Can ‘Intrinsic’ Be Defined Using Only Broadly Logical Notions? 1

Synthese, 2015

It is shown that counterpart theory and the duplication account of intrinsicality-two key pieces of the Lewisian package-are incompatible. In particular, the duplication account yields the result that certain intuitively extrinsic modal properties are intrinsic. Along the way I consider a potentially more general worry concerning certain existential closures of internal relations. One conclusion is that, unless the Lewisian provides an adequate alternative to the duplication account, the reductive nature of their total theory is in jeopardy. 1 The duplication account Lewis attempts to reduce intrinsicality to (at least for him) more fundamental notions. The account he provides, which I shall call the duplication account (DA for short), states that DA. A property is intrinsic iff it never differs among duplicates. Lewis gives two separate accounts of duplication, thereby yielding two separate formulations of DA. According to the first [Lewis, 1986, pp. 61-62], two things are duplicates just in case they share their perfectly natural properties, and their parts can be put into one-one correspondence so that corresponding parts share the same perfectly natural properties and stand in the same perfectly natural relations (to the things and their parts only). Call this account Perfect (since it relies on the sharing of perfectly natural properties). According to the second [Langton and Lewis, 1998], two things are duplicates just in case they share their basic intrinsic properties, where a property is basic intrinsic just in case it is neither disjunctive nor the negation of a disjunctive property, and it is independent of accompaniment, i.e. it can be: (i) had by a lonely thing; (ii) lacked by a lonely thing; (iii) had by an accompanied thing; (iv) lacked by an accompanied thing. A thing (i.e. possible individual, not transworld sum) is lonely just in case it exists without any contingent, wholly distinct worldmate and it is accompanied otherwise, and two things are wholly distinct just in case they share no common part. Finally, a property is disjunctive if it is not natural but is expressible by a disjunction (of conjunctions) of natural properties. One may prefer a characterization involving the less contentious notion of comparative naturalness, according to which a property is disjunctive if it is expressible by a disjunction (of conjunctions) of properties, each sufficiently more natural than the disjunction (where what counts as sufficient is left unclear). Call this second account Disjunctive (since it relies on disjunctiveness). A number of objections to DA apply only to Disjunctive. For instance, Dan Marshall and Josh Parsons [Marshall and Parsons, 2001] argue that the intuitively extrinsic property being such that there is a cube is, according to Disjunctive, intrinsic. In response, Langton and Lewis say:

Over the last thirty years there have been a number of attempts to analyse the distinction between intrinsic and extrinsic properties. This article discusses three leading attempts to analyse this distinction that don’t appeal to the notion of nat-uralness: the duplication analysis endorsed by G. E. Moore and David Lewis, Peter Vallentyne’s analysis in terms of contractions of possible worlds, and the analysis of Gene Witmer, William Butchard and Kelly Trogdon in terms of grounding.

Forthcoming in Philosophy and Phenomenological Research

Intrinsicality is a central notion in metaphysics that can do important work in many areas of philosophy. It is not widely appreciated, however, that there are in fact a number of different notions of intrinsicality, and that these different notions differ in what work they can do. This paper discusses what these notions are, describes how they are related to each other, and argues that each of them can be analysed in terms of a single notion of intrinsic aboutness that relates states of affairs to the things they are intrinsically about.

A sentence or statement or proposition that ascribes intrinsic properties to something is entirely about that thing … A thing has its intrinsic properties in virtue of the way that thing itself, and nothing else, is … The intrinsic properties of something depend only on that thing … If something has an intrinsic property, then so does any perfect duplicate of that thing … (Lewis 1983a, p. 197) I once offered a definition that was meant to capture the notion expressed by the intuitive descriptions in this quote from Lewis.¹ The basic idea is that F is an intrinsic property of an item x just in case x's having F consists entirely in x'sha v-ing certain internal properties, where an internal property is one whose instan-tiation does not consist in one's relation to any distinct items (items other than oneself and one's proper parts). I still think that this relational analysis is largely correct, and here I wish to provide additional support for it and defend it against some objections that have been raised. In the process I aim to make the account somewhat more precise, especially by contrasting it with a grounding approach to defining the intrinsic/extrinsic distinction.

Robert Francescotti (eds.), Companion to Intrinsic Properties: De Gruyter, 2014

Stephen Yablo has pointed out that there are several simple principles connecting intrinsicality with parthood, such as the principle that, if u is part of v, then u cannot intrinsically change without v also intrinsically changing. By exploiting these prinicples, Yablo has provided a striking account of intrinsicality that promises to analyse intrinsicality by appealing only to familiar and well understood notions such as logical, modal and mereological notions, and without appealing to contentious notions such as perfect naturalness or metaphysical grounding. His account also promises to establish that intrinsicality is not a fundamental aspect of reality, but is instead analysable in terms of more fundamental logical, modal and mereological properties, relations and operators. In this paper I argue that these promises can't be kept since Yablo's account fails. I also argue that, while Yablo's account can be modified so that it does a better job at analysing intrinsicality than many other accounts in the literature, these modifications also fail. I conclude that it is likely that any account of intrinsicality in the spirit of Yablo's account will also fail.

Forthcoming in Nous

This paper puts forward a new account of intrinsicality in terms of perfect naturalness.

There are, broadly, three sorts of account of intrinsicality: 'self-sufficiency', 'essentiality' and 'pure qualitativeness'. I argue for the last of these, and urge that we take intrinsic properties of concrete objects to be all and only those shared by actual or possible duplicates, which only differ extrinsically. This approach gains support from Francescotti's approach: defining 'intrinsic' in contradistinction to extrinsic properties which 'consist in' relations which rule out intrinsicality. I answer Weatherson's criticisms of Francescotti, but, to answer criticisms of my own, I amend his account, proposing that possession of an extrinsic property consists in a relation to one or more actual or possible distinct concrete objects. Finally I indicate ways to avoid some apparent objections to this account.

Forthcoming in Philosophy and Phenomenological Research

A number of philosophers have recently claimed that intrinsicality can be analysed in terms of the metaphysical notion of grounding. Since grounding is a hyperintensional notion, accounts of intrinsicality in terms of grounding, unlike most other accounts, promise to be able to discriminate between necessarily coextensive properties that differ in whether they are intrinsic. They therefore promise to be compatible with popular metaphysical theories that posit necessary entities and necessary connections between wholly distinct entities, on which it is plausible that there are such properties. This paper argues that this promise is illusory. It is not possible to give an analysis of intrinsicality in terms of grounding that is consistent with these theories. Given an adequate analysis should be compatible with these theories, it follows that it is not possible to analyse intrinsicality in terms of grounding.

Metaphysica

This paper investigates the plausibility of Witmer, Butchard and Trogdon’s proposal to distinguish intrinsic properties from extrinsic ones in terms of independence from accompaniment and grounding. I argue that the proposed criterion is not adequate to determine intrinsicality, since according to it some intuitively extrinsic properties turn out to be intrinsic. I suggest and evaluate two responses: first, one could characterize a conception of independence which is specific to the individual instantiating the property; and second, one could justify two assumptions about properties which entail that counterexample properties of the kinds I present do not exist, most importantly that there are no fundamental properties which are instantiated in an intrinsic fashion by some individuals and an extrinsic fashion by others. Although the latter seems prima facie plausible, I present some potential counterexamples to it from current physical theory. I conclude that the grounding- and inde...

Possession of any actual physical property depends on the ambient conditions for its bearers, irrespective of one’s particular theory of dispositions. If ‘self-sufficiency’ makes a property intrinsic, then, because of this dependence, things in the actual world cannot have an intrinsic physical resemblance to one another or to things in other possible worlds. Criteria for the self-sufficiency of intrinsic properties based on, or implying indifference to both ‘loneliness’ and ‘accompaniment’ entail that no self-sufficient property can require its bearers to be extended in space or time, yet all physical properties of concrete objects do require this. These outcomes undermine the vindication of physicalism claimed by neo-Humeans for their metaphysical project. For physical properties dependent on ambient conditions cannot supervene on intrinsic properties independent of ambient conditions: when ambient conditions change we get a change in the former without a change in the latter.