page 2 of Berkeley's Argument for Immaterialism

Level 2 phenomena are constituted by level 1 data – not reductively, but mediated in a way revealed by a third, metaphysical, level of explanation (level 3), which describes the causal-intentional activity of mind (ultimately: of an infinite mind) in producing the level 1 data and the level 2 world constituted for us by the organisation, coherence, and character of the level 1 data (P25-9, 51-2, 2D216).

The analysis can be illustrated by Berkeley's account of causality, which is fundamental to his thesis (P25-9, 51-2, 2D216). At level 3 the world is described as consisting of spirits (minds) and their ideas. Spirits are active, ideas inert. What we take at level 2 to be a case of natural causality– the heat of a fire causing water in a kettle to boil – is, strictly, a succession of individual ideas (composed of level 1 data) caused in us by God (level 3) in such a way that the regularity and consistency of their relations establishes in us a custom of thinking in the familiar level 2 way. This application of the distinction of levels provides, moreover, the basis of the proto-Positivistic philosophy of science sketched by Berkeley later in P (P86-117).

It is a common mistake among commentators to describe Berkeley as a phenomenalist. The distinction of levels shows why they are wrong. Briefly, classical phenomenalism is the view that physical objects are ("logical") constructions out of actual and possible sense-data. The modal adverbs in that sentence serve to explain how the desk in my study exists when not currently being perceived, by showing that we take as true a counterfactual conditional stating that the desk could be perceived if any perceiver were suitably placed. That indeed defines what, on the phenomenalist view, it is for such objects to exist: namely, as at least enduring possibilities of perception. An essential commitment of phenomenalism, therefore, is that certain counterfactuals are to be taken as barely (that is, non-reductively) true; which says, in material mode, that the world contains irreducible possibilia.

Berkeley's view is completely different. The esse est percipi principle requires that a thing must be perceived – actually perceived – in order to exist. The perceivability of my desk when it is not currently being perceived (by a finite mind) is therefore cashed in terms of its actually being perceived (by an infinite mind). In phenomenalism there are only levels 1 and 2. It is a familiar problem for phenomenalism that level 2 cannot be reduced to level 1 without remainder, and that therefore level 1 can only be sufficient for level 2 if suitably supplemented. The supplement is acceptance of the bare truth of appropriate counterfactuals (and thus an ontology of possibilia). This exacts a high price for the explanatory shortfall. But for Berkeley there is no such shortfall; his third level of explanation shows how level 1 constitutes level 2, and simultaneously gives us a simple account of counterfactuals by having their truth-conditions fully statable in indicative terms: "If I were in my study I would see my desk" is true just in case "My desk is perceived by the infinite mind" is true ( = "the desk exists"). So on Berkeley's view possibility is relative to finite minds only – for the infinite mind whatever is, is actual. (Whether any of it is also necessary is of course a different and further matter).

So, says Berkeley, to say that things exist is to say that they are perceived, and therefore "so long as they are not perceived by me, or do not exist in my mind or that of any created spirit, they must either have no existence at all, or else subsist in the mind of some eternal spirit" (P6). And from this the conclusion it follows that "there is not any other substance than spirit, or that which perceives" (P7).

In sum argument is this: the things we encounter in episodes of perceptual experience – apples, stones, trees – are collections of "ideas." Ideas are the immediate objects of awareness. To exist they must be perceived; they cannot exist "without the mind." Therefore mind is the substance of the world.

Berkeley's defence of this argument from P7 onwards reveals the machinery that drives it, consisting of the interplay between three crucial commitments and the application of an analytic method which requires us to recognise three different levels of explanation – whose own interrelations, in turn, are pivotal to his case.

The Machinery of Berkeley's Argument

Let us take the question of the three levels first. Berkeley distinguishes between "strict", "speculative" or "philosophical" ways of understanding matters, and ordinary or "vulgar" ways of doing so. When we "think with the wise" we find it necessary to give explanations at what I shall label "level 1" and "level 3." When we "talk with the vulgar" we do so at "level 2" (see e.g. P34-40, esp. P37; 45-8, 3D234-5, C274).

Level 1 concerns the phenomenology of experience, consisting of the data of sensory awareness in the form of minima of colour, sound, and so for the other senses. Level 2 concerns the phenomena of experience ‹ the tables, trees, and so forth, that we see and touch in the normal course of perception. The phenomenological level (call it level 1) is apparent to us only on a "strict and speculative" examination of experience. Level 2 phenomena are constituted by level 1 data – not reductively, but mediated in a way revealed by a third, metaphysical, level of explanation (level 3), which describes the causal-intentional activity of mind (ultimately: of an infinite mind) in producing the level 1 data and the level 2 world constituted for us by the organisation, coherence, and character of the level 1 data (P25-9, 51-2, 2D216).

The analysis can be illustrated by Berkeley's account of causality, which is fundamental to his thesis (P25-9, 51-2, 2D216). At level 3 the world is described as consisting of spirits (minds) and their ideas. Spirits are active, ideas inert. What we take at level 2 to be a case of natural causality– the heat of a fire causing water in a kettle to boil – is, strictly, a succession of individual ideas (composed of level 1 data) caused in us by God (level 3) in such a way that the regularity and consistency of their relations establishes in us a custom of thinking in the familiar level 2 way. This application of the distinction of levels provides, moreover, the basis of the proto-Positivistic philosophy of science sketched by Berkeley later in P (P86-117).

It is a common mistake among commentators to describe Berkeley as a phenomenalist. The distinction of levels shows why they are wrong. Briefly, classical phenomenalism is the view that physical objects are ("logical") constructions out of actual and possible sense-data. The modal adverbs in that sentence serve to explain how the desk in my study exists when not currently being perceived, by showing that we take as true a counterfactual conditional stating that the desk could be perceived if any perceiver were suitably placed. That indeed defines what, on the phenomenalist view, it is for such objects to exist: namely, as at least enduring possibilities of perception. An essential commitment of phenomenalism, therefore, is that certain counterfactuals are to be taken as barely (that is, non-reductively) true; which says, in material mode, that the world contains irreducible possibilia.

Berkeley's view is completely different. The esse est percipi principle requires that a thing must be perceived – actually perceived – in order to exist. The perceivability of my desk when it is not currently being perceived (by a finite mind) is therefore cashed in terms of its actually being perceived (by an infinite mind). In phenomenalism there are only levels 1 and 2. It is a familiar problem for phenomenalism that level 2 cannot be reduced to level 1 without remainder, and that therefore level 1 can only be sufficient for level 2 if suitably supplemented. The supplement is acceptance of the bare truth of appropriate counterfactuals (and thus an ontology of possibilia). This exacts a high price for the explanatory shortfall. But for Berkeley there is no such shortfall; his third level of explanation shows how level 1 constitutes level 2, and simultaneously gives us a simple account of counterfactuals by having their truth-conditions fully statable in indicative terms: "If I were in my study I would see my desk" is true just in case "My desk is perceived by the infinite mind" is true ( = "the desk exists"). So on Berkeley's view possibility is relative to finite minds only – for the infinite mind whatever is, is actual. (Whether any of it is also necessary is of course a different and further matter).

Berkeley's view is completely different. The esse est percipi principle requires that a thing must be perceived – actually perceived – in order to exist. The perceivability of my desk when it is not currently being perceived (by a finite mind) is therefore cashed in terms of its actually being perceived (by an infinite mind). In phenomenalism there are only levels 1 and 2. It is a familiar problem for phenomenalism that level 2 cannot be reduced to level 1 without remainder, and that therefore level 1 can only be sufficient for level 2 if suitably supplemented. The supplement is acceptance of the bare truth of appropriate counterfactuals (and thus an ontology of possibilia). This exacts a high price for the explanatory shortfall. But for Berkeley there is no such shortfall; his third level of explanation shows how level 1 constitutes level 2, and simultaneously gives us a simple account of counterfactuals by having their truth-conditions fully statable in indicative terms: "If I were in my study I would see my desk" is true just in case "My desk is perceived by the infinite mind" is true ( = "the desk exists"). So on Berkeley's view possibility is relative to finite minds only – for the infinite mind whatever is, is actual. (Whether any of it is also necessary is of course a different and further matter).

Many of the difficulties standardly alleged in Berkeley's argument vanish when understood in light of the three-level analysis. Illustrations of this occur in due place below.

As noted, three crucial commitments interact with the distinction-of-levels thesis to underwrite Berkeley's argument. They are commitments to empiricism, to the epistemic character of modality, and, as we have already seen, to the vacuity of the notion of abstract ideas. It might be more accurate to describe the two first as commitments and the third as the conclusion of an argument; but because the two first are premisses of that argument, and because all three powerfully combine in the process of refuting scepticism and establishing spirit as the only possible substance, it is convenient to take them together.

Berkeley is a rigorous empiricist; we are not entitled to assert, believe, or regard as meaningful, anything not justified by experience. The constraint is austerely applied: level 2 is exhaustively explained by level 1 under government of the level 3 causal-intentional story (see e.g. P38). It might appear that Berkeley is less rigorous in his empiricism than Hume because he introduces the notion of "notions" to explain our knowledge of spirit (other minds and God), which seems expressly to involve a non-sensuous epistemic source, and therefore to conflict with his notebook commitment to the strong principle nihil est in intellectu quod non prius fuit in sensu (C779).But we should allow Berkeley at least as much latitude as Locke claims in countenancing intellectual sources of experience. In this sense notions are the counterpart of "ideas" in Berkeley's sense (mental contents) in the experience of encountering minds through a certain class of their effects.

....3/