Against Infinite Nothingness: Ultimate Ground vs Metaphysical Nihilism in Indian Philosophy

1 The Problem of Ultimate Foundations

In philosophy – while many past projects (of Plato, Aristotle, Aquinas, Spinoza, Leibniz, Hegel and others) have pursued a fundamental analysis of reality’s grounds and most basic structures – contemporary philosophers may see ultimate ground as a rather abstruse point of metaphysical speculation. It can seem to have little relevance to the more pragmatic, empirically grounded areas of philosophy of language and science-led naturalistic metaphysics. Yet for its modern supporters many concerns of metaphysics seem to require it, and it reminds us of metaphysics’ grandest ambitions to achieve an ultimate and unifying explanation, giving a complete account of constitution, causation, individuation, contingency, and complex or emergent phenomena.^[11] Thus some recent metaphysicians take grounding relations by their nature to imply an ultimate ground, ensuring the world is not based on “turtles all the way down”.^[12] Yet it is important to note that “ultimate ground” does not necessarily mean a “bottom” level in the hierarchy of grounds, nor a “concrete” physical ground in nature. Consequently foundationalist critiques targeting the traditional model of a “Great Chain of Being”^[13] rooted in matter tend to miss the central thrust of the view.

In theology, by contrast, ultimate unified grounding is of central concern because it is crucial to most ideas of divinity; arguably even those theologians in the post-Heideggerian tradition committed to a God beyond Being would not claim the world is self-grounded or grounded in something other than the divine. It is grounding that relates the divine to the world ontologically; as Spinoza puts it, reflection on the world seems to show that substance, consisting of infinite attributes, of which each expresses eternal and infinite essentiality, necessarily exists.^[14] On this idea of vast ontological and explanatory power, a picture of reality’s hierarchies can be built. Foundationalism thus points to something with a double-sided quality of self-existence on one hand, and an ability to anchor and generate dependent phenomena on the other.

But at least two problems bedevil the foundation-finding project: The first is that it remains a matter of prolific controversy as to whether such a foundation really is implied by the world. There is an enormous array of astute anti-foundationalist views to be found across most schools of Buddhism over the last two millennia, and a recent wave of scepticism about ultimate foundations that includes Zimmerman and Bliss on the possibility of real metaphysical ‘gunk’, Priest and Della Rocca on the incogency of the notion of ultimate real structure, McKenzie on the unlikelihood of any real fundamental level in nature, and Westerhoff on a wide-ranging, wholesale rejection of different brands of foundationalism, and (with Schaffer and Tahko) defence of the real possibility of a coherentist or regress-embracing metaphysics of infinite grounding (without termination in an ultimate bottom level).^[15]

The second problem is that even if we show that an ultimate grounding function is a necessary prerequisite of the world, it is simply very hard to have an idea of what kind of thing the ultimate ground could be. The argument for a complete foundation that is independent, uncaused, or unformed, tends to start from the observation that worldly things do not possess these qualities. Arguably we see no thing in the world that is groundlessly a ground, although abstract objects like numbers or simple properties are sometimes offered as counterexamples (it is with this in mind that theologians sometimes treat such abstract objects as a challenge to divine aseity). But the usual view is that these are non-actual and do not have the kind of fundamentality a ground of reality would need. On the other hand, it may be that no concrete level of concrete things could fulfil the required grounding functions. Indeed, this may be one of many reasons why theologians have been so cautious about attributing ‘existence’ in any normal sense to any God they define as the ultimate ground of reality. The simple reality that grounds all would exist in a different ‘mode’, as Aquinas put it.^[16] We should not be surprised if it is not atoms or matter, but a different kind of foundation we discover.

Substance is one of the ideas history has offered in some form across different cultural traditions to ground the properties and the structure of reality. As Robinson puts it,^[17] “most, if not all philosophers acknowledge that we cannot function without using substance concepts in the narrower sense, for the notion of an enduring particular of individual substance is essential to our making sense of the world as we live it.” Yet from a philosophical perspective these ideas are skeleton keys, philosophical Deus ex Machinas, or “McGuffins” as Hollywood puts it – they perform a crucial function in structuring the cosmos, but they do so in a vague, hand-waving, “don’t know how” kind of way. Michael Della Rocca has made this the starting point of one recent wholesale attack on substance (that mirrors aspects of some Madhyamaka critiques).^[18]

The usual defence of foundationalism draws on arguments for “inheritance” – the idea that hierarchies of grounding need some baseline in order to get going – rather as if there were some ontological capital that is parcelled out in each grounding relation. But as Westerhoff has recently made clear,^[19] along with other defenders of infinite grounding relations, not all of the relevant kinds of relation actually seem to function in this way. It is possible that i) properties may supervene infinitely downward onto underlying levels, and that ii) constitution may be infinitely analysable down into ever smaller “parts”, or that iii) causal events could regress infinitely backward through time, or indeed that iv) these grounding relations have a mutually-supporting or circular character (on which possibility more below). In each of these cases, as McKenzie and others have noted, it may not be pure reason but scientific research that would resolve the question. But as we will see below, other kinds of fundamentality may indeed be a matter of reasoning to the fullest picture, deepest structure, governing measure, or some other phenomenon strictly entailed by the empirical world as we know it.

The Indian idea of foundation that is implied in the inheritance argument discussed here, is based on the Satkārya theory of causation and identity, and aims to a different kind of foundation than the “level of nature” of which philosophers like McKenzie speak. The Satkārya views makes use of a structural-determination ontology that sees things in terms of the determining nature that gives rise to them – the formative factor that makes a seed into a banyan, or an egg into a peacock. It is this formal cause rather than the material or temporally efficient cause that matters in this theory.

Counterfactually reasoning that things could apparently have been otherwise, these thinkers take our radically improbable cosmological order to require some sufficient reason, a consistent modal anchor, as it were. It was this argument that was adopted for monistic and theological purposes by the Vedāntic school when it needed a defence against Buddhism’s arguments for metaphysical groundlessness and global emptiness of structure. But more was needed to turn the idea of ultimate ontological ground into a theology-ready concept: it needed to show a unified ground rather than a pluralistic, piecemeal one. The arguments from coherence and connectivity that we explore below aimed to show that the foundation must be unified in its nature and unified as an object-constituting medium. On the critical side, the debate set in motion a long history of dispute over whether we should feel liberated by freefall in an infinite, empty ‘gunk’, or whether we should positively re-orient toward participation in a sublime, encompassing, radically different kind of reality from all we see at the merely local level. The spiritual significance we accord to different metaphysical pictures was in question.

2 Ultimate Grounding: Inheritance and the Satkārya argument

In ancient India as in ancient Greece, a number of different metaphysical schools competed to give the most convincing account of reality. One of the most intriguing views of individual objects, and the nature of the immanent causation that shapes them, is found in the Satkarya-vāda or “existent-effects theory” of the Sāṃkhya school. Arguing that things already exist (in some sense) in the prior causes that lead to them, the Satkārya theory makes a point about the grounding of causal determination that has been touched on in recent debates by Kelly Trogdon on “causal capacity inheritance”, Jonathan Schaffer on “internal constraining relations”, Jeremy Skrzypek on potency, and by others.^[20] Satkārya claims that the future forms of a thing pre-existent in potentio within them, committing to a kind of subtle presence of future things in their present versions, even when they are currently invisible. The tree must already be contained within the seed, and the bird present in the yolk. Why? Because each form that arises must get its form from some prior determining cause. Otherwise where would new forms come from? They would have to arise ex nihilo, without cause or explanation – and if this were possible then anything could arise anywhere anytime... but this is not what we see to be the case.

The first explicit argument for this idea seems to be found in the c. 300 CE Sāṃkhya Kārikā. The reasons used here then found their way into later scholastic commentaries, and were used to justify metaphysical foundationalism from at least the time of the great 8^th century idealist Śaṃkara onward. Having set out these views in more detail elsewhere I summarise them here.^[21]

In the Sāṃkhya Kārikā, this argument aimed only to show a non-divine foundation of the empirical world from which all else flows. The original argument runs as follow:

[We know that] the effect exists [prior to its empirical appearance] because of:

a) non-being not being a cause [of anything],

b) from grasping that there is a material cause [prerequisite to particular kinds of effects],

c) from it not being the case that everything comes to be,

d) from [a thing’s own] capacity’s power to do [only] what is possible [to it], and

e) from the existence of [the phenomenon of] causes.^[22]

The reasoning here is that a) spontaneous generation out of nothing is never seen and so the universe could not have arisen out of a total causal vacuum; b) the relation of materials and later forms/developments of them follows consistent patterns (remembering that intransitive development rather than transitive influence is the main focus here); c) things do not develop in a random fashion, but rather according to the aforementioned patterns, d) the phenomena we see accord with certain capacities and possibilities that we may infer from the patterns, justifying the idea of powers, e) causation itself is an expression of this phenomenon, and the idea would not exist without something generating it that would have to derive from this observed structure. So there must be some existence of the effect prior to its arising from the materials or causes that precede them. The pradhāna or prakṛti – the foundation – was conceived as the determinative storehouse of these possibilities.

This Satkārya argument was countered by the Asatkārya theory, a counter-model proposed by the atomists of ancient India, the Vaiśeṣika school. This held that each set of new forms do not exist at all until they are constituted freshly by juxtaposing pre-existing factors – aggregates of atoms, or some such basic units – in a new way. The two theories competed largely on the basis of their respective explanatory power. Satkārya could explain the way that (as Aristotle has also noted) natural phenomena seem to know how to develop, and Asatkārya could explain the way that new determinations could arise from multiple grounds or mere interactions. As the Sanskrit example had it, Satkārya could explain why milk naturally curdles, and seeds become trees, but Asatkārya could explain why cooked seeds do not grow into trees, and threads need a weaver to become cloth.^[23] Satkārya’s ontological foundation could not be reduced to any one form that a thing takes, but rather contains them all so that creation is “actually a process of unfolding, or manifestation of, what had always been there”.^[24]

This Satkārya idea of a substratum “cause” containing the sufficient conditions for each effect’s emergence was used by the monistic foundationalist system of Vedānta, and came to be embedded in its core text, the Brahma Sūtras, a redacted series of aphorisms dated to around 400 CE. A long section spread across chapters 2.1 and 2.2 of that text sought to champion Vedānta’s idea of a single complete divine foundation for all things (termed “Brahman”) over competing accounts that included the atomists, and the radical anti-realist and metaphysically nihilist Buddhist traditions. The Satkārya argument of the earlier Sāṃkhya school was used alongside a number of other arguments, to depict the universe as unified by a shared causal ground and constitutive medium.

It may have been Gauḍapāda, the c.6^th century Vedāntic author of a commentary on the Saṃkhya Kārikā who first unpacked Sāṃkhya’s modal argument as an argument for a divine foundation. He wrote:

In this world one selects the material of what one desires to have; one desirous of curds selects milk and not water... and also, because it is not the case that everything everywhere arises, just as gold cannot be produced from silver, grass, dust and sand. So since not everything comes to exist, the effect exists [in the cause before its appearance]... also, because the power of a potter can cause a pot to arise from a lump of clay only when there is [the appropriately] potent means which are the soil [for the clay], the potter’s wheel, a thread of cloth, and water.^[25]

This important passage is echoed by Śaṃkara, a more famous expositor of the idealist Advaita branch of Vedānta who thrived around two generations later. He uses it in his commentary on the Vedāntic (rather than the Sāṃkhya) core text, and implied a single foundation through the observation that we see an overall coherence of causal trajectories across cases, indicating a consistent foundation underpinning ontological types. As Ram-prasad has put it,^[26] this new conception of foundation was of something that not only shapes individual things, but also operates globally across cases as the requisite “causal regulator of materiality”. Here is Śaṃkara’s version:

In the world it is seen that people wanting curds, pots, necklaces, etc. take up their well-established respective (material) causes – milk, clay, gold, etc. – not that a man wanting curds takes up earth, or a man wanting a pot takes up milk. This fact does not fit in with the theory of the non-existence of the effect before origination. If everything were equally non-existent everywhere before creation, why should curds be produced from milk alone and not from clay; and why should a pot come out of clay and not out of milk?^[27]

This brings out the pragmatic epistemic force of the observation: if we do not believe in some formative factor then we ought to expect to be able to churn delicious curds from clay, and shape a nice hardy pot out of milk. To this he adds the further counterfactual observation that if things simply continued as they are once in existence, nothing would change and there would be no cosmos of development and interaction. Or if things had no governing nature to which they adhere then they would change randomly:

...when from one combination another emerges, it would either be regularly similar, or randomly be similar or dissimilar. If regularity be admitted, then a human body can have no possibility of being transformed into divine, worldly, or hellish bodies. And if irregularity be admitted, then a human body may momentarily become an elephant, or be transformed into a god or back into a human form again. But both views contradict your position.^[28]

Śaṃkara was an idealist but other realists and theists made use of the same argument. Indeed, the argument seems meant to apply to the empirical world no matter what metaphysics we apply to it. Here is a later realist Vedāntin, Śrīnivāsa, who sees all reality as an immanent transformation happening within the real being of Brahman, the foundation:

There is no ability to manifest or conceive of a pot in things like wind that are different from clay; if things arose from non-being then it would happen that everywhere everything would arise, and the function of agency would be obsolete.^[29]

Śrīnivāsa tries to strengthen the insight with both natural experience (pots are never formed out of the wind) and the naturalistic sense that if there were no such ground then the very notions of causation and agency would have no meaning – indeed, the conditions would be absent for them even to have arisen. He adduces the natural content of our concept of agency to show that a random model of arising goes against our experience. This is a causal inheritance argument for an ultimate foundation of things – in this sense it operates somewhat like the kind of cosmological argument that is familiar to Philosophers of Religion. But it takes a specifically “modal” form focusing on explanation for why things develop in the contingent, modally particular way they do, rather than the counterfactually possible ways they do not.

Linked to this inheritance argument from modal specificity is a second inheritance argument from measure, based on empirical reality’s need for a shared consistent measure of size or spatial extension for all forms. Otherwise no juxtaposition and comparison would be possible. Although the argument speaks of “parts”, it actually only requires determinate measures. The theistic Vedāntin Rāmānuja puts it this way:

...if atoms possessed no distinctions, even an abundance of atoms would not be able to have minuteness, smallness, largeness, length and so on^[30]... But it must be said that it is the small or great quantity of the parts that differentiates the mustard seed from the mountain, since if it had infinite parts then there would be the same number of infinite parts in a mustard seed and a mountain, so that their difference would remain unaccounted for.^[31]

Here what matters is not that there be a most small part, but that parts preserve the same structural measure at each level, regardless of addicity or divisibility. This scalar notion of inheritance aims to provide a notion of fundamental “building blocks of reality” without necessarily committing to the “more mereologically inclined approach”.^[32]

Similarly, Śaṃkara uses an argument for the need for some consistent overspanning measure of positionality in time and space. If there were no shared measure of locus (an idea for which he uses the idea of ākāśa or “space”), then we could never tell if two birds were approaching each other in the sky, and about to meet each other as an “obstruction” preventing each other’s progress:^[33]

That very thing with the help of which “the existence of an obstruction” is specified, will itself be the positive (substantial or grounding; vastu) entity space and it will not be a mere absence of obstruction... Thus it is not correct to say space is insubstantial.^[34]

Space here means that field from which things get their positionality in relation to others. Where nihilists used concerns about how we can tell the relative movement of two birds in the sky to deny the coherence of motion,^[35] Vedānta’s foundationalist used motion problems to show that a consistent field of positionality is a necessary inheritance for all such relational talk. Śaṃkara here seems to accord with a view attributed to Aristotle that “the cosmic natural places are defined with respect to the universe.... The identity of these places cannot be understood without reference to the structure of the whole... they do not exist without it”.^[36] The idea of a positional field thus provided a further kind of unified inheritance necessary for the cosmos we know.

Arguably, these two forms of modal and positional inheritance are orthogonal to the problems of divisibility into atoms or gunk that motivate many arguments for and against an ultimate ground. We can see this from Tahko’s example,^[37] where one could have a medium of reality that descends infinitely to ever-new subsidiary structures, but contains a foundation because it descends “boringly” in such a way that the pattern repeats after some point so the same basic structure is preserved from there downward. In this case the regress would nevertheless support a structural foundationalism, and everything would be reducible to universal basic structural features. The scalar (mustard seeds and mountains), positional (birds in the sky), and modal (seeds becoming trees) arguments for inheritance were tools for arguing that there must be a consistent foundation that orients and shapes all of the reality we know. For Vedāntins the unique ontological autonomy of this ground meant it had to be divine.

3 Fundamental Unity: Coherence, Entanglement and Medium

In many ways, the argument for an ultimate ground – familiar from cosmological arguments and applicable to classical Aristotelian substantialism – was less controversial than the argument for a unity of ground. It was this idea that had led Europe toward debates about monism in its Spinozist formulation. For the Vedāntic thinkers, at least three points were cited as showing the unity of the foundation. One was the coherence of the empirical patterns we see, implying a coherence of the modal grounding that makes them occur that way. All other things being equal, all seeds become trees, and all eggs become birds. Given the counterfactual reasoning we have seen from Śaṃkara and Śrīnivāsa above, this continuity (seen both in each case moment by moment, and in all cases notwithstanding countervailing factors) seems extremely improbable without some shared determinant. In the Satkārya account, that shared effect is attributed to a shared pradhāna or foundation that encodes the forms that arise out of it with the same structure.

In explaining the Satkārya theory, Larson emphasized that the sufficient conditions for any form coming to be lies in its total ground;^[45] but what if the ground appears divided into plural things? Jeremy Skrzypek has made the point that “dynamically complex” entities which maintain an overall coherence rather than changing randomly, imply “a guarantor of diachronic identity”.^[46] This parallels a recent “modal argument for the causal principle” by Koons, drawing on Gale and Pruss, for the existence of a divine ground based on coherence in nature and the idea that ‘‘the grounding of modality requires a global principle of causation”.^[47]

Outside of theological circles, Schaffer expresses this idea that contingent coherence requires foundational unity counterfactually (as Vedānta does) in terms of a failure of unconstrained appearances:

... the argument from free recombination... If there really were multiple basic independent units of being – as the priority pluralist claims – then these units should be, in Hume’s words, “entirely loose and separate.”^[48] They should be freely recombinable in any which way. Any way the one can be and any way the other can be ought to be compossible. A disconnected pluralistic heap thus should be amenable to free recombination, and thus failure of free recombination is the modal signature of an interconnected monistic cosmos.^[49]

He takes this combinatory constraint as a sign of “internalconstrainingrelations”^[50] and cites Plato, Plotinus, Hegel, Joad, Lotze, Blanshard, Ewing, and Charles Baylis as allies pointing to the same observation. Yet Schaffer’s account is not concerned with some shared ground but rather with the idea of a direct mutual influence between the parts, or a necessarily interlinked essence. They are “interdependent”, “interwoven”, or “internally related”, rather than co-grounded and sharing a nature, as the Satkārya view thinks (indeed, Zimmerman questions the reasons we might treat this as a unity).^[51]

Schaffer does consider that a “shared material origin” seems a strong explanation of a common nature, analogous to “two fragments... produced from one explosion” – an idea that conveniently evokes the Big Bang.^[52] He also cites Mumford’s “holistic” view of properties and a view from Ellis that refers to ongoing “causal powers” in which events activating these powers necessitate other events that are their displays, and lead to a “bound and connected world” where the dispositional trajectories that shape the world effectively unfold from a single power.^[53] Matching Larson’s structuralist account of Satkārya theory, Schaffer also considers formulating this as a “structuralist thesis”.^[54] The structural approach suggests that we might opt to ignore ontological questions of this kind, and accept that whether it is a force, material, power, or mind that grounds the ordered cosmos, that ground is structurally coherent in nature. Sider says something like this, objecting that Schaffer’s monism might merely be a unity of “statespace structure” shaping things rather than a single thing or material.^[55] Whatever its metaphysical medium, the structure would nevertheless be fundamental in its “distributive” pattern,^[56] for on this account it is types not tokens define the world. One might even attribute this kind of view to Aristotle, respecting his theory that “parts lose their essential character when severed from the whole”, since their essential character includes their “natural places” in a formal relation that is “teleologically subordinated” to each other.^[57] Here both Aristotle and Vedānta share a concern to configure part-and-whole in terms of a relational “placing” that indicates holism.

A further conception of fundamental unity is found in the argument that powers have an un-individuatable and entangled nature. In what I have elsewhere called the “complexity” argument or the “argument from shared powers”,^[58] Vedāntic thinkers pointed out that emergent properties cannot be reduced to any subsidiary constitutive “part” of their material or causal base. Śaṃkara’s example was that a cow’s moo is a product of the whole organism and no single atom; similarly conscious experience arises from no single atom of the body (or even of the brain, following the example given by Rāmānuja). Such properties must supervene onto to the whole base which together possesses them.

This suggests a particularly curious kind of unity: no single concrete individual appears at the bottom level, but rather a power or disposition to generate something shared. Further, what the power generates is itself complex but not in such a way that its joints supervene directly onto the lower levels; it is like a story made of syllables, or imagination arising from a brain. As for modern non-eliminative reductionists about emergent properties, novel “holistic” phenomena are indicative of holistic powers at the foundational level.^[59] Since emergent phenomena are recombinable, often shifting materials (as with the ever changing cells of living organisms), and since they are found everywhere, sometimes coalescing or splitting, it seems like the task of carving reality at the joints of its “powers” would be very difficult. As Heil says:

...if you cannot define (or analyze) the concept of a person, or a table, or a promise into ‘lower level’ physical terms, [then] persons, tables, and promises must either be something more than physical entities and their interactions, or such things must fail to exist.^[60]

Buddhist metaphysical nihilists choose the latter option and abandoned the explanatory project at the heart of commonsense, scientific, or metaphysical discourse. Sāṃkhya and Vedānta chose the latter option and strove to reverse-engineer an explanatory foundation from the world it grounds.

A final Vedāntic argument for unity attacked the possibility of fundamental individuation between the constituents of reality, insisting on the consequent ontological continuity of all things. Where the first argument sees unity in the coherence of natures, and the second sees it in entanglement of power, the third sees it in the constitutive ontological continuity of the different kinds of “parts” that make up the cosmos. To borrow Schaffer’s taxonomy of monistic claims, it argues not merely for priority monism of a unified ground, but existence monism where the unity takes place on the same level at which we would normally individuate objects. This argument is echoed by a number of modern Western thinkers from F. H. Bradley to Vallicella, Horgan and Potrc, Priest, and Della Rocca.^[61] It critiques the idea of a “linking” mechanism (usually of inherence or aggregation) between fundamental entities, arguing that the notion generates an infinite regress (or in Della Rocca a sorites sequence) which – if shown to be incoherent – indicates that reality must in truth already be linked by some medium that allows for connections spanning the apparent boundaries of things. This ontological continuity might fruitfully be juxtaposed with the positional argument’s claim that there must be a shared fundamental measure of positionality which could also provide a connective tissue between things. The nature of the unity indicated by an anti-fundamental-individuation argument would depend upon the sense it which absorbs objects “in” it into a single reality. But we can note that the argument has a certain power pertaining to the nature of any plurality of connected entities such as we would call a cosmos. Whether we speak of causal, or constitutive, or inherence connections, in any ontology that involves genuine interactions we would need a means of connection at the fundamental level. As the positional argument shows, even an ontology of juxtaposed tropes where phenomena float unconnected, need some shared field in order to be in juxtaposition.

This last approach to unity most explicitly makes the controversial claim that we should infer a shared ground and medium of all reality – a view that takes us toward metaphysical monism, and theological pantheism or panentheism. The idea that what links what we normally think of as discrete “things” into a shared world that is part of the things should not appear too shocking, since it is arguably close to physicists’ accounts of particle, waves, energy, fields and so on. As Della Rocca and Priest note, this idea dates from at least Parmenides and perhaps Zeno’s paradoxes, and as Priest in particular points out, it invites us to rethink our inclination to individuate things according to our common language use – since we see a different story at the fundamental level.

4 Against Infinite Nothingness: Nihilists and Monists as allies

Here we come to an important feature of arguments for unity: almost the same arguments against clearly separate basic entities can be used to support both ontological nihilism and ontological monism. This ambiguity is pleasingly reflected in the title of Graham Priest’s book One, which is an attack on “apparently simple philosophical concepts, such as: being, existing through time, intentionality, identity” and leads to a wide-ranging rejection of the existence of any “unities”, including common-sense objects.^[62] In his arguments, he takes inter-dependence to indicate “emptiness”, even though it is precisely one of the things that urges classical Vedāntins and modern metaphysicians like Schaffer toward monism. As he puts it:

...what it is that makes [something] empty... [is merely] its relatedness to all things... For a thing to be empty is for it to be what it is by relating to other things... For something to be one thing is for it to have its being by relating to other things: something could not be a one thing unless it was located in a field of relations.^[63]

Of course Priest is thinking of essences and Schaffer of material parts, and it is possible that we are more inclined to take internal interdependence as a sign of a complexly unified thing in the case of physical objects (e. g. a sphere, a body), than we are inclined to take interdependent features as aspects of a single identity in the case of essences (as perhaps of the three angled-ness of a triangle, family, siblinghood and generations in the idea of an ‘uncle’, or of human-ness, embodiedness, and spatio-temporality in the identity of ‘John’). But arguably in both cases it is exactly inter-dependence that is the mark of a complex unity.

Similarly, when Westerhoff speaks of a sum of things structured in complete circular dependence “all the way around”,^[64] he takes this as an anti-foundationalist picture which shows there is no ultimate ground. But it seems that he might equally take it as a monistic one where the total structure, replete with internal intra-dependence but free from any external dependence, is self-grounded and as a whole possesses surprising sovereignty and aseity. From this we can conclude that Buddhist metaphysical nihilism is not only dualistic, but also determinedly insistent on an ontology of non-dependent things – and finding all things dependent, it declares unmitigated nothingness. But this is a nothing full of patterns, relations, shifting identities, dynamic developments, shared powers, and an interwoven coherence that seems the hallmark of both fullness and unity of being. Indeed, if the connectivity argument shows the non-fundamentality of individual things, then the coherence argument reminds us of the rich range of forms and structures for which the foundation is responsible.

Overall we may find that these Indian monists and nihilists are also allies against traditional ontologies of overly-naïve realism, and pluralistic basic items. Both oppose the idea that the separateness we see traces right down to the fundamental level of things. Both also allow for mind-dependent indeterminacy of individuation as one way of cashing out empirical reality’s phenomenological evidence, since they seek an ontology that would apply to appearances as much as to external objects. Arguably they both also invite scepticism about the idea that a foundation could be a concrete object in an ontology of “objects on which other objects depend, but which do not in turn depend on anything”.^[65] Either there is no ultimate ground, or it its existence and one-ness would have to be quite different from that of other subsidiary things. Yet the foundation-monist does see a “full” rather than an “empty” reality, and her key objections to the ontologically nihilistic account could be construed as follows: firstly, despite talk of “emptiness” even Madhyamaka accepts that something exists in some sense. Otherwise the characteristic Buddhist idea that we should accept the simple “thusness” (tathatā) of things would have no sense, and nor would the Madhyamaka idea that in everyday life we deal in “conventional” truths have purchase on anything.

Secondly, the rejection of truth as correspondence to an external reality in no way obviates there being a real reality worthy of valid metaphysical reflection. We can limit our data to the evidence of empirical experience, and still deal with a complex, structured, dependence-layered ontological medium. Indeed in places Buddhist claims to emptiness sound more like methodological pleas against externalist naïve realism. Nāgārjuna’s argument against our ability to “establish” any truths about an external reality assumes that legitimate truths can only be about mind-dependent realities, but this is very much up for question in Vedāntic (and many other) ontologies. He says:

[1] If the means of knowing are self-established, they are established independently of the objects known. Self-establishment does not require anything else. [2] If you think that the means of knowing are established independently of the objects known then those means of knowing are not means of knowing about anything. [3] If the means of knowing are established only in relation to the objects known, the objects known are not established by the means of knowing.^[66]

This passage seems to happily establish that the “means of knowing” (which we take to be our raw empirical data, the phenomena, let us call them) have the reality or truth they have independently of some separate object to which they might refer [1], and do not simply by their nature tell us about anything else in addition to them [2], nor require something they refer to or contain to tell us that they are true [3]. They are simply self-established: they attest to themselves alone. This seems uncontroversial for any phenomenologists or advocates of a similarly minimal ontology. But in preceding verses the author has tried to discount taking those means of knowing or phenomena as what is self-established. He does this by saying:

[4] If the means of knowing are established by other means of knowing, there is an infinite regress... [5] If you think that means of knowing are established without means of knowing, you have abandoned your own doctrine [that we need some basis for knowing a thing]. [6] A means of knowing cannot establish itself because something cannot exercise its characteristic means of knowing upon itself.^[67]

The reason Madhyamaka takes this to show the impossibility of knowledge about reality is that the argument implicitly assumes correspondence realism and is already committed to dualism, as we noted earlier. One of Madhyamaka’s most prolific modern defenders, Mark Siderits, treats just this argument as a rejection of metaphysical realism which he describes as having three theses: “(1) truth is correspondence between proposition and reality; (2) reality is mind-independent; (3) there is one true theory that correctly describes reality”.^[68] But here 2 is a strawman that only a few Indian schools would have agreed with in the first place. The Advaita, Pratyabhijñā, and Yogācāra idealist schools, as well as a number of other Vedāntic thinkers who agreed that reality is made of consciousness or language (Bhartṛhari), would start their ontologies from the phenomena and/or any transcendentally prerequisite features of them, and explicitly acknowledged reality’s openness to multiple interpretations.

If the point of 1 and 3 is that propositional language never captures the total unequivocal truth of the way a thing is, then this is fine (many Indians, again, were relativists, holists, or pragmatists about language), but it says nothing about whether there is something there to be multiply interpreted. Jainism, for instance, was happy to accept that reality exists and has a “naya-vāda” or perspective-relative nature. As Ferraro puts it, this form of nihilism seems rightly unhappy with the process of describing or verbalizing reality, but there should be no problem with simply posing it.^[69] The metaphysical nihilist may defend herself by shifting to nihilism about meaning, saying that nothing is sayable and all truths are self-undermining. But if it is determinate description that is the problem, then it is difficult to see why this would apply to affirming the fact that some kind of occurrences of appearance happen – and this seems, after all, transcendentally entailed in the very process of deliberating about it.

Here it seems that Madhyamaka nihilism uses the rhetorical strategy of willingly targeting the strawman of a determinate fundamentally-individuated non-dependent reality, for the purposes of deconstructing the concept, then eliminating all of the phenomena that some construe in that way. We can call this ‘destructive metaphysical framing’ (in the negative sense of setting someone up for a fall by attaching something false to them). Madhyamaka nihilists of the more extreme types – such as Candrakīrti as interpreted by recent philosophers like Jay Garfield^[70] – also try explicitly to write off direct phenomenal awareness (called svasaṃvitti) as evidence of anything, when it seems that the original critique of phenomenological awareness was meant only to question the idea of a certain somewhat Cartesian idea of a monadic self looking out onto the world. However this shifts the claim back into the dualistic discourse of “egological conceptions of self-consciousness” in order to reject the strawman it creates (as Dan Arnold also notes).^[71] Again, the claim that admitting the existence of phenomena means admitting the existence of some mind-independent determinate Self and the whole dualistic ontology, is obviated if you have rejected that phenomena need externality as a separate ontological base. One need only be a phenomenologist after Yogācāra in India, or Hegel, Husserl and Heidegger in Europe, to know that it is possible to conceive of such an ontology.

Thirdly, Madhyamaka remains committed to statements that it should consider to be true, for even if there is not a determinate way that what we perceive is, there is after all still an overall state of things, a way they are. It is common for the metaphysical or semantic nihilist to say this this is not true... nor even not untrue – it means nothing, or fails somehow to capture the indeterminacy, uncharacterizability, or unassertability of reality. Yet this seems to confuse two meanings of the “state” or “nature” of things. Something that is a particular way need have no single interpretation to which it corresponds. Arguably this is true of most, maybe all things. Perhaps this is why Madhyamaka is wary of Jainism and Bhartṛhari who both seem to hold a doctrine that reality is never univocal, but always subject to perspectival truths: their doctrine is too similar for comfort. But despite these worries, it simply would be the case that “reality is such that it has no single description”. Indeed, it seems that this kind of nihilist really does want to say that reality has no determinately characterisable structure. Denying that there is such a structure could allow that there are some ways reality is and is not, though not any one complete single characterization that captures the way it is. Yet in the larger sense of being non-fundamentally-structured, there is a way that reality is. Some Madhyamaka-inspired nihilists do try to avoid this conclusion by questioning the law of excluded middle or of non-contradiction.^[72] There is not space to address this here, but for the moment we simply point out that the nihilist may want to consider how bizarre the claims they then made would be, and how profoundly they be abandoning any interest in analysing or explaining the world we experience when they deny that law.

Fourthly, as a consequence of implicitly acknowledging the existence of phenomenological occurrences, nihilists of both semantic and metaphysical kinds do actually hold to there being structures aplenty – strikingly consistent and constitutive of all we know – that occur in the phenomena (to continue avoiding the loaded phrase “exist”). Despite Buddhist nihilists’ attacks on objecthood, subject-object distinctions, language’s referential structure and truth’s correspondence status, nevertheless the nihilist is not asserting that the reality we know is simple without any distinctions at all, or a kind of “gunk”, ultimately structureless without any foundational parts or shapes. For the phenomena are filled with complex and surprisingly consistent structures, and to clarify this we can think modally: the gunk might never have had any structures at all. Perception might be empty, static, or chaotic – as Satkārya’s argument points out. This sentence might not have been typed, or might have been otherwise. But instead the structures we have (e. g. are experiencing now) are there. Thus regardless of how we seek to construe the structure ontologically, it is there.

In summary, it is wise to take on board the nihilists’ very helpful critique of traditional ontology, but this does not lead to an empty world. Even through their critique, there would still be a number of things they reveal to us and to which they are themselves committed. They even have an ontology that rejects foundations, but is otherwise fairly uncontroversial on the Indian scene, so that one might want to stop calling them metaphysical nihilists and simply call them metaphysical reformers, ontological post-structuralists, or just phenomenological Gunkists.

So need unified foundationalism suffer from the metaphysical nihilist’s objections? Borrowing from Westerhoff’s helpful taxonomy of Madhyamaka Buddhist non-foundationalist positions,^[73] it should agree with ontological anti-foundationalism that there is little reason to posit one-to-one correspondence of grounds to the discrete objects we see, but it will insist that there are likely grounds for the enduringly consistent trajectories we see, and the grounds cannot obviously be discretely separated between different objects – leading to a grounding holism.^[74] As for epistemological anti-foundationalism, holding that “epistemic instruments and the epistemic objects they give us access to are related to each other by mutual dependence”,^[75] the foundationalist should be happy with this once any implicit dualism is dropped: the interdependence of mind and its objects of knowledge might be an argument for some sort of idealist or phenomenological ontology, but this is not an argument against the grounding claim in question. The unified foundationalist should be happy to agree with linguistic anti-foundationalism that language need not correspond to a mind-independent and determinate reality, but she will insist that some real-world structure determines and explains the structure of the language that arises. Even if language is nothing but a set of “cookie-cutters” (to use Westerhoff’s analogy) that shape perceived experience, this too requires modal explanation, and thus a ground.

The same goes for the non-foundationalist about truth who says that there is no actual, ultimate, or true way the world is, only the way it appears. In this the unified foundationalist would be a “light foundationalist” uncommitted to a mind-independent reality, but agreeing with Buddhist “light-nihilists” that other sorts of foundations could exist “in a [different] manner of speaking”.^[76] Such nihilists would be the allies of their Vedāntic cousins, both fighting to revise our ontological commitments, and resituate themselves in life by doing so.

5 The Puzzle and Promise of Foundationalism

In arguing that there is probably a unified global ground of reality that generates, governs, sustains and explains all things just as Dante and the Upaniṣads describe, these foundationalist arguments together describe something quite strange. They bring out what is so odd about the idea of an ungrounded ground, regardless of whether it is proposed by secular metaphysicians or theologians of a divine creator, by idealists or physicalists, or pluralists or monists. Foundationalists believe in some curiously double-sided phenomenon that grounds without needing a ground, and that is required for the world we know but must be deeply different from it. This difference from familiar worldly things would consist in having an unformed nature that in turns grounds and generates all subsidiary forms (a claim made even by pluralist foundationalists, like atomists, as Cornell points out).^[77] Further, it would have a connecting function that underpins all relation, co-constitution, aggregation and causation. Together these features contribute to an Indian version of what O’Connor calls the “Causal Unity of Nature Thesis”, finding a unity of nature and medium at the fundamental level.^[78]

This foundation bears comparison with Spinoza’s notion of a single substance that is also an immanent cause and determinant of all subsidiary modes,^[79] as well as with Eckhart’s pronouncement that “being is God” who “boils” in and into himself in creating the world, and perhaps even with the all-grounding nature of the Thomist God who is an unmoved mover present in all things by essence, presence and power.^[80] But as implied by these arguments, it need not be theistic, nor considered divine in many of the ways celebrated by theologians and attacked by atheists. It might be “natural” – but if so, it represents something we have not yet seen in nature.

But while these arguments provide a set of explanatory desiderata for the capacities it must have relative to the empirical world, a weakness of the foundationalist’s picture is that it remains hazy about the sort of thing the foundation would have to be to have these capacities. One can understand why the anti-foundationalist might oppose such an idea if there is any way around it – not least for reasons of simple parsimony. To some extent this element of mystery is paralleled in physics, where evidence indicates that “spacetime is not part of the fundamental ontology of physics” so that reality’s “dynamics play out in a structure that can’t be identified with [general relativity] spacetime”.^[81] Yet it is not yet clear how we could study or even model what grounds spacetime. Even physics, then, bids us expand our ontology when thinking in the direction of the kinds of thing that an ultimate ground of the cosmos could be.

Finally, why bother to fight so hard for an ultimate foundation to all things? Metaphysically, these foundationalist arguments bid the philosopher to rethink certain ideas that he habitually applies to the world, like “things”, “causation”, “explanation”, and “relations”. But psychologically and perhaps spiritually, foundationalism lends rational shape and perhaps some details to our sense that we are in the midst of a strangely coherent single phenomenon unfolding around and through us, taking its development from some source that has an extraordinary generative power. For some, foundationalism is the bridge between metaphysics and theology; regardless of one’s view on theism or atheism, religion or science, it marks out the widest and most basic truths we can find about reality as we know it. It is a theory of the ultimate understanding available to us.

More, the kind of philosophizing that we see at work in ontological foundationalism is meant to lead from the sensible, to the intelligible, to a “sublime” reality that stands beyond the rules of our world. Even Schaffer, who does not generally take his monism in theological directions, ends his consideration of global modal constraints with Elizabeth Barrett Browning’s earth “crammed with heaven, And every common bush afire with God”.^[82] One might say that the metaphysical nihilist aspires to freedom as a philosophically informed ethos of living, released from reliance on ultimate truths and liberated to let our linguistic conventions play out and to freely relinquish them once the show is done. But the monistic foundationalist aspires to wonder, using philosophy to develop a picture of that source that Dante describes when he says that at the end of his journey he saw:

...whatever

The universe unfolds; all properties

Of substance and of accident, beheld,

Compounded, yet one individual light

The whole.

And why ponder such things? Because the thought shapes the mind that thinks it:

...methinks I saw

The universal form: for that whenever

I do but speak of it, my soul dilates

Beyond her proper self...