Two Problems of Dispositions and the Material World I. Introduction In previous chapters I have been concerned to sort out some of the central ontological questions brought by the existence of dispositions. In this chapter I outline two traditional problems associated with a philosophical analysis of matter: first, the justification for the primary/secondary qualities distinction, and second, the existence of matter. I shall argue in the former case the establishing of the primary/secondary qualities distinction brings involvement with the problems I have been considering here. I will show why the most important distinction to grasp for an understanding of the primary/secondary qualities distinction is that between a disposition, or as Locke said, a 'power', and the categorical base of that power. I will argue further that the primary/secondary qualities distinction is a worthwhile distinction if, and only if, dispositions are non-reductively identical to such categorical bases. In the second part of the chapter I am concerned with an argument advanced by Howard Robinson, which evidently appeared also in Hume, to the effect that matter has no categorical grounding - specifically, no categorical grounding in solidity - and so has no legitimate claim to existence. The claim that matter has no categorical grounding derives from two sources: one conceptual, that an analysis of matter yields only disposition concepts; and the other factual, that scientists have similarly produced only a dispositional account of the physical world. I shall argue however, that to derive the non-existence of the world from its dispositional nature is to misunderstand what dispositions are.
PART ONE: PRIMARY AND SECONDARY QUALITIES II. Another problem from Locke Locke most famously1 divided all the properties of things into three categories: primary; secondary and powers proper. It is often claimed, however, that this
1 It is arguable that the primary/secondary qualities distinction goes back to Democritus who is
reported, by Sextus Empiricus, to have said 'By convention sweet and by convention bitter, by
I
DISPOSITIONS AND THE MATERIAL WORLD
II
distinction is misunderstood (cf. Jackson, 1929,2 Bennett, 1965 and Mackie, 1976). The distinction has been defended against such confusions and I give a qualified defence here though my main concern is with the connection of 'real' properties with powers. Powers I will take to be exactly the same as dispositions.3 First, on qualities, Locke says: 'Whatsoever the Mind perceives in itself, or is the immediate object of Perception, Thought, or Understanding, that I call Idea; and the Power to produce any Idea in our mind, I call Quality of the Subject wherein the power is' (1690, Bk. II, Ch. viii, section 8 (henceforth references given thus: II, viii, 8.)).
Thus the notion of a power or disposition is central to Locke's account of reality, for all qualities or properties of things are known to us only through their dispositions to cause ideas in us. Primary qualities are. . . '. . . such as are utterly inseparable from the Body . . . such as in all the alterations and changes it suffers, all the force can be used upon it, it constantly keeps . . . and the Mind finds inseparable from every particle of Matter . . . viz., Solidity, Extension, Figure, Motion or Rest, and Number' (II, viii, 9).
Secondary qualities, on the other hand, are . . . '. . . such Qualities, which in truth are nothing in the Objects themselves, but Powers to produce various Sensations in us by their primary Qualities, i.e., by the Bulk, Figure, Texture, and Motion of their insensible parts' (II, viii, 10).
This may appear curious in that, in II, viii, 8, all qualities were said to be powers to produce ideas. Locke must not have intended this to be a sufficient condition of secondary qualities, though a necessary one, to which must be added (II, viii, 9) that the primary qualities are those that are 'inseparable from body'. What of secondary qualities, then? Are they to be understood as 'separable from body'? Is convention hot, by convention cold, by convention colour: in reality atoms and void' (Barnes, 1987, pp.252-3). 2 Jackson gave a list of his contemporaries who were guilty of the error.
3 Though as I will mention below (§9.3), Boyle may have intended to distinguish dispositions
from powers.
DISPOSITIONS AND THE MATERIAL WORLD
III
Locke suggesting that while bodies have no colour, sound or smell they do have solidity, extension and all the other primary qualities? How did Locke attempt to state the distinction? Of primary qualities, he adds that they: '. . . are really in them, whether any ones Senses perceive them or no: and therefore they may be called real Qualities, because they really exist in those Bodies. But Light, Heat, Whiteness, or Coldness [that is, the secondary qualities], are no more really in them, than Sickness or Pain is in Manna. Take away the Sensation of them; let not the Eyes see Light, or Colours, nor the Ears hear Sounds; let the Palate not Taste, nor the Nose Smell; and all Colours, Tastes, Odors, and Sounds, as they are such particular Ideas, vanish and cease, and are reduced to their Causes, i.e. Bulk, Figure, and Motion of Parts' (II, viii, 17).
Here Locke evidently is trying to demonstrate that primary qualities are the real qualities of material substances and that secondary qualities are not real. The argument does not show that secondary qualities are unreal however. It is suggested that colours, sounds and smells do not exist unperceived whereas the primary qualities do. But Locke speaks only of the sensations of such qualities, and taking them away, 'as they are such particular ideas', they then 'vanish and cease'. It is the ideas of secondary qualities which vanish and cease, therefore, and such ideas are not to be conflated with the secondary qualities themselves. If we take away the idea of redness or the smell of a rose, then such qualities do still exist in their objects - as powers - only our ideas or sensations, excited by such powers, have ceased to exist; and this is no great revelation, given that 'idea' in this sense is something that necessarily cannot exist unperceived. Therefore, we can say that if ideas, by definition, cannot exist unperceived, then nor can the ideas excited by primary qualities. Primary qualities, like secondary qualities, are powers to excite ideas, and it is these powers that exist unperceived and inhere in objects when we do not perceive them. Locke cannot, therefore, make the distinction by using a notion of 'unperceived ideas'. How then, are we to know which are the primary and which are the secondary qualities? Locke provides us with another criterion: the criterion of 'resemblance'. The ideas of primary qualities do resemble their causes in the objects whereas the ideas of secondary qualities do not. When I see an object as circular, or in motion, or of a one metre width, my ideas are not deceptive: these qualities really are in the objects I see; that is, the ideas resemble their causes. Any sounds, smells or colours of such objects however, in so far as the causes of sounds, smells and colours are not themselves sounds, smells and colours but combinations of minute primary qualities, do not resemble their causes. The
DISPOSITIONS AND THE MATERIAL WORLD
IV
causes of such secondary qualities are solid, extended and at motion or rest; nothing like our experiences of them. This criterion is quite easily attacked on Berkeleyan lines, however. Resemblance, it turns out, is a hopeless criterion of demarcation. All our perceptions of the qualities of substances are ideas, thus there is nothing to compare these ideas with to discover the relative degree of resemblance between quality and idea. All we have is the idea; thus comparison is an impossibility. If we want to maintain a primary/secondary distinction we had better find firmer grounds. A more serious problem for the distinction is to follow; one where the dispositional/categorical relationship is of central importance. We have seen that when secondary qualities are unperceived they are reduced to their causes: bulk, figure, and the other primary qualities. This is a view which Locke has a number of attempts to express, for example: '. . . those secondary and imputed Qualities, which are but the Powers of several Combinations of those primary ones' (II, viii, 22) and '. . . whiteness or redness are not in it at any time, but such a texture that hath the power to produce such a sensation in us' (II, viii, 19). What it is about secondary qualities that exists unperceived is actually some collection of primary qualities, and this does seem consistent with Locke's frequent claim that it is the primary qualities that are the real qualities of physical objects. This is made explicit at II, viii, 25: '. . . the Ideas we have of distinct Colours, Sounds, etc. containing nothing at all in them, of Bulk, Figure, or Motion, we are not apt to think them the Effects of these primary Qualities which appear not to our Senses to operate in their Production; and with which, they have not any apparent Congruity, or conceivable Connexion. Hence it is, that we are so forward to imagine, that those Ideas are the resemblances of something really existing in the Objects themselves: Since Sensation discovers nothing of Bulk, Figure, or Motion of parts in their production; nor can Reason shew, how Bodies by their Bulk, Figure and Motion, should produce in the Mind the Ideas of Blue or Yellow etc.'
Locke's ontology doesn't seem to be one in which secondary qualities play much of a role. The grounds of the powers whose manifestations are our ideas of colours, smells, etc. are actually the primary qualities of their possessors. In the case of colour perception, for example, light hits the substance, which because of the arrangements of its small parts - its 'real', primary qualities - reflects light back in a certain way, such that when it hits the sense organs of an ordinary human perceiver, an idea of a colour is stimulated. As Alexander (1974) rightly says, it seems plausible that we can account for colours in terms of textures but we cannot
DISPOSITIONS AND THE MATERIAL WORLD
V
account for shape and size in terms of colour, or even conceive of such an account. Hence shape and size are regarded as primary qualities. This raises a serious question; one that threatens the whole motivation for a primary/secondary qualities distinction and is analogous to a problem that arose in the previous chapters. If primary qualities are, according to Locke, really in their objects, but all the causes of our ideas of secondary qualities are really primary qualities, then what is left of secondary qualities? Is there anything of secondary qualities remaining? Is there any role left for secondary qualities to play? In sum; is there any justifiable reason to say that secondary qualities exist? The following passage suggests not: 'What I have said concerning Colours and Smells, may be understood also of Tastes and Sounds, and other the like sensible Qualities; which, whatever reality we, by mistake, attribute to them, are in truth nothing in the Objects themselves, but Powers to produce various Sensations in us, and depend on those primary Qualities, viz., Bulk, Figure, Texture, and Motion of parts, as I have said' (II, viii, 14).
This makes the temptation to view secondary qualities as purely mind-dependent entities very easy to understand, for if only primary qualities exist in objects, then where do secondary qualities exist? In our minds only? If secondary qualities are mind-dependent what reason could there be for ascribing them to objects rather than minds? Are they not properties of perceivers rather than properties of objects? Peter Alexander (1974 and 1974a) claims that Locke was not putting forward an argument for the primary/secondary qualities distinction, rather, he already had accepted the distinction from Boyle and adapted it to his own needs. Alexander mentions that 'Locke actually worked in Boyle's laboratory on relatively routine tasks, such as recording weather conditions, and from time to time commenting on Boyle's manuscripts before publication' (1974, p.63). If we are to understand the primary/secondary qualities distinction, therefore, we need to understand the context of ideas from which Locke took it. The ideas that are important are corpuscularianism and mechanism and, in addition to Boyle, this tradition also includes Galileo, Descartes, Gassendi and Newton (Harré, 1964). I will now show that the distinction of primary and secondary qualities is best understood as a distinction between a disposition and the causal or categorical base of a disposition. It is in Locke's predecessors, particularly Boyle, that the importance of dispositions for the distinction becomes clear.
DISPOSITIONS AND THE MATERIAL WORLD
VI
III. Galileo, Boyle, Mechanism and Corpuscularianism G.J. Warnock says that 'Locke in effect erects the current physicists' atomic, or "corpuscular", theory of matter into the ultimate metaphysical truth' (1960, p.159). The mechanistic, corpuscularian philosophy of Galileo, Boyle and Locke is an attempt to dispense with the occult notion of powers and replace them with less ethereal and more empirically accessible entities.4 All change is to be accounted for by the corpuscularian in terms of what is actual, and what is actual - the 'real' properties of matter - is the primary qualities. What, then, is 'unreal' about secondary qualities? That they are minddependent, though a misinterpretation of Locke, is an explicit thesis in Galileo's presentation of the distinction: 'Now I say that whenever I conceive any material or corporeal substance, I immediately feel the need to think of it as bounded, and as having this of that shape; as being large or small in relation to other things, and in some specific place at any given time; as being in motion or at rest; as touching or not touching some other body; and as being one in number, or few or many. From these conditions I cannot separate such a substance by any stretch of my imagination. But that it must be white or red, bitter or sweet, noisy or silent, and of sweet or foul odour, my mind does not feel compelled to bring in as necessary accompaniments. Without the senses as our guides, reason or imagination unaided would probably never arrive at qualities like these. Hence I think that tastes, odours, colours, and so on are no more than mere names so far as the object in which we place them is concerned, and that they reside only in the consciousness. Hence if the living creature were removed, all these qualities would be wiped away and annihilated. But since we have imposed upon them special names, distinct from those of other and real qualities mentioned previously, we wish to believe that they really exist as actually different from those' (1623, p.274).
Galileo's account of secondary qualities is a completely reductive one. They are not real qualities and are believed to be so only through ignorance of their real nature. In detail, this is substantially different to Locke's account. For Locke, it is only the idea of a quality that is taken away when not perceived. All ideas belong to minds and all qualities to objects; thus 'mind-dependent quality' is a contradiction in terms on Locke's account, as is 'mind-independent idea'. Though the details differ, however, the general thesis of primary qualities being the real qualities is common to Galileo, Boyle and Locke. These are the qualities that do all the explanatory work, as we find fully expressed in Boyle's mechanistic account of powers. 4 See also Urmson, 1982, ch.1 for a short outline of the corpuscularian philosophy.
DISPOSITIONS AND THE MATERIAL WORLD
VII
Boyle tells us that 'the Corpuscularians will show that the very qualities of this or that ingredient flow from its peculiar texture and the mechanical affections of the corpuscles it is made up of' (1674, p.151). All the mysterious powers in the philosophy passed down from Aristotle can be explained in terms of shapes and structures of things. For example: 'the solidity, taste, &c., of salt may be fairly accounted for by the stiffness, sharpness, and other mechanical affections, of the minute particles whereof salt consists' (1674, p.149), or the power of a lock to be opened by a key can be explained in terms of the shape of the lock and the shape of the key (1666, p.23). Similarly 'gunpowder itself owes its aptness to be fired and exploded to the mechanical contexture of more simple portions of matter nitre, charcoal, and sulphur' (1674, p.147). Boyle goes into detail on the power of beaten glass to poison. Though the power to poison is regarded as a real quality of the glass it 'is really nothing distinct from [the primary qualities which compose] the glass itself' (1666, p.25). We merely state the mechanism by which substance produces its effect: small, glassy fragments, being 'many [and] rigid . . . and endowed with sharp points and cutting edges. . . pierce or wound the tender membranes of the stomach and guts, . . . whereby naturally ensue great gripings and contortions of the injured parts' (op. cit., p.25). Further, ground glass is no poison to some animals because their guts 'are usually lined with a slimy substance' which sheaths the minute powders (p.26). Boyle offered various other explanations of powers in terms of primary qualities5 and Stewart (1979, p.xix) speaks of it being Boyle's 'lifetime ambition' to provide mechanistic accounts of all the powers. An argument is offered against the allowance of powers. If there are powers, then: 'we must admit that a body may have an almost infinite number of new real entities accruing to it without the intervention of any physical change in the body itself: as, for example, gold was the same natural body immediately before aqua regis [in which it is soluble] and aqua fortis [in which it is insoluble] were made, as it was immediately after' (1666, p.24).
Another example of this kind is the various powers of the Sun. We would think the Sun to be endowed with only a finite list of real physical properties but it apparently has an endless list of powers: to harden clay; to melt butter; to thaw ice; to evaporate water; to ripen fruit; to make plants grow and so on (1666, p.27).
5 (1671), see also separate papers on Fluidity and firmness, Colours and Cold (Stewart 1979,
p.xix).
DISPOSITIONS AND THE MATERIAL WORLD
VIII
Boyle's distinction of primary and secondary qualities can be stated in mechanistic terms. The secondary qualities of colours, tastes, etc. belong to objects purely in virtue of their primary qualities. All our perceptions of secondary qualities can be accounted for in terms of some mechanism and need refer only to primary qualities in various combinations. The universe contains a multitude of corpuscles which mingle together in various ways; 'And when many corpuscles do so convene together as to compose any distinct body, as a stone or metal, then from their other accidents (or modes) . . . there doth emerge a certain disposition or contrivance of parts in the whole, which we may call the texture of it' (1666, p.30)
which then causes us to have ideas of colours, for example. Thus the secondary qualities are just special cases of powers in general, namely, powers that exercise themselves directly on our senses, and the mechanistic explanation of powers suffices also for the explanation of secondary qualities. Boyle's position is thus one that allows no unsupported powers; that is, no powers for which there is no mechanistic explanation in terms of real properties. A power for which there is such an explanation can be called a disposition and is acceptable in the corpuscularian philosophy as a disposition is nothing but a contrivance of primary qualities. Ungrounded powers thus have no place in the corpuscularian's ontology and this makes the corpuscularian the Seventeenthcentury equivalent of Quine (1974, pp.8-12). IV. Powers in Locke's Essay Why does Locke, in his Essay, list qualities that appear to be ontologically superfluous (§9.2)? At best, it would seem, secondary qualities can be described as epiphenomenal properties; for they apparently have no causal role of their own. The manifestations of the powers Locke calls secondary qualities are our ideas of secondary qualities but these ideas are always caused by primary qualities. What have been called tertiary qualities, Locke's 'powers proper', or what we now recognise as common dispositional properties, also have their manifestations caused by primary qualities so they also seem to have no causal role and are thus mere epiphenomena.6 Why then does Locke include the secondary and tertiary in his list of qualities when it seems that once all the primary qualities have been listed, nothing is left unaccounted for; specifically, no 'occult' powers are 6 What distinguishes 'powers proper' from other powers is that they first have an effect upon
another substance, whose changes then affect our senses.
DISPOSITIONS AND THE MATERIAL WORLD
IX
unaccounted for, for all change and all manifestations of 'powers' have received a causal explanation? If, when secondary qualities are unsensed, they 'vanish and cease, and are reduced to their causes' (II, viii, 17), then there is an argument for saying that they are nothing over and above such causes. All three types of quality discussed by Locke at II, viii, 23, are explained through the existence of primary qualities alone. All the qualities that really belong to objects are primary qualities; these are the grounds of the powers to produce our ideas of primary qualities. In their 'minute parts' however, they are also the grounds of the powers to produce ideas of colours, sounds, and smells, which do not resemble anything in the objects themselves. Also, the primary qualities are the grounds of the powers of an object to affect changes in other objects, i.e. powers proper. Jackson (1929) argues that this is how we should understand the distinction: primary qualities are the 'real' qualities, secondary and tertiary qualities are not really qualities at all, but just powers; but then there is no reason for Locke listing them as separate qualities as if they were as real as primary qualities, for as we have seen, powers that are ungrounded in actual mechanisms are inadmissible for the corpuscularian. The notion of 'power' is central to Locke's account; what does Locke say of such powers? He says that we form the idea of a power when we observe regular behaviour of the ideas of one thing to be changed and the idea of another thing as making the change (II, xxi, 1). Thus the idea of a power is an idea without content, for a power cannot be observed. All that can be observed is the idea of a stimulus event and the idea of a manifestation event. Powers are thus 'dummy concepts' (Ayers 1977, p.81) arising when such repeated and regular changes occur. The power is posited as the cause of this change, that is, 'the source from whence all action proceeds' (II, xxi, 4). We know that these sources of action can also be described in terms of the mechanisms that they actually are, as a matter of fact, and, given the identity of causal role (cf. §6.3) and exclusion of overdetermination (cf. §6.4) it seems that an identity of powers with their 'real', primary quality causes is unavoidable. The separate listing of secondary and tertiary qualities, being powers, is, I suggest, justified only if they stand in a non-reductive identity relation to primary qualities. Locke's justification for listing secondary qualities separately must be that we can know of the possession of such a power without knowing the modus operandi, or manner of operation, of the powers. This is what I call the factual occupier of the causal role. Locke puts it thus: '. . . many words which seem to express some Action, signify nothing of the Action or Modus Operandi at all, but barely the effect, with some circumstances of the Subject wrought on, or Cause
DISPOSITIONS AND THE MATERIAL WORLD
X
operating; v.g. Creation, Annihilation, contain in them no Idea of the Action or Manner, whereby they are produced, but barely of the Cause and the thing done' (II, xxii, 11).
It can be seen, therefore, that Locke was sensitive to the distinction between functional role and occupier of functional role. It is this distinction that is the key to understanding the primary/secondary qualities distinction. If a power is the power that it is purely in virtue of its satisfaction of functional criteria, rather than structural criteria of its small parts, then such powers can be separated as differing in sense from such structures, even though such powers refer to the same qualities as those which are also describable as primary qualities. The quality of 'greenness', for instance, cannot be reduced away into primary qualities, for what it is that makes any instance of greenness green, is that it has the functional role of producing ideas of green in normal perceivers. Even though some combination of primary qualities is responsible for such ideas of greenness, greenness does not mean the same thing as these primary qualities, it means that the quality satisfies specifically functional criteria. I offer this as the best manner of presenting Locke's distinction, but is it how Locke intended us to interpret the distinction? Cummins (1975) argues to the effect that it is. In explaining Locke's picture of scientific explanation Cummins states that '[t]o have a power is simply to behave in conformity with a certain rule. This does not explain a power: we may very sensible ask why something has a certain power - that is, why it behaves in conformity to a certain rule. But this is another inquiry' (pp.410-411). If this is so, then Locke's view of the meaning of a power ascription can be seen as distinct from his account of the ontological question of powers. To have a power just is to satisfy certain functional criteria, and how such criteria are satisfied is no part of the meaning of a power ascription, rather it is a question for the scientist to answer. On the ontological question Locke had clear commitments towards corpuscularianism; a philosophy where unexplained, 'brute' powers, had no place. What I hope to have shown however, is that unless Locke argues for the conceptual irreducibility, there is no reason to posit secondary qualities at all. For Locke, therefore, the meaning of secondary quality terms is not reducible to primary quality terms,7 though powers are ontologically to be identified with primary qualities. The only reason we would have to speak of secondary qualities is their conceptual distinctness and so Locke must have an ontology that does not differ greatly from the one I have developed. A particular 7 Quine argues for the reduction in meaning of disposition terms (1960, p.224) as well as the
ontological reduction of the dispositional to the categorical.
DISPOSITIONS AND THE MATERIAL WORLD
XI
property can be referred to in two ways: as a power or as a mechanism. The difference between my position and the mechanistic one however, is that Boyle and Locke are committed to the view that the one description of the property expresses the true nature of reality while the other description does not. My position has no favourite in this respect.8
PART TWO: THE EXISTENCE OF MATTER V. Robinson's Attack There is a major problem inherent in Locke's account of matter which remains unresolved and is used by Howard Robinson (1982, ch. 7) to attack the very existence of matter. It is argued that 'matter' is an incoherent notion and that the only things that we can say exist categorically, are minds with mental states as their objects. The same problem has been touched upon by Blackburn (1990) where, unlike Robinson, the self-sufficiency of dispositions is recognised. I will argue that it is in Robinson's understanding of dispositions that the argument faulters and if we treat dispositions correctly we can save the material world from the attack. In the defence of matter I shall also defend an identity relation between a generic disposition and categorical term. Robinson begins by looking at the philosophical attempts to characterise matter - to find a simple, unconditional and categorical account of physical things - thus his initial concern is with conceptual analysis. Descartes (1641, second meditation) said that extension was of the essence of material bodies; that is, a body is a geometrically definable space. But this is obviously an incomplete characterisation of body. Extension provides only the idea of a 'bare geometrical figure' which may just be a volume of empty space. What needs to be added to Descartes' idea of extension is that the extended space be an occupied one. Locke attempted to make such an addition by including solidity in his list of primary qualities (§9.2), the other primary qualities being characterising of shape, size and position. Only by the addition of solidity to such 'geometrical' qualities is the area of space made an occupied one. Robinson's critique of this analysis is not new; it can be found also in Hume. Locke claimed, as we saw, that primary qualities are really in substances and that the ideas they produce resemble the qualities themselves. But solidity seems to be problematic in this respect. It could be argued that we have no idea of 8 The question is taken up in chapter XI.
DISPOSITIONS AND THE MATERIAL WORLD
XII
such a quality at all, let alone an idea which resembles the quality itself. Hume points out that our ideas reveal only a property of impenetrability, and impenetrability is a dispositional property: 'In order to form an idea of solidity, we must conceive two bodies pressing on each other without penetration; and 'tis impossible to arrive at this idea, when we confine ourselves to one object, much more without conceiving any' (1739-40, Bk. I, Part IV, section 4).
Without any independent idea of solidity we have no coherent account of matter, for no other supposed quality can fill the role solidity is supposed to play: 'Now what idea have we of these bodies? The ideas of colours, sounds, and other secondary qualities are excluded.9 The idea of motion depends on that of extension, and the idea of extension on that of solidity. 'Tis impossible, therefore, that the idea of solidity can depend on either of them. For that wou'd be to run in a circle and make one idea depend on another, while at the same time the latter depends on the former. Our modern philosophy, therefore, leaves us no just nor satisfactory idea of solidity; nor consequently of matter' (Hume, loc. cit.).
Robinson's argument is to the same effect. We have arrived at a conception of material bodies as volumes of impenetrability. But impenetrability is a dispositional property only, that is, it describes how an object will act under certain circumstances when it collides with another body of its own kind, not what it is in itself. We are lead to look, then, at what the occupant of the impenetrable volume of space is; we want to find the actual properties that support this possibility of resisting penetration. The clear assumption here is that only the non-dispositional properties of a thing tell us what it is really like. We would not characterise water simply as 'that which dissolves sugar', for this tells us only how water behaves in certain circumstances. What we need is a non-dispositional account, such as 'water is H2O'. This 'categorical base' will explain the dispositional properties of the substance and Robinson argues that we need just such a 'base' for our concept of matter. However, if Robinson is correct, no such categorical base can be found which would be responsible for a material body's powers or dispositions; these being its impenetrability, perceptible qualities and effects on other objects.
9 The reason for this being that '[t]he impressions, which enter by the sight and hearing, the smell
and taste, are affirm'd by modern philosophy to be without any resembling objects; and consequently the idea of solidity, which is suppos'd to be real, can never be deriv'd from any of these senses' (loc. cit.).
DISPOSITIONS AND THE MATERIAL WORLD
XIII
Conceptual analysis reveals, therefore, a conception of matter which is entirely dispositional. The argument then proceeds to a quasi-Berkeleyan conclusion as follows: Matter, and all material states, are understood now in terms of dispositions to produce effects only, that is, in terms of quantitatively discernible forces, fields and energies, which are all instances of dispositions. The effects matter produces are the perceptible qualities which are accessible to our understanding. Such perceptions are the only things which are actual or categorical because, as Robinson established in the preceding chapters of Matter and Sense, they are the only things which cannot be further reduced to dispositions or potentialities. Furthermore, such perceptible qualities are mind-dependent because they vary according to the perceptual apparatus of the perceiver. Therefore, we arrive at the conclusion that 'the only categorical entities . . . are mental states with sensible qualities as their objects' (p.117). Such mental states cannot be physical states as modern philosophies of mind assert, for if they were physical states then we would have to give them a dispositional account in terms of powers only, and no categorical base would have been discovered at all. The postulation of any mindbrain identity thesis will therefore be self-defeating according to Robinson's argument. Is Robinson correct? If he is then the conclusion is of the greatest importance, for he has, with one argument, dismissed psychophysical identities, proven the incoherence of matter, and thus proven Idealism. One move to block Robinson's argument is to attempt to defend solidity as the categorical grounding of matter. One could argue that solidity is not observable but is, rather, inferred from the observation of impenetrability. Robinson is aware of this move and raises a number of objections to such an inference. First, we would need to specify the relationship between the quality of solidity and the power of impenetrability and Robinson thinks that the connection is only a contingent one; it is nothing stronger than a Humean constant conjunction and, lacking any a priori connection, the two could possibly become separated. Without there being a necessary connection we can imagine solidity without impenetrability or impenetrability without solidity so the suggested analysis fails. The problem is, explains Robinson, that the solidity must explain the impenetrability: '[w]hat is desired . . .' he says, 'is that the quality is a genuine quality, the power a power, and that the two are necessarily connected such that there could not be a quality like that if it were not also a power: the quality entails and therefore explains the power' (pp.110-111). Such an account is lacking. Second, Robinson says that we are unjustified in inferring a categorical quality from a perceptible or dispositional quality, for such perceptible qualities
DISPOSITIONS AND THE MATERIAL WORLD
XIV
are sense-receptor and mind-dependent, thus according to Robinson, lacking any objective status. We have no reason to believe that just because we perceive the world in a certain way, the world actually corresponds to our perception. As Berkeley argues, the true cause of our ideas could be something totally different to those ideas, such as an omnipotent God. Third, Robinson introduces the scientific account of matter to back up his conceptual analysis. He argues that any such solidity is only a feature of matter at the macroscopic level. If modern scientific accounts are to be believed, as materialists want us to believe, at the microscopic level there is no such thing as solidity. Modern physics conceives of matter in terms of fields, energies and charges which are not solid: '[t]herefore, if the ontology of modern science is correct, solidity is not an intrinsic feature of matter' (p.112). It seems, then, that modern physicists want to tell us that no physical world exists! This is Robinson's interpretation at least. How we understand dispositional properties is crucial to the success or failure of Robinson's argument and I argue that the correct understanding is one which makes Robinson's attack a failure. We can see the importance of the construal of dispositions when Robinson's spells out the stages of his argument systematically. I call this the regress argument, which proceeds as follows (pp.114-5): 1) 'Every real (categorical) object must possess a determinate nature'. 2) 'The nature of any power P is given by what would constitute its actualisation'. From which it follows that: 3) 'If P is a real object it must be a power to a determinate actualisation'. The regress becomes apparent in the following: 4) 'If a power Q is the actualisation of P, the determinacy of P will depend upon the determinacy of Q; that is, of Q's actualisation'. 5) 'The list of effects constituting the determinate and complete nature of P will be finite only if the list contains (and thereby terminates at) an effect which is not a power'. and
DISPOSITIONS AND THE MATERIAL WORLD
XV
6) 'An infinite list constitutes indeterminacy' and in the case of matter: 7) There can only be a list of powers for the nature of matter, so the nature of matter is indeterminate. If material entities can only be analysed in terms of powers or dispositions then we have no specification of what they actually are, for a power is a power to alter another physical body; but that physical body is only a set of dispositions or powers to alter other physical bodies, which are powers acting on other physical bodies, and so on. Thus Robinson's statement: 'a power: P is the power to produce (alter) a power to produce (alter) a power to produce . . . Such a formula seems to tell us nothing about what is actually done' (p.116). Robinson has argued for two main theses. First that powers exist only in virtue of a determinate actualisation. Second, claiming the support of modern science, that physical entities are not categorical entities. So what, asks Robinson, are these categorical entities upon which everything depends? Sensed qualities are the only possible candidates because they represent actuality rather than potentiality; 'the phenomenal realm is that area which cannot be reduced to potentialities . . . the only categorical entities that can end the regress are mental states with sensible qualities as their objects' (p.117), objects which cannot be physical if the regress is to end. Robinson has allowed, therefore, only a dispositional existence to matter, an existence which depends entirely on categorical mental states. VI. Solidity, Impenetrability and Identity Robinson thought that solidity was not an intrinsic categorical property of matter for a number of reasons, one being that science tells us that matter is only solid at the macroscopic level, another being that we cannot find a strong enough relationship between solidity and impenetrability over and above a contingent constant conjunction. This last point is raised again by Robinson when he considers the 'last ditch' for materialists; the thesis that there is an unknown, nameless residue behind sensible qualities. Again Robinson argues that there can only be a contingent relationship between the unknown centre and the qualities of the object. It is therefore imaginable that the two could become separated in the way that we can imagine a magnetic field persisting in the absence of the magnet. Something stronger than a constant conjunction is required; something that makes the field of powers 'essentially . . . tied to their centres in a manner which defies
DISPOSITIONS AND THE MATERIAL WORLD
XVI
intuition' (p.123). Robinson thinks any such stronger connection between the object itself and the appearance of the object to be unimaginable but I will suggest a connection which will be sufficiently strict to exclude the separation of a body from its perceptible qualities. This will be a relationship which will allow us to reintroduce a certain understanding of solidity as a criterion of the material and also allow us to answer Robinson's criticisms of the notion outlined in §9.5. The relationship I shall suggest is one of identity. It is no mystery, nor is it in the slightest defiant of intuition, that the morning star is 'tied to' the evening star, and that water is 'tied to' H 2O. We are dealing with a relationship of identity here, where we do not have two things 'tied together' in such a way that they could become separated, rather we have a single thing or kind under more than one name or description. It is just such a relationship which we could argue existed between an object at the centre of appearances and the object as it appears to us in the phenomenal realm. Robinson discusses these objects as if they were distinct, as he does for instance when he dismisses the position I wish to take: 'It is difficult to prove that the importation of synthetic necessities into the phenomenal realm in this way is improper. It does, however, look as if the peculiar desired position seems legitimate only because it is a confusion . . . the contingently connected quality . . . becomes confused with the directly experiencable power . . . Put at its weakest, such synthetic necessities should be objects of suspicion' (p.111).
But surely we have a single object with both qualities and powers. The object at the centre of appearance has the essential quality of solidity but this is that same thing that we experience in the phenomenal realm as the manifestations of impenetrability. We can say that because we are dealing with two essential properties of a single object, one categorical and one dispositional, then there is a necessary connection between the two, even though there is no a priori relationship between solidity and impenetrability when we analyse them in isolation. Where, then, does Robinson go wrong in his understanding of dispositional properties that allows his regress argument to go through. Robinson says (pp.108-9) that to ascribe a disposition is only '. . . to say how a thing will sometimes act' and 'not to say what it is'. Such a view is reminiscent of Ryle's phenomenalism, where the ascription of a disposition was nothing more than affirming the truth of a subjunctive conditional: (DfR) 'If it were the case that , then it would be the case that '.
DISPOSITIONS AND THE MATERIAL WORLD
XVII
Such a conditional analysis of dispositions does give the appearance that they are only allusions to some future possibility and that there is nothing in actuality that dispositions are. Robinson is clearly looking for, or claiming to look for, the actualisation of dispositions in the actuality of their manifestations. Thus when matter is characterised only dispositionally, as impenetrable volumes of space, Robinson looks to the manifestations of such impenetrability in the categorical properties that are mental states. But Robinson ought to have considered the following two points: the ascription of a disposition cannot be simply equivalent to a true conditional (Martin, 1994 and §8.5 above), and we need not look for the actualisation of a dispositional property solely in its manifestation, for we understand dispositions to be actual whenever they are ascribed. If, as I have suggested, to ascribe a disposition is to say that: (DfM) x ( Dx P ( Px & P is causally efficacious of -ing upon being -ed)) then we can understand why dispositions are actual even when not currently manifested. Their actuality consists in the possession of a property which will play a certain causal role in certain circumstances. This offers a way out of Robinson's regress. We can find the actuality of matter in the property which is causally efficacious of impenetrability when in contact with other matter. 'Solidity' thus stands for the categorical correlate to the disposition of impenetrability. What this categorical correlate is will be a complex matter and is for physicists to discover. The categorical correlate may even be described using further disposition terms, which would seem to leave the account open to Robinson's criticism that characterisations of matter will never be able to eliminate dispositions from the account and in modern physics there is no solidity at the microscopic level. The following considerations should be noted, however. First, solidity names whatever property it is which is type-identical to impenetrability. It may well be that as physics advances it offers more complete accounts of the causal base of impenetrability. In this case, 'solidity' can be understood as standing in place of whatever the full account is. Second, it may be objected that I am ignoring the whole of modern physics in my defence of solidity, for physics does characterise matter in dispositional concepts: in terms of forces, fields, and energies. I claim, however, that the authority of modern physics is insufficient to falsify the identity theory of dispositional properties. As physical theory advances it offers structural explanations for dispositional phenomena; this may well involve reference to further dispositions. But even if there was nothing solid at the microscopic level this would not mean that there was no such thing as solidity. The matter we encounter is macroscopic and it seems perfectly reasonable to say that at this level matter is solid, even if it is not solid at the
DISPOSITIONS AND THE MATERIAL WORLD
XVIII
microscopic level. As Joske (1967, p.15) points out, on this very issue, '. . . although we speak of elements, such as lead, as composed of molecules of lead, these molecules are not themselves leaden in the everyday sense of that word'. Similarly, the components of motor cars are not themselves motor cars, but this does not mean that motor cars are a fiction. In sum, that a property is not present at the microscopic level is no proof that it is also absent at the macroscopic level. VII. Conclusion For Hume and Robinson, and, we suppose, Locke, the definition of matter as an impenetrable volume of space was unsatisfactory. In Locke's case this lead to a new quality being brought in - solidity - which could be the ground of the disposition of impenetrability. Hume noted that no such quality is found in experience, Robinson brought forward further objections. All three assumed, however, that to ascribe impenetrability was not to ascribe something that a substance actually was, rather it was to refer to a possible event that could happen. All these views therefore, begin from the assumption that dispositions are not real qualities or properties of things and do not suffice to confer reality upon their subjects. The identity theory asserts that dispositions are real properties which are inherent in their owners, even when they are not manifested. Thus the characterisation of matter as an impenetrable volume of space is a legitimate characterisation of matter. The objection that no quality of solidity is found in our experience is ineffective. Impenetrability is a part of our experience and this is a dispositional description of the property which could also be referred to as solidity, hence solidity is experienced.
