts are to distinguish semantics from metaphysics, while respecting the inter-relations between them, and to discern whatever may be special, semantically or metaphysically, about the modal case. Following detailed presentation, we amplify and advocate our interpretation by providing a conforming genealogy of Lewis’s theory of modality and applying it to construct a detailed and newly illuminating version of the Lewisian theory of modality de re.

McFetridge (in Logical necessity and other essays . London: Blackwell, 1990 ) suggests that to treat a proposition as logically necessary—to believe a proposition logically necessary, and to manifest that belief—is a matter of preparedness to deploy that proposition as a premise in reasoning from any supposition. We consider whether a suggestion in that spirit can be generalized to cover all cases of absolute necessity, both logical and non-logical, and we conclude that it can. In Sect. 2, we explain the significance that such an account of manifestation of belief in absolute necessity has for the prospects of a non-realist theory of modality. In Sect. 3, we offer a sympathetic articulation of the detail that underlies the McFetridge conception of belief in logical necessity. In Sects. 4 and 5, we show that the conception so articulated will not generalize to encompass all cases of belief in absolute necessity and proceed to offer a remedy. Our proposal is based upon a distinction between two kinds of suppositional act: A-supposing and C-supposing (Sect. 6). In Sect. 7, we then explain and defend our central thesis: (roughly) that (manifestation of) belief in absolute necessity is a matter of preparedness to deploy as a premise in reasoning under any C-supposition. Finally, we indicate that there is some promise in the parallel thesis that manifestation of the treatment of a proposition as a priori is a matter of preparedness to deploy as a premise in reasoning under any A-supposition (Sect. 8)

At the core of the concept of supervenience are certain general maxims— notably, that there can be no A-differences without B-differences and that Bindiscernibility must bring A-discernibility. Supervenience is thus conceived as a matter of modal covariance between two sets of things in a given category, usually properties. The perennial issues surrounding supervenience concern: (a) the variety of specifically formulated theses that serve the core maxims and the patterns of entailment that obtain among these theses, and (b) the relations that obtain between supervenience theses and such salient concepts of metaphysics as reduction, dependence and determination. The regions of philosophical inquiry that gave rise to contemporary analytic philosophical interest in supervenience, and in which the notion is still most prominently applied are, famously, meta-ethics and the philosophy of mind. In the former case the history of supervenience can be traced back through Hare to Moore; in the latter case the line can be traced back from the pathfinding work of Kim through the famous resurrection of the notion of supervenience in Davidson’s work to the emergentist theorists of mind in the early part of this century.1 Publication on the topic of supervenience has exploded so that a brief treatment of recent work could provide an even remotely comprehensive survey only by degenerating into a list.2 So I shall settle for concentrating on the two groups of issues that were identified above, that is those concerning the logical character and metaphysical significance of supervenience theses, interposing some comments on the specific topic of supervenience in the philosophy of mind. I will emphasise and sympathise with that deflationary strand of recent work that invites soberand cautious re-appraisal of the role of supervenience theses in philosophical enquiry

In section 1 I motivate and execute the presentation of a well-defined Lewisian conception of analysis and of what it would be to analyse modality successfully. That conception is then put to two applications. In section 2 various inadequacies are exposed in a (recently popular) separatist approach to the understanding and/or evaluation of Lewis's analysis of modality. Section 3 provides a defence against a resilient argument for the claim that Lewis's analysis of modality cannot be fully reductive while also dealing adequately with alien possibility.

If a possible-worlds semantic theory for modal logics is pure, then the assertion of the theory, taken at face-value, can bring no commitment to the existence of a plurality of possible worlds (genuine or ersatz). But if we consider an applied theory (an application of the pure theory) in which the elements of the models are required to be possible worlds, then assertion of such a theory, taken at face-value, does appear to bring commitment to the existence of a plurality of possible worlds. Or at least that is so if the applied theory is adequate. For an applied possible-worlds semantic theory that is constrained to contain only one-world models is bound to deliver results on validity, soundness and completeness that are apt to seem disastrous. I attempt to steer a course between commitment to the existence of a plurality of possible worlds and commitment to such a disastrous applied possible-worlds semantics by noting, and developing, the position of one who asserts such a theory at face-value but who remains agnostic about the existence of other (non-actualized) possible worlds. Thus, a novel interpretation of applied possible-worlds semantics is offered on which we may lay claim to whatever benefits such a theory offers while avoiding realism about (other) possible worlds. Thereby, the contention that applied possible-worlds semantics gives us reason to be realists about possible worlds is (further) undermined.

This chapter has three principal aims. Firstly, to promote interest in the question of the function, or utility, of judgements of modality. Secondly, to endorse an alternative to orthodox contemporary methodology, advocating that we prioritize the question of function in modal philosophy. Thirdly, to consider among our modal judgements exactly which are the proper and exact source of various different kinds of substantial philosophical commitments in ontology, epistemology, and elsewhere. An illustration is offered, in the de dicto case, of how minimal a philosophical theory of modality might be if constructed properly, and carefully, according to functional methodology.