p28

The heart of the issue is that phenomenal experiences have all sorts of capacities to affect us in various ways. This is most obvious in the case of painful experiences, but it is also clearly the case for bad odors, itches, orgasms, etc. Insofar as experiences have these capacities, then it seems like there needs to be a corresponding subject that is affected by these capacities. Moreover, the ways that experiences affect us are many and varied: the impact that experiences have on us can’t be fully captured with a single dimension of “positive” or “negative”: experiences can be more or less interesting, or vivid, or surprising, or beautiful, or sublime, or nostalgic, or visceral, or disgusting, etc. What we would need for an Edenic quality to be possible is to have an experience that has no capacity to affect us in any way whatsoever, and it is unclear whether there are any examples of such experiences. The case of Edenic color is perhaps the best case for such an example, but on reflection, many of the same things that can be said for other sensory modalities could also be said for color. Consider, for example, being shown an intensely bright white light as soon as you wake up from a deep sleep. Such an experience would be unpleasant (just like sourness and spiciness), and it is moreover unclear where to draw a precise line between such an intense visual experience from other more common visual experiences. Or consider how a visual experience of a beautiful sunset affects you, and how greatly it differs from an experience of uniform grey. Or consider a uniformly white visual field with a flashing bright red circle in the center, and how the red seems to be intrinsically “attention grabbing” in a way that its surroundings are not. The ways that colors affect us are clearly not as striking as the way that (say) placing one’s hand on a hot stove affects us, but it’s unclear whether there is a sharp line to be drawn between these experiences, which is what would 20 be needed to delineate metaphysically possible Edenic qualities from metaphysically impossible ones. In sum, I have argued that neither Question-1 nor Question-2 admits of a precise and non-arbitrary answer. But, if at least one of them does not admit of a precise and non-arbitrary answer, then our main Question will not admit of a precise and non-arbitrary answer either, which is incompatible with the possibility of Edenic qualities. 6. Interlude: Getting to Modal Idealism I’ve finally finished arguing that it is necessary that every quality is a phenomenal quality. But how do we get from this claim to Modal Idealism? Recall that Modal Idealism is the claim that it is necessary that (i) every fundamental entity is conscious and (ii) every fundamental property is a phenomenal property. However, we can disambiguate the second condition in two ways. According to Impure Modal Idealism, every fundamental one-place property is phenomenal. Such a view is neutral on whether there might be fundamental relations that are not consciousnessinvolving. According to Pure Modal Realism, every fundamental property and relation is consciousness-involving. In order to keep the discussion manageable, I will argue for Impure Modal Idealism in the main text, and I will further argue for Pure Modal Idealism in an appendix. It turns out that there are three ways to resist (Impure) Modal Idealism while granting that every possible quality is a phenomenal quality. However, all three of these ways are highly controversial. In this section, I’ll briefly describe these three ways, and mention why they are controversial. First, one could reject Modal Idealism by arguing that there could be fundamental entities that are bare particulars, which do not possess any (non-haecceitistic) fundamental properties at all.44 If there could be bare particulars, then it would be false that it is necessary that every fundamental entity is conscious (contrary to Modal Idealism), since bare particulars would not be conscious. 44 By “bare particular” I mean to refer to what Sider (2006) calls “truly bare particulars” (rather than any “substratum” in the context of the substratum theory of objects and properties). I am also assuming that “merely negative” properties, like lacking mass or lacking electric charge are not fundamental properties (because otherwise bare particulars would have such non-haecceitistic fundamental properties). If such merely negative properties were fundamental properties, then there might be an infinity of extra fundamental properties that we would have to ascribe to everything: for every fundamental alien property P that is not instantiated in our world, there would seem to be a corresponding fundamental property of lacking P that would be instantiated by everything in our world. I am also assuming that “trivial” properties like being self-identical are not fundamental properties, since fundamental properties are typically thought of as substantive properties that are contingently instantiated. For more on properties concerning identity and distinctness, see note 47. Even if one does recognize such merely negative or trivial fundamental properties, then bare particulars could be defined as entities that do not possess any (non-haecceitistic) fundamental non-trivial positive properties. 21 My first response to this worry is that many philosophers, both historically and in contemporary times, have thought that bare particulars are in some way incoherent or unintelligible.45 For one, it is not clear that we can positively conceive of bare particulars. Try to positively conceive of a world that contains nothing but a single bare particular. Such a world would be completely devoid of colors, shapes, masses, particles, fields, conscious experiences, etc. Now conceive of a world that contains nothing at all. Did you manage to conceive of two clearly distinct possibilities? To many, the answer will be no. Insofar as one finds bare particulars to be inconceivable, this gives one some (defeasible) reason to think that bare particulars are not genuinely possible. In writing about the “obvious incoherence” of bare particulars, Strawson (2017) writes, “Clearly there can no more be objects without properties than there can be closed plane rectilinear figures that have three angles without having three sides…to be is necessarily to be somehow or other, i.e. to have some nature or other, i.e. to have properties” (69). Armstrong (1997: 109–10) claimed that bare particulars were “vicious abstractions” and was explicit about their impossibility when building his own metaphysics. Although Sider (2006) defends the intelligibility of bare particulars, he aptly summarizes a common attitude towards bare particulars in saying that “bare particulars are widely regarded as the grossest of metaphysical errors” (392). The second response to make is that there are independently motivated metaphysical views that are incompatible with the existence of bare particulars. For example, according to the “bundle” theory of objects, objects are mere bundles of properties, so if there are no properties to bundle together (as with bare particulars), then there is no corresponding object. The same could be said for a wide variety of other views that reject the existence of an underlying “substratum”, whose role it is to “instantiate” various properties.46 The idea of a bare particular only seems to make sense according to such a theory – a bare particular is simply an underlying substratum that does not instantiate any fundamental properties. For arguments against this kind of substratum theory, see Dasgupta (2017), Builes (2021), and Strawson (2021). The second way to resist Modal Idealism is by arguing that there could be fundamental entities that possess fundamental dispositions or powers (and such fundamental powers would not be phenomenal properties, contrary to Modal Idealism). Of course, it is common to believe in dispositions or powers, like the fragility of a glass, but the posit of fundamental dispositions or powers, which lack any underlying categorical grounding, is a posit that is highly controversial. For example, while it is intuitive to think that a glass is fragile (i.e. disposed to break if struck) in virtue of its microphysical organization, the idea of a primitively fragile glass, which isn’t fragile in virtue of any other features that it possesses, seems dubiously coherent. Sider (2001) argues against such brute dispositions on the grounds that they posit brute “hypothetical properties [that] ‘point beyond’ their instances” (41). More generally, anyone who believes that modal facts must 45 See, for example, Plato’s Timaeus (48c-53c), Aristotle’s Metaphysics (1029a20-33), Locke (1689/1997), Russell (1996), Mertz (2003), Lowe (2003: 86), Bailey (2012), and Giberman (2012). 46 See, for example, Dasgupta (2009) and Turner (forthcoming). 22 be grounded in non-modal facts will reject the possibility of such brute dispositions. For further criticisms of metaphysical views that posit fundamental powers, see Barker and Smart (2012), Barker (2013), and Jaag (2014).47 Lastly, one could resist Modal Idealism by arguing that there could be fundamental entities that possess bare quiddities, i.e. quiddities whose nature is wholly characterized merely by which other quiddities they are (non-)identical with. Perhaps the primary metaphysical reason to be wary of bare quiddities is that they seem to imply a proliferation of metaphysical distinctions that are ultimately illusory. For example, because bare quiddities lack any substantial nature, it is unclear what the difference is supposed to be between a bare particular and an object whose only fundamental property is a bare quiddity. It is furthermore unclear what the difference is supposed to be between a world where a single object x instantiates bare quiddity Q1, and a world where instead x instantiates a numerically distinct bare quiddity Q2. Because bare quiddities do not have any substantive qualitative nature, such possibilities would be qualitatively indiscernible. Lastly, the main motivation for adopting quiddities (namely, to avoid positing fundamental dispositions or causal powers) is satisfied by adopting qualitative quiddities, which Modal Idealism already does. As Hildebrand (2016) argues, the reasons to posit quiddities are perfectly satisfied by qualitative quiddities, and the many arguments against quiddities are only persuasive against bare quiddities.48 In sum, the case from “all possible qualities are phenomenal qualities” to Modal Idealism is not airtight. One could avoid Modal Idealism by upholding the possibility of bare particulars, fundamental powers, or bare quiddities. Still, all of these posits are highly controversial, and the claim that all possible qualities are phenomenal qualities at least makes Modal Idealism much more plausible than it otherwise might be. 47 It should be noted that a popular argument against fundamental powers, namely that positing fundamental powers involves a certain kind of vicious regress (e.g. see Robinson (2022: ch. 9)), only applies to views where all fundamental properties are powers. However, what is needed here is the stronger claim that it is impossible for there to be any fundamental powers. 48 There are some other more exotic ways to resist Modal Idealism. For example, one could think that there are possible fundamental haecceitistic properties, such as the property of being identical to p or the property of being distinct from q. However, such properties are not needed to have fundamental haecceitistic facts. For example, suppose the fundamental facts were of the form [p1 is F], [p2 is G], …, [p1 bears relation R to p2], …, etc. where all of the properties and relations are non-haecceitistic, but p1, p2, etc. are primitively distinct individuals. Such a view would have fundamental haecceitistic facts, but not because of the fundamental properties that the view recognizes. One could also think that everything has a fundamental non-haecceitistic property of being self-identical, which would also be a counterexample to Modal Idealism. However, because everything “trivially” has this property, it seems to be better to not posit this as an additional fundamental property, since fundamental properties are often thought of as contingently instantiated. Lastly, even if these kinds of counterexamples to Modal Idealism succeed, one could refine the definition of Modal Idealism to be “necessarily, every fundamental entity is conscious, and all non-haecceitistic contingently instantiated fundamental properties are phenomenal properties”. Defending such a view would still be of great metaphysical interest. For different accounts of the grounding of identity and distinctness facts, which don’t commit to fundamental properties concerning identity and distinctness, see Litland (2022) and Rubenstein (2023). 7. The Second Argument: The Eleatic Principle 23 The second argument for Modal Idealism appeals to the following principle: The Eleatic Principle (EP): Necessarily, every concrete object has causal powers.49 First, I’ll argue that the best way to develop a metaphysics based on EP is to endorse Modal Idealism. Second, I’ll argue that several potential counterexamples to Modal Idealism can be resisted by appealing to EP. In sum, I’ll be arguing that EP strongly supports Modal Idealism. Could there be a concrete object that is entirely causally inert? For example, could there be a spatiotemporally located object (hence, a concrete object) that is entirely causally inert? At the very least, it is hard to see how we could ever know about the existence of such an entity, since we could never causally interact with it. So perhaps we could never justifiably believe in any counterexamples to EP. However, many philosophers, both historically and in contemporary times, have gone further and defended the metaphysical requirement that having causal powers is a necessary condition for (concretely) existing in the first place. According to Leibniz (1714: 65), “activity…is of the essence of substance”, and Schopenhauer (1813: 119) wrote that “matter is throughout pure causality, its essence is action in general”.50 Further contemporary supporters of EP include Armstrong (1978), Ellis (1990: 22), Field (1989: 68), and Strawson (2016). My goal in this section won’t be to further argue for EP. Instead, my goal will be to draw out some of the consequences of EP for those who have some antecedent attraction to the principle. My overall strategy for arguing that the best way to endorse EP is to endorse Modal Idealism will be to argue that Modal Idealism allows the defender of EP to avoid having to posit all sorts of brute necessities. Since EP is crucially about causation, and since philosophical views about causation are inextricably linked to views about laws of nature, I will be considering how EP might be developed according to different views about the nature of causation and laws. To start, consider a “Humean” view of causation and laws. According to Humeanism, there is a spatiotemporal mosaic that is filled with local, point-sized qualities, and there are no metaphysically necessary connections between the intrinsic state of one spatiotemporal region and any other non-overlapping spatiotemporal region. According to the most influential development of Humeanism by David Lewis, laws of nature are merely simple and informative summaries of the mosaic, and facts about counterfactuals and causation are determined by reference to such laws.51 The problem with combining Humeanism with EP is that there are metaphysically possible 49 A popular strengthening of EP is that “to be is to have causal powers”. I will appeal to this strengthening of EP towards the end of this section. 50 Strawson (2016) also attributes something like EP to both Plato and Aristotle. 51 Versions of this “best systems account” of laws are developed and defended by Earman (1986), Lewis (1994), Loewer (2007), Cohen and Callender (2009), and Hall (2015). 24 mosaics that do not exhibit any simple or informative patterns at all, so such possible mosaics would have no laws at all. However, in the absence of any laws, entities in such a mosaic would not stand in any causal relations, violating EP.52 In response, the Humean could simply reject the metaphysical possibility of such mosaics, but without any independent reason to do so, such a rejection would be committing to brute constraints on modal space. Moreover, as Cowling (2015) has noted, there might be mosaics where it is indeterminate whether there are any laws, because it might be indeterminate whether there are any sufficiently simple and informative regularities in such mosaics. In order to comply with EP, the Humean would either have to say that the space of possible mosaics is itself indeterminate, or else they could admit the possibility of things that “indeterminately exist” (since it is indeterminate whether they stand in causal relations). Neither option is particularly attractive. Among Non-Humean approaches, some account for causation in terms of laws of nature, and others account for laws of nature in terms of causation. Let’s start by considering the former. According to law-based views, laws of nature can either be understood as fundamental entities in one’s ontology, or else as relations of “necessitation” between universals.53 A familiar objection to such views is that the metaphysically necessary connection between their Non-Humean posit (“laws” or “necessitation relations”) and corresponding events in the concrete world is mysterious. Why is it absolutely necessary that if a universal F-ness stands in a certain relation to a universal G-ness, then every F must also be G?54 However, going beyond this standard objection, in order to comply with EP, such views would have to say that all possible worlds have primitive laws or necessitation relations, because in the absence of any such laws, nothing would stand in any causal relations.55 Given that it seems easily conceivable for there to be worlds that lack these NonHumean posits, this seems to be a problematic brute necessity. In addition, there not only have to be some laws in every world, every world has to have laws that govern every property in that world. Otherwise, there could be objects with causally inert properties that are not governed by any corresponding law. It would be better if the defender of EP could do without these brute constraints on modal space. Third, there are Non-Humean approaches that posit fundamental causal powers.56 The first problem for this view is one we’ve already seen: while it is familiar to posit causal powers or dispositions that obtain in virtue of underlying categorical facts, fundamental powers are much more controversial. Furthermore, this view faces a similar problem to the previous view: it would 52 There are other versions of Humeanism that do not reduce causation to laws (e.g. see Wilson (2009)), but such views face similar problems. 53 See Carroll (1994), and Maudlin (2007) for views of the former sort, and see Dretske (1977), Tooley (1977), Armstrong (1983) for views of the latter sort. 54 For more on this “inference problem”, see van Fraassen (1989), Sider (1992), and Schaffer (2016). 55 Maybe other views about causation could hold in worlds without laws of nature (e.g. Humeanism). But such a position would then also face the problems faced by other views of causation. 56 See Mumford (2004), Bird (2007), Chakravarty (2007). 25 have to be combined with the view that it is impossible for there to be fundamental properties that are not powers: otherwise, it seems like something could exist that only possess such non-powerful properties. In the absence of any independent motivation for this modal constraint, this would be a brute necessity that it would be better to avoid. Let us finally see how Modal Idealism might secure EP. According to EP, everything must have causal powers, but according to Modal Idealism, there are no fundamental causal powers. The only fundamental properties are phenomenal properties. So, the combination of EP and Modal Idealism entails that everything must have non-fundamental causal powers. This immediately raises a worry about brute modal facts. Facts about the distribution of non-fundamental properties must be necessitated by facts about the distribution of fundamental properties, since non-fundamental facts supervene on fundamental facts. So to avoid brute modal facts, the Modal Idealist must be able to explain why things have the non-fundamental causal powers they do by appealing to facts about the distribution of fundamental phenomenal properties. How could this be done? The answer is that it can be done by the Phenomenal Powers view, which is precisely the view that phenomenal qualities ground corresponding causal powers.57 The fundamental insight behind the Phenomenal Powers view is that the connection between phenomenal qualities and their corresponding causal powers seems to make sense. For example, it is not a mystery why someone who is in pain would try to get rid of that pain: the reason why people try to avoid pain is because of its intrinsically repulsive character. Similarly, it is not a mystery why someone who feels hungry or thirsty would be motivated to try to satisfy their hunger or thirst, or why someone who feels an itch would be motivated to try to scratch the itch, or why someone who entertains an obvious proposition like “2+2=4” would be motivated to endorse it58, or why someone who feels exhausted would want to stay in bed, etc. So, the Phenomenal Powers view not only entails that phenomenal qualities ground corresponding causal powers, but it also claims that this grounding relationship is intelligible.