Langton and Lewis on “Intrinsic”*

intrinsicality in terms of modality and naturalness. Their key idea, drawing on earlier work by Jaegwon Kim, was that an intrinsic property is one that is independent of accompaniment, which is to say that P is intrinsic iff the following four conditions are all met: 1. It is possible for a lonely object to have P. 2. It is possible for an accompanied object to have P. 3. It is possible for a lonely object to lack P. 4. It is possible for an accompanied object to lack P. Langton and Lewis say that an object is “accompanied ” iff it coexists “with some contingent object wholly distinct from itself. ” (Langton and Lewis 1998, p. 333) A “lonely ” or “unaccompa-nied ” object is one that is not accompanied. We will also speak of an object being “accompanied by an F ” iff it coexists with some F wholly distinct from itself. This works very nicely for the obvious examples. It works for being cubical, being 50 km from a capital city, and being lonely (intrinsic, extrinsic, extrinsic, respectively). But it doesn’t work for every property. Langton and Lewis note that disjunctive properties cause trouble: they give the example of being cubical and lonely, or else non-cubical and accompanied. This prop-erty is independent of accompaniment, but intuitively is extrinsic. Other disjunctive properties are intuitively intrinsic (the property of being cubical or spherical, for example), so a new test must be prescribed for them. And that is, indeed, what Langton and Lewis do: they call the non-disjunctive intrinsic prop-erties “basic intrinsic”, and apply the independent of accompaniment test to them. So, a property is basic intrinsic iff it is: 1. not a disjunctive property, and, 2. not the negation of a disjunctive property, and, *Thanks for comments on and discussion of this paper go to Toby Handfield, Lloyd Humberstone, Frank Jackson