1 See Sellars, W. (a) Science, Perception and Reality (London: Routledge & Kegan Paul 1963),Google Scholar (b) Science and Metaphysics (London: Routledge & Kegan Paul 1967), Chapter I, (c) ‘Metaphysics and the Concept of a Person’ in Lambert, K. ed., The Logical Way of Doing Things (New Haven: Yale U.P. 1969),Google Scholar (d) ‘Science, Sense Impressions and Sensa: A Reply to Cornman’, Review of Metaphysics, 24 (1971), 391-447, (e) ‘The Adverbial Theory of the Objects of Sensation’, Metaphilosophy, 6(1975) 144-160.

2 See Cornman, J.W. (a) Materialism and Sensations (New Haven: Yale U.P. 1971),Google Scholar (b) Perception, Common Sense and Science (New Haven: Yale U.P. 1975).

3 See Davidson, D. (a) ‘The Logical Form of Action Sentences’, in Rescher, N. ed., Logic of Action and Preference (Pittsburgh: U. of Pittsburgh Press 1967),Google Scholar (b) ‘The Individuation of Events’, reprinted in Rescher, N. ed., Essays in Honour of Carl G. Hempel (Dordrecht: D. Reidel 1969).CrossRefGoogle Scholar

4 See Sellars (a), Chapter I.

5 See Sellars (c).

6 See Sellars (d).

7 In a more recent paper on the ontology of perception, ‘Sensa or Sensings: Reflections on the Ontology of Perception’ (read at the University of North Carolina Colloquium, October 1976; forthcoming in Philosophical Studies), Sellars seems to take an event-mentioning formulation as a genuine bearer of ontological commitment. More on this in section VIII. I wish to thank George Pappas for drawing this paper to my attention.

8 See Jackson, F. (a) ‘On the Adverbial Analysis of Visual Experience’, Metaphilosophy 6 (1975) 127–135,CrossRefGoogle Scholar (b) ‘The Existence of Mental Objects’, American Philosophical Quarterly 13 (1976) 33-40.

9 See Sellars (b), Chapter I, Sellars (c), Sellars (e).

10 Sellars (e). I will mention page references to this paper in the text. In Jackson (a), Jackson asserts that ‘The number of cases of persons breathing in a room cannot exceed the number of persons breathing; likewise, the number of after-imagings cannot exceed the number of persons after-imaging.’ (130) Presumably we should read ‘persons’ breathings’ for ‘persons breathing’ in this quotation. Apart from this, it is not at all clear what general principle Jackson is appealing to here. Is it that if a person engages in an event of kind K, then that event cannot be composed of several events of kind K? But what, then, of a person engaging in a complex dance-step whose components are themselves dance-steps? In any case, since this issue could reasonably be seen to be of great importance to an adverbial theorist it is surprising that Jackson does not discuss it more deeply.

11 In a sequel to his original paper Jackson (b), Jackson alludes to the ‘predicate-modifier’ treatment of adverbs, the account proposed by Sellars. He notes that treatments of this sort require an intensional semantics and objects to this in the case of sensory adverbs on the grounds that ‘“I have a pain” and “I have a red image” are statements about the actual world, if any are’ (page 40). I leave the reader to decide what cogency there is to that objection.

12 Sellars himself acknowledges a problem here (Sellars (e), pages 156-157) and sends us to consult Sellars (d).

13 Sellars communicated this argument to me in conversation.

14 See Sellars (b), Chapter I, page 42 ff.

15 Another plausible candidate is ‘looks’-talk, but because the distinctions one needs to draw if such talk is to be useful for present purposes are both complex and contentious, it will be best not to rely on such talk.

16 See Sellars (b), sections 45-69.

17 There are several reasons for this, but perhaps the most convincing is that describing hallucinations by describing their contents can probably be reduced to describing sensations by describing their standard causes (see pages 13-15 below). Pace Smart, the really interesting task of a philosophical theory of sensations is to say not Just that a given sensation is a sensation of the kind standardly caused by red objects but also what kind of sensation it is.

18 Id-2 implies that if s ₁ and s ₂ are identical, then there are counterpart objects for s ₁ and s ₂. But this makes Id-2 too strong, for we may well have theoretical reasons for postulating sensations for which there are no standard causes in the external world. We will, therefore, want to restrict the domain of application of ld-2 to only those sensations that have counterparts. With this proviso in mind we can see that ld-2 is simply a formulation of the lsormophism Condition for identity structures.

19 See Sellars (b), page 55.

20 Sellars (c). Page references will be included in the text.

21 Vendler, Zeno ‘Causal Relations’, The Journal of Philosophy 64 (1967), 704–713.CrossRefGoogle Scholar

22 See Davidson, Donald ‘Causal Relations’, The Journal of Philosophy 64 (1967), 691–703.CrossRefGoogle Scholar

23 See Pollock, John Subjunctive Reasoning, Philosophical Studies Series in Philosophy, Vol. 8 (Dordrecht: D. Reidel 1978), pages 145-151Google Scholar.

24 See Parsons, T. ‘Problems Concerning the Logic of Grammatical Modifiers’, Synthese 21 (1970) 131,CrossRefGoogle Scholar and Thomason, R.H. and Stalnaker, R. ‘A Semantic Theory of Adverbs’, Linguistic Inquiry 4 (1973) 197–220.Google Scholar

25 See footnote (7) for first reference.

26 ibid., section 90.

27 I wish to thank Robert Audi, Joseph Camp, Jr., William Blackburn, George Pappas and Terry Tomkow for helpful comments on earlier drafts.