The Powers of Megaric Actualism

The Powers of Megaric Actualism De Quincey describes ‘a tranquillity that seemed no product of inertia, but as if resulting from mighty and equal antagonisms; infinite activities, infinite repose.’1 This comes remarkably close to the conception of the material world of Megaric Actualism, the theory of powers defended in this paper. Molnar2 finds the position in Aristotle, who characterises it as the thesis ‘that a thing ‘can’ act only when it is acting, and when it is not acting it ‘cannot’ act, e.g. that he who is not building cannot build, but only he who is building, when he is building, and so in all other cases.’3 Proponents of MA reject the dualist conception of powers, dominant since Aristotle, on which powers have an actual/active and a potential/inactive mode of existence. 4 According to MA, the being of powers is univocal: they are always exercised. So stated the view faces obvious difficulties, which Molnar deftly presses. But duly clarified and explicated, MA confers two significant benefits: it simplifies and demystifies our understanding of powers, as compared with current alternative views. In part one I lay out MA, by visiting some of the problematic philosophical issues around powers. I show how MA solves the problems. In part two I consider objections to MA and attempt to meet them. I. Powers: Philosophical Issues Powers and Conditionals Let me begin by agreeing with Heil that the project to analyse power ascriptions in terms of conditionals is relatively uninteresting from a metaphysical standpoint. Conditionals like ‘would break if struck’ are handy pointers to the presence of a power in an object (in this case fragility). But the endeavour to reduce power ascriptions to such conditionals cannot hope to shed much light, certainly not as much as we need to see them clearly, on powers themselves. For as Heil observes, even if we found a counterexample-proof conditional analysis of power ascriptions, we ‘would still be faced with the question, What are the truth 1 2 De Quincey 1822. Molnar 2003, chapter 4. 3 Aristotle Met: 822. 4 Psillos (2006:138) calls this the ‘dual life’ of powers. 1 makers for dispositional claims?’5 Since my interest is in the ontology of powers—what we are talking about when we speak of powers—I shan’t concentrate on issues relating to the analysis of power ascriptions,6 but rather on the analysis of powers themselves.7 It might be objected that for aught we know all there is to powers is the truth of certain conditionals. That’s of course one of the possible options (although I don’t believe it is the correct option). But still, it seems that if we end up embracing the thesis that all there is to powers is the truth of such conditionals, this will be based upon our metaphysical investigations; it won’t proceed from consideration of the conditionals as our primary focus. So my decision largely to ignore this arm of the debate is vindicated. Having said this, we will have occasion later to formulate a rough conditional expression of power, for those who care for such things, firmly premised upon our metaphysical conclusions. Powers and Manifestation It is widely held that there is more to a power than its manifestation. The charge of a given electron (‘Ed’), for example, is a power. A clear manifestation of this power is when Ed repels other, identically charged, electrons. But even if Ed is transported into an otherwise empty universe, where no manifestation is possible, it retains the power to repel other electrons; it is still (negatively) charged—so we want to say. Ed’s charge is an actual, ongoing affair, a way it is with the electron constantly. Though this powerful property cannot do anything, in the empty universe, it still exists within Ed. I agree that manifestations do not exhaust the being of powers, and that powers exist even if they never have the possibility of manifesting. But I interpret this idea in an unconventional way, one that eventually enables us to view powers in a pleasingly unproblematic light. Powers are made to seem mysterious entities on account of their apparently ethereal8 nature. Where two objects possess two different powers that are never able to manifest, it seems that one of the real respects in which these objects differ is as to things they would or might do, but never in fact do. These are real respects of difference as long as we take powers to be, in some sense, genuine properties of things. The difficulty is this: the nature of a power appears only to be fully actualised when it is manifesting—Ed’s charge only exists in a way that expresses its nature as and when it is busy repelling other electrons (or attracting protons, or doing other causal work). If we want to say that Ed is always charged, even when this is not manifesting—even if it never manifests—the puzzle is how to understand the reality of the power in the meanwhile. In what manner do we conceive of Ed’s charge-power outside of contexts of occurrent repulsion or attraction?9 We need a positive conception of this sort in order to make proper sense of the real respect in which an electron differs from a proton by having a negative as opposed to a positive charge in a scenario where neither particle is ever given an opportunity to put these powers to work. Otherwise, unmanifested powers are left unacceptably ethereal. 5 6 2005:345. Another view is that power ascriptions entail certain conditionals, without being reducible to them (e.g. Mumford 1998). This view will not be of much interest to us for similar reasons. 7 Mellor (2000:758) also distinguishes the two issues, but disagrees about the order in which they should be tackled. 8 The apt term is Goodman’s (1954). 9 The title of Psillos’ (2006) paper ‘What Do Powers Do When They Are Not Manifested?’ reflects the seriousness of this issue. 2 MA’s solution to this difficulty is to hold that Ed’s negative charge is always fully expressing its nature. It is, contrary to appearances perhaps, ever-active: it is never merely dormant, awaiting the stimulus that will make it leap into life. The term ‘power’ (as opposed to ‘disposition’) is helpful for the purpose of accurately conceiving of Ed’s charge. For we can easily imagine power as ‘switched on’, as continuously humming-away, and this is the essential form of MA’s conception of powers. When we say of an electrical system ‘the power is on’, we have in mind that the system is active, as an ongoing matter. It is because of this ongoing activity that the system’s power can make things happen: if we plug in the kettle, water boils as a result. The power does not activate somehow just when the kettle is plugged in: it is not ‘triggered’ by the kettle being plugged in. Rather the power is fully real and occurrent, trying to do all that it can do, before the kettle is plugged in. For another analogy, consider a fast-flowing river. The river has the power to carry a stick we throw into it away to sea. But that power is an occurrent affair: the river is flowing. Throwing the stick into the river causes no change in the river, it does not cause the river’s power to ‘come on’. What throwing the stick into the river does is allow something to be acted upon by the river’s (ever-active, fully realised) power. Ed’s charge is powerful in the same sense: as well as being causally potent it is never less than fully active. Let us put the idea in this way: When Ed’s charge does have a manifest effect, such as repelling another electron, the power is doing no more, in its own right, than it does while Ed is too far away from anything else to interact. Ed’s charge is fully exerted at all times, we may say. The only difference between contexts with and contexts without manifestation is that in the former the world co-operates with the powerful charge, providing it with things it can act upon in ways we can notice. In order to explain the effects of powers we might imagine that powers remain dormant and are ‘activated’ by their so-called ‘stimuli’. Or we could imagine that powers, where they exist, are ever-active, but cannot have manifest effects except when suitable conditions come their way (for example when things come near enough to be affected). This second conception of powers is simpler, and avoids the mystery concerning the nature of unmanifesting powers. On MA manifestations acquire a primarily epistemological character. On conventional conceptions manifestations are essential to the ontology of powers: for it is thought that it is when a power is fully realising its nature that manifestations occur. This is mistaken. In fact, in one good sense, powers are always manifesting, because that is what goes with the notion that such properties are always fully ‘on’, they are fully active as long as they exist. It will help us to switch examples in order to see the appeal of this point. Let us take it that talk of the fragility of a certain vase picks out some structural property—the vase’s molecules being bonded together thus-and-so. It is their being bonded together thus-and-so which means that when dropped, the vase’s collision with the floor will split a good many of these molecules apart, producing the vase’s shattering. According to MA, we must think of the vase’s fragility as existing in just as full a sense before the vase is dropped. All the time it is sat quietly on the mantelpiece it is fragile in an ongoing way, which is to say that the structural property concerning its constituent molecules is continuously, occurrently, in place. With this much many authors claim to agree, for the power is said to be logically distinct from its manifestation. But it is simply not true that this structural property, what we are getting at with our talk of the vase’s fragility, exists somehow in dormant form before the vase hits the floor. Rather that structural property is constantly exerting itself, because it is always fully and powerfully realised. To understand this consider that the manifestation of the vase breaking is the upshot of the structural property combining with the circumstances of colliding with the floor under normal gravitational conditions (etc.). But then it is easy to see that the same structural property has other ‘manifestations’ under other conditions. For example, that same pattern of molecular bonding, in the context of the vase sitting on the 3 mantelpiece, is responsible for the vase remaining solid instead of, say, forming a puddle. It is not that there are no collisions occurring while the vase is on the mantelpiece: air molecules are constantly colliding with it. Your elbow may gently knock it as you pass, without dislodging it from its perch. Yet none of these collisions is sufficient to break the vase’s molecular bonds—its molecular structure resists these collisions. But resisting is a perfectly active and ongoing state of affairs; as is the ever-tense, ever-taut state of molecules bonded into a solid, actively held together by the charge they inherit from their constituent particles. Therefore, it is another ‘manifestation’ of the vase’s molecular bonding that it doesn’t do anything much while sat on the mantelpiece. It is only the ongoing powerful exercise of the vase’s molecular bonding that ensures nothing at all exciting happens to it until it is dropped. We ignore the manifestations of powers that simply keep things the same at our peril.10 What we conventionally consider manifestations—such as vases breaking—are just exercises of power that are particularly salient for us. Other, less exciting, manifestations are constantly occurring. The way things are, however they are, is really the combinatorial upshot of all the powers in play. The relatively peaceful world we sometimes contemplate is in reality a shunted gridlock of exerting complimentary powers: a perfect, perfectly tense, stalemate. 11 And once we have recognised the role that the ongoing activity of powers plays in preserving this status quo, we ought not so easily to imagine that powers exist without exerting themselves. Powers are in fact always doing all that they can do, just not with all possible effects they could have. We need therefore to distinguish between the exertion of powers, and their manifestation. Powers are always exerting: they are always doing all that they can do, and bringing a ‘stimulus’ near to them occasions no change in them. But not all exertion causes manifestation. If Ed inhabits an empty universe, there is no manifestation that can occur; nonetheless it is negatively charged, actively so, all the while. In the actual world, the exertions of powers do not always cause manifestations that we recognise as such: as with the vase being held quietly together on the mantelpiece, or the table I am currently writing upon remaining in one piece. We can distinguish the exertion of a power, the activity of its powerful nature, from its ability in a given context to cause anything to happen. If there is nothing sufficiently nearby then the power, for all that it exerts, will not make anything happen. On the other hand a power’s exertion may cause things to stay the same. This, technically, is a form of manifestation. But we usually reserve the term ‘manifestation’ for effects of powers that are salient to us, such as vases breaking, not for effects which merely keep the peace. As a corollary of this point about manifestations, we must break another plank in the orthodoxy concerning powers: the idea that ‘a disposition has a characteristic manifestation. Some event-type is associated with a disposition that occurs when the disposition is ‘triggered’12 In fact, for a given power, since it is on the present conception always doing as much as it can do (in its own right), which particular manifestation manifests at a time depends almost entirely on what context the power is in, what other things it is interacting with, what the laws of nature are,13 and so on. The potential manifestations of a given power are, in reality, practically never-ending, due to the variety of other powerful things with 10 11 Williams 2005 is alert to this point. See again de Quincey. 12 McKitrick 2003:157. 13 Though I leave open that these laws are fixed by the natures of the powers that exist. 4 which it could interact. We therefore cannot, with an attitude of appropriate ontological seriousness, type a given power-nature existing within an object with reference to a single effect it happens to cause in situations that happen to be familiar to us (such as causing a vase to break on collision). Nothing exists in objects which corresponds so closely, in its own nature, with such a narrow pattern of effects, no matter how salient these effects are for us. Of course Place is correct that ‘We invariably characterize a [power] in terms of its manifestations’14. But this constitutes a claim about our classificatory practices, not about the nature of the properties these practices aim at. That nature far outstrips the ability to cause any particular effect. This is not to say that powers are not identified by what effects they would cause: A power may be individuated by its total causal profile, by its overall contribution to causation. Powers and Triggering Heil reports: ‘Philosophical lore has it that a disposition is ‘triggered’ by a ‘stimulus’, and the result is a manifestation of the disposition.’15 Philosophical lore is mistaken. Powers are always ‘on’, they are always doing all they can do in their own right. They will not always have their possible effects: they have effects according to what other powers are around for them to interact with. Thus there is no such thing as the triggering of a power, in the sense of some occurrence that makes the power go from an ‘inactive’ to an ‘active’ state. The power, where it exists, is always exerting: like a running motor, it is constantly straining to cause all it, in its own right, can cause. Whether it has any of the effects at its disposal depends on whether suitable partners for interaction enter its ambit. In a similar way, an engine that is running needs to be put into gear, or attached to axles, to make anything go. The concept of ‘triggering’ belongs to the mistaken, mystery-inducing, conception of powers as dormant between manifestations: On the conventional conception a vase’s fragility has two modes: dormant and active. The fragility only becomes active, causally efficacious, when the vase hits the floor. Two difficulties lurk here. First, what is fragility’s form of being in an undropped vase?16 Second, what is it that makes fragility ‘react’ to the vase-dropping, how is this power ‘activated’? Because fragility is conventionally conceived of in such close relation to its manifestation of breaking, the first difficulty is especially pressing. All we are able to contemplate, in the absence of a vase actually breaking, is an absence of breaking. But in what sense is an absence of breaking a case of being fragile? The conventional view is thus at risk of letting slip of the important truth that the vase is fragile whether dropped or not. Natural reactions are to relegate fragility to an inert status supervenient upon a categorical base—then fragility never activates, or makes anything happen—or to say the nature of powers is exhausted by the relevant conditionals concerning their manifestations. This latter approach has never quite overcome the problems of finks17 and masks.18 The former relegates powers to mere figures of speech: they do not do anything, they are ontological good-for-nothings. 19 An answer to the second difficulty, for its part, might seem to require that fragility have the power to activate on collision, which power also needs the further power to activate just in time…and so on. Hence we face an unwelcome and implausible, perhaps infinite, proliferation of 14 15 1999:226. 2005: 350. 16 Apologies to the reader for some repetition over the next half-paragraph. But the issue of powers’ nature between manifestations is both sufficiently serious, and closely connected to the puzzle around triggering, that it bears re-stating here. 17 Martin 1994. 18 Bird 1998. 19 Prior Pargetter and Jackson (1982) mount an influential defence of this view. 5 ‘powers’, whatever those are (for we have not settled the first difficulty). Nor would such higher-order powers help to explain what triggering is; each power to activate only passes the metaphysical buck to its descendant. We simply do not understand what it means for a power to go from an inactive to an active state, nor how this might be achieved. Consequently these mysterious ‘triggering’ events only serve to clutter our ontology: Any theory of powers requires interactions between the powers of different objects. The conventional dualist view additionally requires first the triggering of a power so that it might interact efficaciously. This obscure extra posit does not earn its ontological keep. MA posits properties, active powers, which defy the categorical/dispositional dichotomy. Categorical properties are understood to be occurrent—fully present from moment to moment —but also inert: they don’t make things happen, except perhaps through realising some disposition on a given occasion.20 Contrastingly, dispositional properties are seen as active and as making things happen, but also as fleeting affairs that only ‘activate’—come into full being—from time to time, and perhaps never (for dispositions allegedly need never manifest). But away from these troublesome conceptions we can imagine a property both powerful and occurrent: it would have to be powerful—exerting itself—as a matter of its ongoing existence, like a muscle constantly flexed. This is the sort of property we must posit in fundamental ontology if we are to understand the powers of things. Such powers are never genuinely ‘triggered’; at best they are ‘put into gear’ by the proximity of suitable co-actors. Unfortunately this conception runs the unavoidable risk of satisfying no side in the debate. To the dispositional essentialist I might seem to be eliminating genuine dispositions, or at best reducing them to something else; while to the reductionist I’ll likely appear to be smuggling in dispositions—these active powerful properties—under another name. But hopefully there’ll be some, perhaps those with nothing particularly at stake in the discussion, who will detect the appeal of the present alternative. Power Semantics Ed is charged; this is the ongoing, active way it is with Ed. In terms of power-talk, we say Ed has a power to repel other electrons. But Ed also has, clearly, a power to attract protons. We have now ascribed two powers to Ed. What, we may ask, do these two power ascriptions pick out? In the act of ascribing two powers, must we be ascribing two distinct ontological features to Ed, one corresponding to each ascription? In reality it seems that what there is in Ed corresponding to the two power ascriptions is simply the single property of being negatively charged—this power active within Ed as an ongoing affair. It is Ed’s unit of negative charge that causes other electrons to be repelled (in conjunction with their own negative charges) and it is similarly Ed’s negative charge which causes (or would cause) protons to be attracted to it (in conjunction with their positive charges). Thus underlying the two powers we ascribed is really just a single ontological feature. What is the relationship between the two power ascriptions and this single active power property of Ed? When we say Ed has a power to repel other electrons we must, one way or another, be picking out the charge property that actually does (or would do) the repelling, that much is clear. I suggest we are using the known possible manifest effects of the charge as contingent reference-fixers for the charge-power itself. The charge is all set up, so long as the world co-operates suitably, to repel another electron. That it has the power to repel another electron can be read as saying that there is something within the nature of Ed which is striving to repel electrons. That same something can have other effects (manifestations), such 20 Or in concert with natural laws, as on Armstrong’s view. 6 as attracting a proton. So, all told, the two power ascriptions refer to different aspects of the one charge property, namely to the fact that it is set up on the one hand so as to repel other electrons and, by the same token, by the same nature, to attract protons. This is the way it is throughout the natural world: the many powers we ascribe turn out to pick out many fewer underlying natural (‘sparse’) properties via reference to their possible (salient) effects.21 These effects are contingent in the sense that they may never happen, for the powerful object may never see suitable circumstances wherein its power can have the effect it would have. But the effects are necessary in the sense that given the setup favourable to the effective exercise of the power, the effects must happen of metaphysical necessity. Micro/Macro-Powers and Change What is the relationship between micro- and macro-level powers? In particular, what is the relationship between the powers of the fundamental material components of a macro-object and that object’s higher-level powers? The powers of macro-objects are due to the interlocking powers of their constituents. Consider a fragile vase. The vase is composed of basic particles with properties like spin, charge and mass. On MA, these are ever-exerting powers. The particles are associated with each other, forming atoms, which in turn form molecules. The flow of powers from micro to macro is, as Heil suggests,22 combinatorial. Micro-powers interact, bonding particles into atoms with their own powers. These atoms likewise bond into molecules, which in turn are bonded to form the vase. On the hypothesis that the ultimate particles are of one or a limited number of kinds, the powerful upshot of their combination is due to their arrangement: different configurations of ultimates will yield different larger confluences of power. The upshot of the micro-to-macro confluence of powers in the case of our vase is that the vase has an (actively sustained) irregular pattern of molecular bonding. That is why when the vase suffers a minor fall onto the wooden floor, it breaks. So long as the vase’s micro-structural arrangement stays relevantly the same, it will remain fragile. Fragility has a dispositional appearance because the manifestation salient for us, breaking, need never happen. At times other than manifestation, the fragility may appear to lie dormant within the vase: it doesn’t seem to be doing anything. But when we turn to the vase’s lowerlevel constitution we reveal this appearance as deceptive. The vase’s fragility is constituted by the bonds between its molecules, their motion, and ultimately by the charges of the particles composing its atoms. But the taut circumstance of the molecules being bonded, and of the basic particles going on being charged, is a perfectly active state of affairs featuring the ongoing exercise of powerful properties. In just the same way, when we play ring-a-roses, our circle of friends is held together by each of us actively holding the hands of those either side of us. Thus the fragility of the macro-object, although it doesn’t seem so from the macroperspective, is a perfectly occurrent, ongoing and active property. It is a Megaric power. There is nothing merely potential, merely dispositional, about it. All that is merely potential are certain of its effects, inter alia the manifestation of breaking. It has these when in the presence of suitably powerful co-actors. On the present account the powers of fundamental objects such as electrons are fully-active properties, they are exerted in an ongoing way as part of their nature. Moreover, electrons being what they are, an electron’s charge is a power that is active throughout the life of the 21 The model is Kripke’s (1980) account of our reference to natural kinds via their surface features, such as appearance. 22 2005, Chapter 11. 7 electron, for as long as it exists and in constant fashion. But it is clear that many powers of macro-objects, by contrast, can be gained or lost during those objects’ lifetimes. If the powers of a macro-object are simply those due to its complex of powerful constituents, how can we explain the fact that macro-powers come and go, while micro-powers are forever ‘switched on’ and invariant? At least one macro-power appears to be invariant: mass. As Molnar says ‘Massive objects are constantly manifesting their gravitational power in continuous interaction with spacetime.’23 On the present understanding a macro-object’s mass is constituted by the masses of its basic parts. These micro-masses, like any power on MA, are constantly exerting themselves. As regards the mass of the macro-object, it matters not what happens to the parts, how they twist and turn, or are aligned, still the confluence of these mass-powers is the macro-mass of the whole, as long as the whole exists. This is a case of an invariant macropower: mass cannot be (in Molnar’s phrase) ‘switched off’. The fragility of a vase, or the solubility of salt, will not vary so long as the micro-structure of the composite persists. By contrast, no mere re-configuration of an object’s micro-structure can rid it of mass, although the object’s determinate mass can thereby be altered. Comings-and-goings of macro-powers are of two sorts. Some are due to external action. Consider Ashwell’s nail: ‘the nail, wrapped in bubble-wrap, is immersed in a canister of liquid nitrogen, which is being constantly shaken. If the nail were not wrapped, it would break, as the liquid nitrogen would render it fragile so that it would break when it hits the side of the container. The wrap protects the nail from breaking, and yet the nail is not fragile.’24 The nail is not fragile, but would become so were the bubble-wrap removed. How? The answer is that the nitrogen would alter the arrangement of the nail’s parts, by chemically reacting with it. This chemical reaction would occur thanks to the powers of the nitrogen and the powers of the nail that are presently exerting. It is the present exertion of these powers which means that we require bubble-wrap to stop the nail disintegrating. If the wrap is removed, and these active powers are allowed to interact, the nail’s structure will change. The new structural configuration, after reacting in nitrogen, is weaker than previously. That is on account of the new powerful interactions of the nail’s parts, re-configured by reacting with the nitrogen, and the effect (breaking) this new interaction would cause were the nail to collide with the container. Conventionally we would say the nail is disposed to become fragile in nitrogen. The truthmaker for this claim, however, is nothing other than the occurrent, ever-exerted, powers of the nail (thanks to those of its parts). In truth, the nail has the power to acquire the power of fragility,25 which power is fully realized and active. I say an object has the power to 0 when, given the right interaction partners, it could 0 without any prior intrinsic change in it.26 The nail does not have the power of fragility before reacting in the nitrogen, since intrinsic changes are needed in its structure to get it into a state where subsequent collision with the container would break it. By contrast, even while in the bubble-wrap the nail has the power to 23 24 2003:86-87. 2010:10. 25 This is one of Molnar’s ‘iterated’ powers (2003:33). 26 This corresponds to Martin’s idea that powers are ‘ready to go’. 8 acquire fragility, thanks to its actively powerful nature. When we know what changes would be produced in an object by some external action upon it—when we know how its parts would re-configure and/or what new powers would thereby be conferred—we tend to speak of dispositions. But there is nothing truly dispositional that exists, in such cases. What exist are the occurrent micro-powers of an object’s parts, which would have new upshots after a reconfiguring interaction with the outside. There is no disposition of the nail which is ‘activated’ after reacting in nitrogen. Instead the nail develops a new active power, thanks to the reconfiguration of its corporation of parts. The truthmakers for disposition-talk are perfectly actual states of affairs. What is non-actual are some of the effects powers would have. The second way a macro-object’s powers can change is through internal evolution. Though the powers of an object’s constituents remain constant (and constantly active), the overall powerful upshot for the composite can change, as these powers interact with one another, causing the whole to evolve. Setting aside the effect of outside influences, the macro-object can be seen, with its micro-composition in mind, as a developing confederate in a state of ongoing flux, as its micro-parts ‘work out’ their powers with respect to one another. Perhaps the whole will come eventually to a state of relative rest, during which its powers remain fairly constant. More likely it will be forever shifting in its properties, as least in a low-level manner. Not so its ultimate constituents, however, the electrons and suchlike: these continue to exercise their basic and defining powers in unfaltering fashion. But their changing configuration, caused by the interaction of their powers, will vary the powers of the whole they compose. In practice these two modes of macro-power change operate in parallel: consider the changing powers of an icecube as its powerful parts interact with the surrounding air, and with each other, while it melts into the puddle that makes me slip when I walk by. Most powers of composites can be lost, and new powers gained, through change in their structure. The ultimate material constituents, however, have their powers forever ‘switched on’. In both cases, where a power exists it is always exerting. The Categorical/Dispositional Identity Some authors27 come tantalisingly close to the correct view of powers by proposing an identity between dispositional and categorical properties. This is not the doctrine which attempts to reduce dispositions to mere categorical properties.28 Rather this view asserts a genuine identity, i.e. a unity on equal footing, of the putative pair of property natures. But the identity claim, and the way these authors explain it, have the unfortunate air of a mystical meditation. Consider this formulation by Heil: ‘Properties are not ‘pure powers’. More likely, every property is at once dispositional and qualitative’29 How are we to understand the possession by one thing, simultaneously, of hopelessly opposed modes of being? Recall that the qualitative/categorical is understood as the occurrent 27 28 Heil (e.g. 2003, 2005), C.B Martin (e.g. 1994, 1996), Strawson (2008). Associated primarily with Armstrong (see e.g. his 1996). 29 2005:346. 9 but inert, and the dispositional as the dormant but (fleetingly) efficacious. I foresee only two outcomes of attempting to embrace Heil’s identity claim—either we swallow something unintelligible, or we force the statement into a more familiar conceptual shape and endorse a form of property dualism. The first option has little to recommend it. A genuine claim of unity would require us to grasp how a property could be dispositional and qualitative in just the same respect and to just the same extent: dispositional in so far as it was qualitative. But we have no grip on what this could mean. Alternatively, we are to understand that powers have two aspects, a categorical/qualitative aspect and a powerful/dispositional aspect. But this is not a categorical/dispositional identity, any more than the view that brains have irreducibly distinct mental and physical aspects establishes a mental/physical identity. This is really dualism: property dualism, to be precise. But clearly a dualism of aspects of properties, of a categorical as well as a dispositional aspect to properties, is no better than a dualism of first-order properties. We may as well say that objects have categorical as well as (distinct) dispositional properties. Instead of struggling to understand how the ongoing and inert, the categorical, could be the same as the potential and fleetingly active, the dispositional, I suggest we relinquish the notions at either end of this conceptual spectrum and try to make sense of something firmly in the middle. What is called for, in order for us to grasp the genuinely unitary nature of powers, is the notion of property-natures that are occurrent, but also perfectly active all the while. This is the conception MA recommends. Then it can truly be said that just in so far as a power is potent (dispositional), so it is occurrent (categorical). Multiple-Realisability Many different kinds of thing are fragile. What to say about the relationship between the variety of perfectly active powers within different objects that can cause them to easily disintegrate and the property they apparently share of being fragile? One approach to avoid is functionalism: identifying fragility with the property of havingsome-property-that-ensures-easy-disintegration.30 That renders fragility a power-less power: anything it might cause is effected by its realiser in a given case. Heil is correct that ‘This is most odd. In an effort to make sense of causal powers…Jackson and his colleagues posit dispositions as higher-level properties. Having introduced these properties, they then express amazement that anyone could imagine that such properties might do anything.’31 What about identifying fragility with the realiser in each case? Then fragility would be: having this realiser, or that, or that other one, picked out by the rough functional profile they all fill. That’s passable as a solution, except for making fragility a mongrel property, with nothing clearly enough in common between instances. Heil maintains that the predicate ‘fragile’ fails to pick out a kind; instead instances of fragility have at most a family resemblance. This would appear tantamount to elimination: it transpires that there is no such ‘thing’ as fragility, strictly speaking, for Heil. In the end I am happy with this result. Some objects (vases, paper castles, soap bubbles) are composed by micro-parts whose combining, ever-interacting powers mean their wholes disintegrate easily. There is at most a family resemblance between the summative powers of each object-type; the objects fit a broad behavioural profile. But the sets of (fully active) micro-powers whose confluence has this upshot are likely to be of quite different configurations in each case. Hence the summative 30 31 See Prior, Pargetter and Jackson 1982. 2005:349-350. 10 powers are really nothing alike. Things differ at the lowest level of ontology. It seems that if the basic constituents of matter are of one kind, or a limited number of kinds, then the powers of these entities straightforwardly belong to types, causally individuated. But quite different arrangements of such ultimates can have, through the confluence of micro-powers, what are to our eyes sufficiently similar macroscopic upshots in familiar situations that we apply one predicate to them. It follows that where two complexes of ultimates are sufficiently similarly configured, the same confluence of micro-powers can result; these complexes therefore share macropowers. Examples might be chemicals of the same kind, or duplicate fragile vases. Intrinsicness Are powers intrinsic to their bearers? Molnar claims intrinsicness ‘is one of the crucial appearances which has to be saved by an analysis’.32 Megaric powers are best construed as intrinsic, since in constantly exerting themselves they fully express their natures whether or not there is anything for their bearers to interact with. Therefore, in this section I attempt to defuse some purported counterexamples to the intrinsicness thesis, due to McKitrick.33 Shoemaker’s Key Adapting an example from Shoemaker,34 McKitrick considers a key with the power to open Shoemaker’s door. What allegedly reveals this as an extrinsic power is that if we change the lock on Shoemaker’s door the key can no longer open it. Without intrinsically altering the key, we appear to have rid it of one of its powers. We might reply that the power to open Shoemaker’s door is not a genuine property. That is Shoemaker’s conclusion. But this reply is strained, since it seems perfectly straightforward to say the key has the power to open Shoemaker’s door, and we do appear to ascribe some feature to the key in virtue of which it can achieve this result, some property or other. Another avenue for the defender of intrinsicness is to claim that the relevant intrinsic power is the power to open a type of lock, and this is a power the key retains even if the lock on Shoemaker’s door changes. The key may no longer be able to open Shoemaker’s door, but it retains the power to open this certain type of lock, no longer tokened on Shoemaker’s door as it happens. There is something to this reply, but by changing our focus to the lock it ignores the simple truth that the key—before changing the lock—has the power to open Shoemaker’s door, not just to undo that sort of lock. It may also have the power to open that sort of lock, but that doesn’t change the fact that it is, in a different sense, able to open Shoemaker’s door. A better response, I think, is to recognise that the statement ‘the key has the power to open Shoemaker’s door’ implicitly refers to the state of the door at the time the statement is made. The claim is that, within certain rough parameters of relevant change, the key has the power to open Shoemaker’s door as it currently is, where this reference to the current state of the door includes its lock having a certain configuration. Now the door can of course survive the change in lock. So, once the lock is changed, it is true to say in one sense that the key does not now have the power to open Shoemaker’s door. But that does not mean the key has lost any power: for it retains the power to open Shoemaker’s door in its past state. This is confirmed if we replace the original lock: the key will once again open the door. Rather than say the power has miraculously re-appeared, we can simply say it was there all along, when relevantly specified: Given the state of the key—its material configuration—it was geared to 32 33 1999: 3. McKitrick 2003. 34 Shoemaker 1980. 11 opening the door given a certain state of that door. Now that the state of the door is reinstated, the key can again do its work. The lock-changing case is really no different in kind, only in degree, from that of transferring the key from our world into an empty universe. We can imagine a series of intermediate stages between the key existing in our world with the power to open Shoemaker’s door as it currently is, and its existing in the empty universe. First we could remove the lock from Shoemaker’s door which the key is able to open, then in the next case destroy the lock, then in the next case destroy the whole class of locks which the key could open, thus leaving it without any locks it could open. And so on. But, if we moved the key from our world where Shoemaker’s door was in its original configuration straight into the empty universe, we would not say the key loses its power to open the door just because it is suddenly made lonesome. It’s simply that said door is no longer around to be acted upon in this case. If it were true that removing the powerful object from all other objects it could possibly act upon altered its powers, then only those powers that concerned an object’s effects on itself could be intrinsic. But everyone agrees that at least some interactive powers are intrinsic. Therefore simply removing the objects with which a power could interact does not remove the power. But the case of changing the lock on Shoemaker’s door is best seen as a case of removing the relevant object. True, as we’ve noted, Shoemaker’s door can survive the change in lock. However once this change has occurred it is no longer the object the key has the power to open. The key has the intrinsic power to open—and retains this power—the door in its former state. To think otherwise is to succumb to a linguistic trick. Weight Adapting an example from Yablo,35 McKitrick argues that weight, an object’s power to depress scales, changes from environment to environment. For example, intrinsic duplicates give different scale readings on Earth and on the Moon. Hence weight is an extrinsic power. Response: Weight is not really a power (as opposed to mass, which is intrinsic). Your weight is just what you read on the scales, and of course that can change from environment to environment, according to gravity. But you, the object to which the putative power belongs, have no more power to depress scales on the Moon than you do on Earth: all that happens is that it’s easier to depress scales on the Moon. Parallel case: If, while arm-wrestling you, I start pushing less hard (say I decide to let you win) that doesn’t mean you become any stronger. Closer parallel: If I put you on easier-to-depress scales on Earth, that doesn’t indicate anything about your powers. McKitrick’s examples of powers, as well as plausibly being extrinsic, have to belong to the objects in question (for they have to be extrinsic to these possessor objects). But the present case is best construed as one where there is no power both belonging to the relevant object and extrinsic. What belongs to you is your mass, and that is an intrinsic power. The Bucket of Water A bucket of water has the power to dissolve the contents of my pocket, given that it contains sugar cubes. But with the contents of my pocket changed to copper coins, the water can no longer dissolve the contents of my pocket. Hence, concludes McKitrick, the power to dissolve the contents of my pocket is extrinsic. What is the natural property of a bucket of water, we may ask, in virtue of which we ascribe to it the ‘power to dissolve the contents of my pocket’ given that I have sugar cubes in my 35 Yablo 1999. 12 pocket? It is the molecular structure of the water, this vibrating, powerful, ever-tense assembly of molecules. Does the water possess the (multifarious) powers conferred on it by this structure even while we change the contents of my pocket (to copper coins, as it may be)? Of course it does. Vulnerability Joan is vulnerable walking alone in Central Park, but not with her bodyguards, so her being vulnerable (easily damaged) is allegedly extrinsic. But this is not so, for the reason Joan has her bodyguards is because she is vulnerable; she herself remains so even when surrounded by them. That’s precisely why she needs bodyguards! One has bodyguards on account of one’s ongoing vulnerability.36 By contrast, the system of Joan plus her bodyguards is not vulnerable (to attack, by nasty park-dwellers), and is intrinsically so. There are two ascriptions of ‘is vulnerable’ in play: one concerns the core object, one concerns this object plus whatever defences it clothes itself in (‘I’m vulnerable to arrows, but with my armour I’m less so’). Both attribute intrinsic powers. Visibility Whether an object is visible changes with lighting conditions and the presence of obstructions, says McKitrick. Hence visibility is extrinsic to its possessors. But this example trades on two senses of ‘visible’: the sense in which something could be seen given the right conditions, and the sense in which it is or is not in fact seen presently; i.e. the sense in which ‘Is it visible?’ just means ‘Do you see it?’ The stone in the darkness is not currently visible: it is not seen, because of the darkness. This, however, is not a dispositional but a categorical matter. The stone remains a visible object in this situation in the sense that its structure is actively such as to reflect and suitably modify light rays, where light rays are available to be so reflected and modified. This object continuously makes the relevant contribution to being seen: hence it is see-able, i.e. visible, and intrinsically so. Recognisability Bill Clinton is recognisable (has the power to be easily recognised), but not intrinsically so, claims McKitrick, because his being recognised depends on facts about other people (that they know him, aren’t terribly short-sighted etc.). Recognisability is not intrinsic, I admit, but it’s not a power of the object either. It’s nothing really to do with some person that she is recognisable. Saying Bill Clinton is recognisable is just shorthand for saying lots of people have the power to recognise him. And these powers of people are intrinsic to them: were they transported, brain-traces and all, to the empty universe, they would retain the power to recognise Bill Clinton. Perhaps there is a sense of ‘recognisable’ on which it properly belongs to an object: If for instance I have an enormous banana-shaped nose then I am recognisable indeed: I’m likely to leave a lasting impression on people who see me. But that power is intrinsic to me: my power to make a striking impression on onlookers goes with me wherever I wander, and persists for as long as I have the amazing nose.37 McKitrick replies to this line of thought thus: ‘When one says that Bill Clinton is recognisable, one is saying something about Bill Clinton.’38 That’s undoubtedly so; but it is consistent with the thought that what we are saying about Bill Clinton is that other people 36 Compare Mellor (1974:173), discussing the precautions taken to prevent nuclear fuel from exploding: ‘It is ridiculous to say that their success robs the fuel of its explosive disposition and thus the precautions of their point’. 37 C.f. Cyrano de Bergerac. 38 2003:XXXX. 13 recognise him easily, that they have a power regarding him. It doesn’t follow that we are ascribing a property to Bill Clinton in anything but a loose, and metaphysically uninteresting, sense. Some apparent property ascriptions really say something primarily about things other than the object mentioned: ‘Bill Clinton is desired’, if it ascribes a property, says that things other than Bill Clinton have an attitude towards him. Having addressed McKitrick’s attempted counterexamples, I submit that Molnar’s intrinsicness-intuition stands undefeated. The Conditional Analysis Having concluded our metaphysical examination of powers, we may now say something useful about power ascriptions. In light of our conception of powers as perfectly occurrent and active, I suggest the following understanding of the ascription of a power to an object: There is some intrinsic property(s) of x which, if given free rein, would cause x to 0 in conjunction with y and z. The key component is the conditional phrase ‘if given free rein’. The phrase reflects the fact that powers are to be conceived of as always exerting themselves, always doing their utmost to make effects happen (they are doing no more when an effect actually occurs, it is just that worldly circumstances have become friendly to the effect’s production). If an antidote or fink is present, then the power will not be given free rein: either it will be blocked, or subverted by the very circumstance wherein it seeks to express itself manifestly. So this way of ascribing powers avoids the problems antidotes and finks have posed. Interestingly, this understanding of power ascriptions seems to bear close relation to Mellor’s account. Of fragility Mellor says ‘Is not a’s fragility the property that is part of what makes it true that, if a were stressed without changing this property, it would break?’39 II. Objections to Megaric Actualism Franklin’s Objection The attempt to dispose of dispositions is nothing new. In particular, there has been a project to reduce dispositional properties to, or one might better say to eliminate them in favour of, categorical properties. This project faces a formidable objection due to Franklin.40 Franklin considers Democritus’ claim that the solidity of objects—their power to resist penetration—is due to atoms having hooks, causing them to stick to one another, and the whole they form to be solid as a result. Franklin notes that it is not enough for atoms to have hooks, they must also retain their hooked shape on contact with other atoms. However the tendency to retain shape is a disposition. Hence Democritus has failed to give a disposition-free account of solidity. Generalising from the example Franklin says ‘A disposition is a disposition to behave in certain ways when certain causes act. If the categorical properties or structures of a thing are to produce this behaviour under these causes, the thing must retain the properties or structures despite the action of the cause…But this rigidity of structures, their tendency to be preserved, is dispositional. Hence dispositionality has not been eliminated.’41 39 40 2000:767, emphasis mine. Franklin 1986. 41 1986:62-63. 14 This ingenious objection to the attempt to reduce dispositions to categorical properties does not apply to MA. MA’s active powers are not inert categorical properties. If Ed has the property of going on being actively charged, there is no need for it to have additionally the disposition to retain this charge. What ensures Ed is charged at the crucial moment, and will repel another electron in its path, is just the fact that it has the property of going on being actively charged. The only way there could be a question over whether Ed retained its charge when entering into a transaction with another particle would be if it did not really have this property of going on being actively charged. On this view, the required ‘maintenance aspect’ of powers which Franklin highlights is covered within the property nature itself. But cannot Franklin’s objection just recur? Even if it has the property of going on being actively charged, doesn’t Ed also need the disposition to retain this property, on pain of perhaps losing it at the crucial moment? But this cannot be a fair reply. For note that the same move could be made against Franklin. Suppose that on his account the fundamental truth about Ed is that it is disposed to repel other electrons; this is what its being charged comes to. Still, we could imagine that Ed runs the risk of losing this disposition at the instant of interaction with another electron. To prevent this, we would seem to have to add to it the disposition to retain its disposition to repel electrons. But then the difficulty reappears, and we face the need to add a further disposition (to retain the disposition to retain the disposition to repel electrons). And so on, without satisfactory end. The bottom line is that the buck must stop somewhere. And once we have said that Ed has the property of going on being actively charged, we ensure the same level of property-robustness in the face of oncoming interactions as would be provided by attributing to Ed the disposition to repel electrons. What, finally, about the ability to interact, must not every basic particle have this disposition, on pain of there being no causation in our world? 42 The answer to this objection is that the interactivity of particles is already ensured by their properties, without needing to be separately legislated for. Since Ed has the property of going on being actively charged, it has the power (always active) to repel other electrons; which is, inter alia, an ability to interact. The ability to interact, in other words, is already included in the perfectly active (i.e. nondispositional) properties of Ed—these properties, by their nature, are interactive—and it would be ontologically superfluous to add this general disposition to Ed’s profile as well. The Objection from Science Thompson observes that in quantum mechanics properties ‘are not continuously definite, but only have specific values in suitable situations such as measurement interactions of certain kinds.’43 Thus such properties appear to be dispositional par excellence. In reply we can note that Thompson’s interpretation of quantum mechanics is non-compulsory. We needn’t take it that a measurement genuinely affects—makes determinate—the state of an electron, which state has existed hitherto in a merely dispositional, shadowy, manner. On the Everett interpretation, every possibility that could show up in measurement is in fact realised, it is just that we the measurers become correlated (entangled) with only one of these outcomes.44 Thus it appears to us that the electron has really been narrowed down by our measurement, to the determinate of the variable we measure. But in reality the electron (or its counterparts— there is leeway on how we read this) exists in every possible state, and corresponding to each of these possible state-measurements is a distinct conscious state of measurement associated with the experimenter. On this conception of the quantum world, we are not forced to give up 42 43 Thompson 1988 argues in this way that dispositionality cannot be eliminated from ontology. 1988: 75-76. 44 I have particularly in mind Lockwood’s (1989) sensible reading of Everett. 15 the idea that the powers of the electron are occurrent and active, perfectly real at each point of time, as opposed to dispositional. I am not here arguing for the Everett interpretation, although it seems clear that the defender of MA will find it congenial. The point is just that quantum mechanics does not provide incontrovertible support for the irreducibility of dispositions (as opposed to Megaric powers), for an interpretation of the data is available that fits MA’s disposition-free ontology. Molnar’s Objections MA has simplified and de-mystified our understanding of powers. But this will be for naught if the position faces decisive objections. Molnar claims that it does.45 Molnar advances three objections: First: MA implies that something which once 0-ed and will 0 again loses its power to 0 between these exercises, miraculously regaining it at the time of the second exercise. Second: MA can’t distinguish between unexercised and absent powers: Two babies who never see the light may differ in that one is sighted while the other is blind even though neither ever sees. Third: The powers MA posits result from ‘metaphysical double vision’. Molnar says ‘It is hard to understand the concept of a power that exists precisely when it manifests and for exactly as long as it manifests. Is such a power…a property in its own right?’46 Such a property might just seem to be a reification of the effects of powers. What is essentially actual in its own nature is the effect. To say there is an always-actualised power as well is to count the effect twice, in effect. Further, since their existence is parasitic on their effects, so conceived, Megaric powers cannot be used in explanation of their effects. This is unacceptable, since such explanation is exactly what we want powers for. These are challenging objections, but I believe they can be answered by developing the framework we have put in place in earlier discussion. Consider an example to illustrate objection one. A catapult consisting of a rubber band and two pins is used to launch a ping-pong ball: the ball is positioned, the rubber band is stretched back, then released, and the ball goes flying. The catapult is then left alone for a day, before being used to launch another ball. Since on MA ‘At time t an object has the power to 0 iff it exercises the power to 0 at t’,47 it seems to follow that while not launching balls the catapult lacks the power to launch balls, which power it mysteriously regains when the time comes for the second launch. Mysterious losses and reacquisitions of powers are exactly the sorts of events MA is formulated to avoid: the Megaric Actualist is suspicious of the orthodox doctrine that powers lie ‘dormant’ and are somehow ‘activated’ by appropriate triggers. It is important to note that there is only a terminological hair’s breadth between the absurdity Molnar lays at MA’s door with objection one, and the orthodox doctrine. For what the orthodox doctrine asserts is that powers exist powerfully at some times, then cease to exist in this way, mysteriously reacquiring this state in the presence of so-called ‘stimuli’—why and how the stimuli are stimuli, we know not. The Megaric Actualist will wish to distance himself from the absurdity, either by explaining an object’s possession of an unmanifesting 45 Prior 1985 also advances objections. Psillos 2006 considers and rejects MA in very perfunctory fashion. Molnar’s case against MA is the fullest and most sophisticated available, for that reason I concentrate on it. 46 2003:97. 47 Molnar 2003:95. 16 power in terms of some occurrent powerful activity that persists between manifestations, or by giving a well-motivated explanation of the object’s loss and reacquisition of the power. Our earlier account of a vase’s fragility is in the first category. The fragility is realised by the micro-structure of the vase, which is a matter of actively exercised bonds between its molecules. As long as the vase retains its structure, it remains fragile in an actively exerted manner. This is so quite outside of circumstances which provoke the manifestation of breaking. The vase constantly has the power to break when lightly stressed, since thanks to the confluence of the active powers of its constituents it is primed for this manifestation without requiring prior intrinsic change. The catapult’s power to launch balls fits into the second category. It is the rubber band’s elasticity—its power to return to its initial shape after deformation—that is responsible for the catapult launching balls.48 This elasticity, on MA, is a matter of the active bonding (‘crosslinks’) between the long, chain-like polymer molecules composing the band, plus the constant motion of these molecules (at room temperature) which causes them to curl up. These molecules are (actively) bonded in such a way that they can be deformed considerably and still their bonding, plus their curling motion, will pull them back to a state whereby the band is relaxed. Now, elasticity is really comprised of a pair of powers, one of which produces the other. First, the rubber band has the power to significantly stretch without breaking. This is thanks to the fact that its molecules are curled up when relaxed and can be stretched out, plus the matter of the bonding between them which means the band does not break when it is stretched in this way. Compare this manifestation of the band’s power with the manifestation of a vase breaking when a different sort of stress is applied to it. Both manifestations are thanks to the constantly exerted powers involved in the interactions between the micro-parts composing these objects. Once stretched a rubber band is like a prisoner struggling to escape his chains: stretching the polymer molecules causes an increase in energy, and thereby an increase in the motion of these molecules, which in turn causes them to try to curl back up into a relaxed state. At this point the band has acquired the power to snap back into shape. It lacks this power while relaxed, since intrinsic change is needed to its structure—stretching—to put it in a position where it may snap back into shape without further prior intrinsic change to it. In turn, the power to snap back into shape—constantly exerted for as long as the band is held stretched— realises the power to launch balls, since when the former power is allowed to manifest, through the band being released, the discharge of energy involved is what launches any ball placed against the band. Thus when relaxed the band has the power to acquire the power to launch balls, but it does not have this latter power until stretched. When we are aware of the powers conferred on objects by repeatable, reliable manipulations of them, we are prone to ascribe dispositions to them. But the relaxed band does not have a disposition (awaiting ‘activation’) to launch balls. What it has is an actively exercised power—the power to stretch without breaking—which, when it manifests (as stretching), endows the band with a further power—inter alia, to launch balls. Once we consider the ever-exerted powers of the micro-constituents of objects, we can easily understand how re-configuration of these parts (in the case of the band, achieved by 48 The following explanation omits for simplicity that the rubber band needs to be held, in order to be stretched. Being held is an active state of affairs (see the example above of playing ring-a-roses) involving exerting powers by both holding object(s) and held. 17 stretching) causes powers of the composite to be gained and lost, through the ongoing interaction of the parts and the new powerful confluences established by their rearrangement. Other powers of objects, such as the vase’s fragility, exist in active form all the time they do not manifest, so long as the relevant micro-structure persists. Thus MA has available to it an utterly un-mysterious account of what goes on in objects outside of salient manifestations of power. This is in contrast to the orthodox account, which only thinly cloaks a metaphysical mystery in talk of a disposition existing now in an ‘unactivated’, now in an ‘activated’ state, thanks to the presence or not of a ‘stimulus’. These stipulative terms express the puzzle to be explained, but they do not explain it. In response to objection two, the Megaric Actualist will say there exists something ongoing and actively powerful in the mind of one these babies that does not exist in the other. Presumably this difference takes the form of a certain neural configuration, in turn to be explicated in terms of the states of bonding of certain molecules in the baby’s brain, states which we have understood to involve the active exercise of the charge (etc.) of the constituents of these molecules, plus electrical activity. Now it will surely be that placing the sighted baby in the light with its eyes open will cause a reaction in this complex neural state: neurons will communicate, molecules will move and reaction to, and so processing of, light will occur. Many effects of actively exercised powers can only occur given suitable partners for interaction. Imagine the Mary Rose facing the Spanish Armada. The captain orders continuous firing of the cannons: this is the ongoing exercise of a power belonging to the ship. Still, nothing will be the upshot of this exercise until the Mary Rose is faced with an enemy ship, at which point that ship is destroyed. The Mary Rose’s firing power is exercised constantly, but will have no manifestation until an enemy ship comes close enough. In the same way we can consider the relevant configuration of the sighted baby’s brain to be ‘firing’, but with no target to interact with: there is no light to be seen. We can say that the baby has the power to see, in an ongoing active way, where the truthmaker for this statement is the neural activity in the baby’s brain. So the Megaric Actualist can point to ongoing, actively powerful states of affairs within one baby that do not exist in the other, as the ground of one having the power (of sight) which the other lacks.49 The distinction between a power’s exertion and its manifestation allows us to reject the accusation of double-counting, of mistaking a power’s effects for the power itself. A power may exert without having effects observationally salient to us, perhaps without having any effects at all. The Mary Rose may be firing its cannons, but unless there is a target facing the broadside this power’s exercise will not cause anything to happen. So it is not true that on MA a power is conceptually parasitic upon its effects. Such powers as exist are always exercised, but that doesn’t mean they are always efficacious. They are always doing all they can to make things happen, but whether things happen often depends on the alignment of worldly-elements. The idea of a Megaric power that is exercised but not effective to make anything happen is perfectly coherent: whether we label this ‘manifesting’ is a terminological decision. On an epistemological reading of ‘manifestation’, it’s no problem for MA that powers exert without manifesting. On a metaphysical reading, they are always manifesting, we just don’t always notice. What is the intrinsic nature of a Megaric power, the nature we say it always actively expresses whether or not it has an effect? This is something we cannot know directly, for we can only access the nature of a power via the effects it has, and we have said that Megaric powers have a life outside of their effects (thus avoiding the double-counting charge). The 49 An alternative explanation will be on the model of the catapult: the sighted baby may have the (active) power to acquire the power of sight given interaction with light. 18 idea that the world contains intrinsic qualitative natures belonging to its ultimate parts which are beyond our powers to experience is a familiar one, largely thanks to Russell’s reconceptualisation of physics as revelatory of the world’s structure as opposed to its intrinsic character.50 If we posit that these intrinsic qualitative natures, whatever they are like, are essentially actively powerful in an ongoing way (always exerting), then we derive MA. Molnar claims we have no good reason to include Megaric powers in our ontology. The good reason is to make sense of the nature of powers in between manifestations: the answer is that they are just the same in between as at manifestation. This view does without the hard-tounderstand notion of ‘triggering a disposition’, in favour of a conception of the world as an ongoing, self-determining confluence of ever-active power natures Bibliography Aristotle, Metaphysics in The Basic Works of Aristotle, ed. R.McKeon (New York: Random House, 1941). Armstrong, D. M. (1996) Contributions to T. Crane (ed.), Dispositions: A Debate, (London: Routledge). Ashwell, L. (2010) ‘Superficial Dispositionalism’, Australasian Journal of Philosophy Vol. 88, No. 3:1-19. Bird, A. (1998), ‘Dispositions and Antidotes’, Philosophical Quarterly 48: 227-234. Goodman, N. (1954) Fact, Fiction and Forecast (Cambridge, Mass.: Harvard University Press). De Quincey, T. (1822), Confessions of an English Opium Eater. Franklin, J. (1986) ‘Are Dispositions Reducible to Categorical Properties?’, Philosophical Quarterly Vol. 36, No. 142:62-64. Heil, J (2005), ‘Dispositions’, Synthese 144: 343-356. —(2003), From an Ontological Point of View. (Oxford: Clarendon Press.) Kripke, S. (1980) Naming and Necessity (Harvard: Harvard University Press). Lockwood, M. (1989) Mind Brain and the Quantum (Oxford: Oxford University Press). Martin, C.B. (1996), Contributions to Dispositions: A Debate, ed. T. Crane. (London: Routledge). —(1994), ‘Dispositions and Conditionals’, Philosophical Quarterly 44: 1-8. McKitrick (2003), ‘A Case for Extrinsic Dispositions’, Australasian Journal of Philosophy Vol. 81 No. 2:155-174. Mellor, D. H. (2000) ‘The Semantics and Ontology of Dispositions’, Mind Vol. 109: 757780. —(1974) ‘In Defense of Dispositions’, Philosophical Review 83: 157-181. Molnar, G. (2003), Powers: A Study in Metaphysics ed. S. Mumford, (Oxford: Oxford University Press). —(1999) ‘Are Dispositions Reducible?’, Philosophical Quarterly 49:1-17. 50 Russell 1927. (See also Heil 2003:114) 19 Mumford, S. (1998) Dispositions (Oxford: Oxford University Press). Place (1999) ‘Intentionality and the Physical: A Reply to Mumford’, Philosophical Quarterly 49:225-31. Prior, E.W. (1985) Dispositions (Aberdeen: Aberdeen University Press). Prior, E. W., Pargetter, R. and Jackson, F. (1982), ‘Three Theses about Dispositions’, American Philosophical Quarterly 19: 251-257. Psillos, S. (2006) ‘What Do Powers Do When They Are Not Manifested?’, Philosophy and Phenomenological Research, 72: 137–156. Russell, B. (1927) The Analysis of Matter (London: Kegan Paul). Shoemaker, S. (1980), ‘Causality and Properties’, in P. van Inwagen, ed., Time and Cause. Springer. Strawson, G. (2008), ‘The Identity of the Categorical and the Dispositional’, Analysis 68: 271-282. Thompson, I. J. (1988) ‘Real Dispositions in the Physical World’, British Journal for the Philosophy of Science Vol. 39 No. 1: 67-79. Williams, N.E. (2005) ‘Static and Dynamic Dispositions’, Synthese 146, 3: 303-324. Yablo, S. (1999) ‘Intrinsicness’, Philosophical Topics 26:590-627. 20