Can the World Be Indeterminate in All Respects?

  • Chien-hsing Ho orcid logo (Institute of Chinese Literature and Philosophy, Academia Sinica)


Especially over the past twenty years, a number of analytic philosophers have embraced the idea that the world itself is vague or indeterminate in one or more respects. The issue then arises as to whether it can be the case that the world itself is indeterminate in all respects. Using as a basis Chinese Madhyamaka Buddhist thought, I offer two reasons for the coherence and intelligibility of the thesis that all concrete things are themselves indeterminate with respect to the ways they are. The first reason draws on a notion reminiscent of the picture of reality as an amorphous lump, while the second reason makes use of the relativity of conceptual perspectives and determinations. Assuming that the world is the mere totality of all concrete things, I show that there is a genuine metaphysical possibility of the world’s being indeterminate in all respects.

Keywords: conceptual relativity, ontic indeterminacy, worldly indeterminacy, Chinese Madhyamaka

How to Cite:

Ho, C., (2023) “Can the World Be Indeterminate in All Respects?”, Ergo an Open Access Journal of Philosophy 9: 22. doi:



Published on
31 Mar 2023
Peer Reviewed

1. Prologue

Many physical objects, such as mountains and rivers, appear to have fuzzy or indeterminate spatial boundaries. It seems indeterminate or vague when living organisms, such as humans and trees, start or cease to exist. If we are to come to an understanding of the nature of reality, it is critical that we account for the source or nature of these and other cases of indeterminacy. In the last century, the dominant approach in analytical philosophy took indeterminacy to be rooted in semantic indecision, in how we represent the world. A few philosophers took indeterminacy to be rooted in the limitations of our epistemic capacities, in how we know the world. These two approaches are committed respectively to semantic and epistemic indeterminacy, but both agree that the world itself is precise and determinate. In a contrasting trend, over the past twenty years a number of analytic philosophers have upheld the view that the world itself is indeterminate in one or more respects (henceforth, worldly indeterminacy).

The issue naturally arises as to how we should account for the nature of worldly indeterminacy. In this regard, I have developed an ontological conception of indeterminacy, termed ontic indeterminacy (OI), which centers on the two complementary ideas of conclusive indeterminability and provisional determinability.1 OI is originally based on the works of Sengzhao (374?–414 CE) and Jizang (549–623 CE), two leading Buddhist thinkers of the Sanlun tradition of Chinese Madhyamaka. It resembles the analytic-philosophical notion of metaphysical indeterminacy,2 whose advocates acknowledge worldly indeterminacy and offer varying accounts that aim to tackle such issues as the constitution of physical objects, their spatio-temporal boundaries, indeterminate identity and existence, vague objects and properties, and the open future.

Another, less natural but still significant, issue is whether it can be the case that the world itself is indeterminate in all or nearly all respects. Herein, I propose what we may call the OI thesis: that all things in the world are ontically indeterminate with respect to the concrete ways they are, such as their existence, nature, property, and form. By ‘all things’ I mean all concrete things (mountains, trees, humans, chairs, and quarks) that common sense and empirical science take to constitute the world. By ‘concrete ways’ I preclude abstractions, such as the property of being self-identical in the strict Leibnizian sense of ‘identical’. For the sake of argument and simplicity, let us assume that the world is simply the mere totality of all concrete things. (It then excludes abstract and imaginary objects such as numbers and unicorns. And I bypass the issue about the indeterminacy of the physical stuff of the world.) Consequently, if the thesis holds good, it is in an important sense true that the world itself is indeterminate in all respects.

In this paper, in addition to appealing to contemporary philosophical views, I partially draw on Sengzhao’s and Jizang’s Madhyamaka ideas to proffer two different reasons for the coherence and intelligibility of the OI thesis. The advocates of worldly or metaphysical indeterminacy generally approach the subject of indeterminacy from within the framework of metaphysical realism. By contrast, the framework that underlies the reasons, as we shall see, is broadly nonrealistic. The two frameworks still share much in common, while their differences might shed fresh light on indeterminacy.

Although I employ Chinese Madhyamaka ideas, I often freely develop them and the resultant thought may not be acceptable to the Madhyamaka thinkers. Consequently, the following discussion is intended to be a theoretical contribution that is interesting and worthwhile in its own right. With these in mind, reference to Chinese Madhyamaka will, whenever feasible, be relegated to a footnote. Significantly, my objective is not to argue for the truth of the thesis, but to show that it is coherent and intelligible, that there is a genuine metaphysical possibility of the world’s being indeterminate in all respects.

Section 2 sketches the basic ideas of OI. Sections 3 and 4 put forth two reasons showing the coherence and intelligibility of the thesis. Section 3 offers the first reason by exploring a notion somewhat reminiscent of Michael Dummett’s picture of reality as an amorphous lump. Section 4 offers the second reason by examining the relativity of conceptual determinations and exploiting Hilary Putnam’s doctrine of conceptual relativity. Section 5 concludes.

2. Basic Ideas of OI

In this section, I provide a sketch of OI, explaining the specific notion of indeterminacy employed here and the two associated ideas of conclusive indeterminability and provisional determinability.

Chinese Madhyamaka developed from Indian Madhyamaka, a prominent philosophical school of Buddhism reputed to have been founded by Nāgārjuna (c. 150–250 CE). On the ground that things in the world arise, abide, and perish in dependence on various causal and noncausal factors, Nāgārjuna contended that all things are devoid of nondependent and invariable nature or existence. This idea is neatly encapsulated in the Madhyamaka catchphrase ‘All things are empty’. However, due to translational and possibly cultural factors, the Chinese Mādhyamikas (Sengzhao and Jizang) tend to understand the catchphrase as meaning that all things are devoid of determinate nature and form. This amounts to saying that all things are indeterminate with respect to their nature and form.

For exegetical and philosophical reasons, this lack of determinate nature and form may best be explicated in terms of conceptual and linguistic indeterminability. Accordingly, I characterize the ontic (ontological) indeterminateness of things as follows:

Madhyamaka-based OI: For a thing X to be ontically indeterminate at time t with respect to the way it is (its existence, nature, property, form, etc.) is for X to be such that no concept or expression can be conclusively applied to X at t in the sense of representing definitively the way it is.

‘To represent a way X is’ means to determine X as containing a certain feature and make known in X that feature; the term ‘feature’, when used to characterize a thing, broadly signifies the particular ways the thing can be.

This characterization is expressed mainly in semantic terms, which may suggest that the indeterminacy involved is semantic rather than worldly. In fact, I take some kinds of semantic and epistemic indeterminacy to be present alongside worldly indeterminacy. Here I focus on the co-presence of worldly and semantic indeterminacy.3 According to Barnes, even if we take vagueness (or indeterminacy) to arise when our words do not latch onto the world appropriately, this does not rule out metaphysical vagueness because it is “coherent to think that vagueness is always such a mismatch [between words and world], but that in cases of metaphysical vagueness, the direction of explanation for the mismatch is from world to words, rather than from words to world” (2010: 955). So, we might explain the mismatch by saying that the world is itself vague such that our words fail to latch onto it. A similar view is expressed by Williams (2008). Barnes and Williams, like most theorists of worldly indeterminacy, think that worldly and semantic indeterminacy are not exclusive. Likewise, if we construe ‘indeterminacy’ in terms of a lack of definitive representability, the representation relation is the joint upshot of that which represents (linguistic representation) and of that which is represented (how the world is).4 For OI, the direction of explanation for this lack is from how the world is to linguistic representation, which signifies worldly indeterminacy.

To borrow an example from Shoemaker (Shoemaker & Swinburne 1984), suppose that Brown’s brain is transplanted surgically into Robinson’s skull such that the resultant person, Brownson, has Brown’s brain and psychology and Robinson’s body. Assuming that it is unsettled as to whether Brown survives the surgery as Brownson, OI would say that, after the transplant, Brown is not definitively representable by such expressions as ‘existent’ and ‘not existent’. Further suppose that this involves a serious legal case: if Brown survives the surgery, Brownson will be bequeathed $10 million; otherwise, the money will go to charities. Obviously, the judge cannot settle the case simply by demanding that the meaning of the predicate ‘existent’ be precisified. In fact, Brown’s objective nature is such that, after the transplant, he is not definitively representable by ‘existent’ or ‘not existent’, and this unrepresentability is due to the indeterminacy of his post-surgical existence. If, as I do, we want to ascertain the ontological nature of things, we should consider worldly indeterminacy. Besides, the Mādhyamikas locate the lack of determinate nature in the myriad things. Thus, it is meet to place emphasis on worldly indeterminacy.

If the word ‘tree’ represents definitively the way X is, it means that X is finally, nonrelatively, and exclusively a tree (containing the feature of being a tree). Yet OI implies that an ontically indeterminate X cannot be so represented. Likewise, no determination we may impose on X is definitive. Thus, X is not conclusively determinable—that is, it is not capable of being determined as definitively (or determinately) such-and-such or as definitively containing this or that feature.

Meanwhile, X is subject to multiple relativized determinations, to use a phrase coined by Wilson (2013: 367). We can determine X as such-and-such where the determinations are relativized to different conceptual perspectives and far from conclusive. These determinations are provisional in the sense that they are not definitive and do not predicate of X any determinate (viz., definitively representable) feature in such a way as to make it conclusively determinable.

Overall, OI centers on two complementary ideas, namely, conclusive indeterminability and provisional determinability: X is conclusively indeterminable (more precisely, not conclusively determinable), yet provisionally determinable, with respect to the way it is. These two ideas approach the same thing from negative and affirmative angles. ‘Conclusive indeterminability’ negates the conclusiveness of any of X’s determinations and characterizes X as failing to contain any determinate feature. ‘Provisional determinability’, in contrast, affirms the feasibility of relativized determinations of X, considerably accommodating our pretheoretical intuitions about the positive determinability of things. Combining and implementing the two ideas, we may have a sustainable account of worldly indeterminacy.

Consider an oft-used example for illustration. Suppose Tibbles the cat is losing a hair, such that at time t the hair is barely connected to his skin. The question then becomes: is the hair part of Tibbles at t? It seems difficult to give a clear-cut ‘yes’ or ‘no’ answer. Assuming this to be a case of worldly indeterminacy, we can state the situation as follows:

S1: The hair is ontically indeterminate at t with respect to its mereological relation to Tibbles.

The way the hair is at t in relation to Tibbles is not represented definitively by such expressions as ‘being part of’ or ‘not being part of’. Given conclusive indeterminability, the hair at t cannot be determined as determinately part or determinately not part of Tibbles. Given provisional determinability, we can, relative to different perspectives, provisionally determine the hair as part of Tibbles (say, since it is still touching his skin) or as not part of Tibbles (since it is hardly moored to his body).

3. Carving Up the Real

As said, the OI thesis takes all things to be ontically indeterminate with respect to the ways they are. In this and the next section, I aim to proffer reasons showing the thesis to be coherent and intelligible. For my purposes, I introduce as a theoretical postulate a distinction between conceptualization at the subconscious level and conceptualization at the conscious level. This section deals with a theme that concerns conceptualization at the subconscious level, while the next section proceeds from the conscious level of conceptualization.

For Madhyamaka, things are empty because they arise, abide, and perish in dependence on various causal and noncausal factors, which include our conceptualization. As things depend on conceptualization for their existence, the Chinese Mādhyamikas would reject the metaphysical realist view that things and their properties exist independently of our conceptual contributions. Indeed, they appear to hold that the myriad things are, to a moderate or considerable extent, molded by our conceptualization.5 Then if I am seeing a daffodil, the plant is, to some moderate extent, formed by my conceptualization. Some explanation is required.

At the conscious level, when we perceive something, we, by dint of conceptualization, differentiate the thing from adjacent things and other kinds of things, and cognize and determine it as, say, yellow-flowered, a jonquil, or a plant. Here our perception involves the deployment of concepts. For instance, if someone cognizes the thing as a tristylous daffodil, and she understands what a daffodil is and what it means to be tristylous, we can say she has the concepts of daffodil and tristylous for classifying the thing and applies them in her perceptual experience of it. This application involves conscious conceptualization, so to speak. If, as we shall see in Section 4, a given thing does not itself legislate what specific concept must be applied to it, it seems sensible to hold that the thing is originally indeterminate, yet the conscious deployment of concepts leads to its determination, its being given a specific identity and being considered determinately such-and-such. All this concerns our conceptual contribution at the conscious level.

Significantly, there is another kind of conceptualization that may be said to function at the subconscious level. Before explaining this, let us ponder what things would be like if we somehow managed to experience them without conceptualization of any kind. Assuming that things do not exist independently of our conceptual contributions, without conceptualization we would not experience things as we normally experience them to be. The demarcated boundaries and metaphysical joints that we normally posit among things would likely cease to be present. Here we may think of Dummett’s (1981: 563–66, 577) picture of reality as an amorphous lump, not yet articulated into discrete objects.6 (The notion of the real, which I introduce below, still differs significantly from this picture.) The reality prior to our conceptual contributions would be amorphous in the sense of being devoid of definite, conceptually representable boundaries and structures, waiting to be carved up conceptually into the things that common sense takes to constitute the world.

On my construal, this primordial conceptual carving, through which a structured world of particular things is formed, is done by conceptualization that functions at the subconscious level. (While some cognitive scientists claim that the brain constructs our perceptual world, we might say that the mind, which may emerge from the brain, subconsciously constructs the world.) As the conceptualization is not pretheoretically apparent to the conscious mind, we think of ourselves as existing in and experiencing an already structured, ready-made world, a world that contains naturally sliced res waiting to be captured by human concepts. In truth, the world as we commonly know it is, to some moderate extent, conceptually molded, although we cannot help but cognize it otherwise.

Let us speak of the reality before the conceptual carving as the real qua the way things really are. As can be inferred from above, the real is empty of definite boundaries, joints, and structures. In addition, because the real outstrips conceptual differentiations and is not properly matched by human concepts, it is conceptually indeterminable and ineffable, which means we are unable to describe (directly and properly express) the real as it is.7

To say that the real is indescribable may seem self-referentially inconsistent insofar as it appears to describe the indescribable. As Goodman (1978: 6) remarks: “Talk of unstructured content or an unconceptualized given or a substratum without properties is self-defeating; for the talk imposes structure, conceptualizes, ascribes properties.” This is a conundrum that one has to tackle if one speaks of something as ineffable.

In response, I submit that the real, while ineffable in the sense of being directly inexpressible, is indirectly expressible. I have elsewhere (Ho 2006) understood the notion of indication as an indirect mode of expression that involves an imposition-cum-negation method. In the sentence ‘X is ineffable,’ the word ‘ineffable’ may be said to directly express ineffability (as a semantic correlate). On the one hand, in using the sentence indicatively, this ineffability is being superimposed onto X such that one learns about the ineffability of X. On the other hand, this superimposition is simultaneously negated in such a way that one takes it as simply a provisional application and not as a real ascription. Thus, the ineffable X is intimated through the superimposition of ineffability and the negation of this superimposition. With the superimposition something about X is intimated; with the negation X is not made effable. Further, the result of this indirect expression of X may be compared to what we know about a landscape from an oil-painting of it: while there is no genuine correspondence between the painting and the landscape, the former still generates an inkling of the latter. On this account, an expression for the ineffable can be understood as indicative so that it somehow conveys information about the ineffable without directly representing the latter and falling into contradiction.8 As the real is indirectly expressible, and I take my expressions for it to be indicative, this talk of the real is not self-defeating. Moreover, it is widely held among contemporary philosophers that there is a close correlation between language and conceptualization. Analogously, in light of the correlation, we can indirectly or indicatively apply concepts to a directly unconceptualizable, unstructured reality without really imposing structure on it.

Although I related the notion of the real to Dummett’s picture of amorphous reality, the notion differs in that the real is not without preconceptual differences that outstrip conceptual imputations and contribute to the way the world is normally experienced by us.9 Indeed, that the real is devoid of conceptually representable joints, boundaries, and structures does not mean that it is completely formless and undistinguished. While conceptualization cannot carve the real at its joints (there are no joints) and inevitably involves conceptual imputations, what are called things in what is called the world are not merely conceptual constructs but rather originate from the coalescence of subconscious conceptualization and the real with its preconceptual differences, which account for the objective, mind-independent factor in experience. Hence, my use of ‘moderate’ to speak of things formed by conceptualization.

For example, in the real, what ordinary people come to cognize as a daffodil is not yet a daffodil, not a thing possessive of various features, not even a distinct thing out there. Given subconscious conceptualization, a cluster of preconceptual differences in the real are conceptually differentiated and structured such that a distinct thing forms in some space-time region, is differentiated from things in its surroundings, possesses various parts and features, and is to be referred to with demonstratives such as ‘this’ and ‘that’. Some of these features may lead one to apply the concept ‘daffodil’ to the thing, cognize it as a daffodil, and express it with the sentence ‘that is a daffodil’. A natural kind term such as ‘daffodil’ is usually thought to carve reality at its joints, say, dividing the world into daffodils and non-daffodils. While this conceptually representable division (this joint) may be partially traced back to certain preconceptual differences of the real, it also results from subconscious conceptualization and may be reinforced by human interests and practices. Simultaneously, although ‘that is a daffodil’ fails to represent the cluster of preconceptual differences concerned, the thing that the sentence describes does have its origin in the cluster. In addition, for ordinary people, the thing is expressed correctly by the sentence but not by sentences such as ‘that is a stick ablaze’.

Since worldly things originate from the afore-mentioned coalescence, the real does not differ greatly from the perceived world and we can be said to have glimpses of it. It is not Kantian things-in-themselves, hidden behind the appearances. It is instead the world when deprived of all describable boundaries, joints, and structures, so to speak. This discussion reveals that the present framework is not committed to an idealism that views everything as ultimately mental. Correlatively, OI would not become mired in an incoherent relativism such that no one conceptual perspective, relativism included, is more valid than any other (see Section 4).

In analytical philosophy, metaphysical realism can be characterized as the view that (most of the) things do not depend for their existence and nature on the conceptual activities of our minds. By contrast, according to Putnam’s version of antirealism (what he termed internal realism), the objects of our experience “do not exist independently of conceptual schemes. We cut up the world into objects when we introduce one or another scheme of description” (1981: 52; cf. 1987: 36).10 It makes no sense to speak of things in themselves and their intrinsic properties, apart from any contribution made by language or the mind. Notably, the present framework reflects a middle path between metaphysical realism and antirealism. Unlike antirealism, it acknowledges the notion of the real such that, in many cases, there can be an objective fact of the matter as to which conceptual perspective is relatively the better or best one to adopt.11 Unlike realism, things do not exist independently of our conceptual activities. Insofar as its rejection of realism is temperate, the framework may be depicted as a form of nonrealism rather than as an antirealism.

We now see how this discussion provides a rationale for taking the OI thesis to be coherent and intelligible. As said above, worldly things originate from the coalescence between the real and our subconscious conceptualization. The real qua the way things really are is closely related to worldly things and accounts for the latter’s final ontic status: it is the way things preconceptually and ultimately are. Since the real is indicated to be conceptually indeterminable, it is reasonable to maintain that all things are not finally conceptually determinable. In view of the intricate indeterminability of the way things really are, what we take to be things should naturally be susceptible to multiple perspectives—similar to how the colors of iridescent hummingbird feathers can be seen from different perspectives—such that they are not determinable as exclusively or nonrelatively such-and-such. Meanwhile, the resultant indeterminacy of worldly things is ontological because it is bound up with their ontic nature. Given that the notion of the real makes coherent and meaningful sense, we ought to acknowledge the coherence and intelligibility of the thesis.

I have offered the first reason in support of the thesis’s coherence and intelligibility. This reason hinges on the notion of the real. The next section presents the second reason, which hinges on the relativity of conceptual perspectives.

4. Relativity of Conceptual Determinations

In the previous section, I spoke conveniently of carving up the real into various concrete, particular things rather than into other types of things. This is not groundless. After all, it seems unlikely that people can sensibly be said to subconsciously carve up the real into such entities as numbers and the mereological sum of two trees and three chairs. This section focuses on the conscious level of conceptualization. While one can consciously carve up the world in various ways, I continue to assume conveniently that we are facing a world of particular things. Now, relative to different conceptual tendencies and perspectives, there can be diverse ways of viewing particular things and their relations at a more general level. For instance, some people tend to see things as enduring substances, whereas some others are inclined to see them as dynamic processes. In light of one conceptual perspective, things are loose and separate; in light of another, they are deeply interrelated and interdependent.12

At a more specific level, relative to different conceptual perspectives, a given thing can be variously cognized and determined with respect to the way it is: there can be seemingly incompatible, yet plausible determinations of the way the thing is. We have great difficulty picking out one perspective, among others, as the privileged perspective that induces the exclusively true representation of the thing. For example, what a villager takes to be a daffodil (call it X) may be food for slugs and snails, a stick ablaze for some meditating yogis, or a mass of wave-particles of indeterminate nature for a reductionist quantum physicist.13 The determinations involved may all be plausible. They are seemingly incompatible because they appear to ascribe different ontic statuses to one and the same thing. They are not truly incompatible insofar as we take the determinations to be relativized to different perspectives.

Furthermore, the fact that X can plausibly be said to be a daffodil signifies that it is not exclusively a mass of wave-particles, and vice versa. The fact that X is a daffodil only relative to a human, commonsensical perspective signifies that it is not nonrelatively a daffodil. Thus, we cannot determine X as definitively this or that. Hence S2:

S2: The thing X is ontically indeterminate with respect to its ontic status.

Here X is conclusively indeterminable with respect to its ontic status in that the status is not represented definitively by such expressions as ‘daffodil’, ‘a mass of wave-particles’, and so forth. X fails to have a determinate ontic status.

Granted that there can always be different perspectives on things, there can be different, but equally plausible, conceptual determinations of things such that none of the determinations is exclusively and nonrelatively correct. This consideration applies to all concrete things, and all concrete ways they are, which gives us a preliminary understanding of what it means to say that all things can be indeterminate with respect to the ways they are.

Let us substantiate the consideration by attending to one aspect of Putnam’s internal realism, namely, the doctrine of conceptual relativity, the gist of which can be seen in this passage:

[W]hat is . . . the same situation can be described in many different ways, depending on how we use the words. The situation does not itself legislate how words like ‘object’, ‘entity’, and ‘exist’ must be used. What is wrong with the notion of objects existing ‘independently’ of conceptual schemes is that there are no standards for the use of even the logical notions apart from conceptual choices. (Putnam 1988: 114)

Putnam means that different and prima facie incompatible descriptions of the same situation can all be true given the different uses of the words in question, and that how many and what kinds of objects and properties there are in the world is relative to conceptual scheme.

Imagine a world (call it w) with just three non-overlapping individuals. According to Putnam, different answers can reasonably be given to the question: how many objects are there in the world w? Common sense says ‘three’. Yet if a mereologist believes that for every two individuals there is an object that is their mereological sum, the answer would be ‘seven’. We then have these two descriptions:

D1: There exist three objects in w.

D2: There exist seven objects in w.

While D1 and D2 describe one and the same world and are prima facie incompatible, they can both be true when considered relatively: relative to a commonsensical and a mereological scheme, respectively. Alternatively, if you take D1 to be true and I take D2 to be true, we may be using the word ‘object’ in different ways (to mean ‘individual object’ and ‘mereological object’, respectively). On Putnam’s view, it is incorrect to think that there is some kind of metaphysical, mind-independent fact of the matter as to the truth of the descriptions (1987: 18–20).14

In the parlance of OI, D1 and D2 are compatible because they result from provisional determinations relativized to two different perspectives. They can both be said to be true if the determinations are plausible. We have as a special case S3:

S3: The world w is ontically indeterminate with respect to the number of its constitutive objects.

For Putnam, relative to different conceptual schemes (which reflect different practical interests), there can be a plurality of irreducible yet true descriptions of the same facts, while none of them can uniquely be said to describe the facts as they really are. He is concerned primarily with the number, existence, and ontic status of things, but seems ready to apply the doctrine of conceptual relativity to the various ways things are. Thus, Putnam writes: “Since I don’t think that any objects are totally mind-independent (or theory-independent), . . . on my view, objects and properties are, in general, vague too” (1983: 301). His point is perhaps that concrete objects and properties are vague because, depending on the conceptual schemes we adopt, they can be described variously as this or that. Clearly, Putnam’s emphasis here is on vagueness in the world, not vagueness in language or thought.

Excepting its antirealist underpinnings, Putnam’s doctrine of conceptual relativity is coherent and intelligible and can be appealed to for making sense of the OI thesis.15 We can say that none of the descriptions of a given thing can represent exclusively and nonrelatively the way the thing is. As the thing does not itself legislate what specific expression must be used to describe it, none of the descriptions can represent finally the way it is. Correlatively, the thing does not itself legislate what specific concept must be applied to it, and none of its conceptual determinations is conclusive.

Expectedly, there are differences between the doctrine and the nonrealist framework introduced in Section 3. According to Putnam, once we choose to adopt a certain scheme or a way of using certain words, there is a fact of the matter as to the truth of a description that hinges on the scheme—or fails to do so—or involves the words. For example, adopting the commonsensical scheme, D1 is true whereas D2 is false. (To evaluate their truth we need to check the empirical facts, and Putnam rightly insists that such facts are there to be discovered and not legislated by us.) However, there is no objective fact of the matter as to which scheme to adopt in the first place. Then, on a given occasion, which scheme to adopt to ascertain the truth of D1 and D2 is merely a matter of conceptual or linguistic convenience (Putnam 1987: 32–36; 2004: 238–43; Sosa 1993: 614–19).16

By contrast, the nonrealist framework acknowledges that in many cases there can be an objective fact of the matter as to which conceptual perspective is relatively the better one to adopt.17 To clarify this, consider the two ideas of OI: conclusive indeterminability and provisional determinability. Given conclusive indeterminability as well as the OI thesis, no description or statement, not even the thesis, is determinately true in the sense of properly representing a state of affairs that is actual and determinate. Here a state of affairs is determinate if the thing (or things) which it constitutively involves is conclusively determinable. Since nothing is conclusively determinable, there is no determinate and actual state of affairs. Consequently, there is no determinately true description. (OI acknowledges both first-order and higher-order indeterminacy such that, concerning the thesis, not only are all things indeterminate with respect to the ways they are, they are also not conclusively determinable as indeterminate. This acknowledgement coheres well with the idea of conclusive indeterminability.)

However, given provisional determinability, we can provisionally determine the way a thing is as such-and-such and the resultant description can be said to be true or false. Remarkably, different provisional determinations based on different perspectives are not necessarily epistemically equal: some determinations may seem plausible while others not so much. I submit that the plausibility of a provisional determination is to be judged with reference to experiential evidence, good reasons, convention, and conformity to the framework. The same goes for the truth of the resultant description. The point is that whether a given description is true or false may often depend on the way the world, and eventually the real, mind-independently is such that there can be an objective fact of the matter as to which perspective to adopt. Some illustrations may be helpful.

In the Tibbles example, perhaps there is no determinate fact of the matter concerning the truth of the description ‘the hair is part of Tibbles’. People with different perspectives have different views on whether to treat it as true or false, and no perspective stands out as the favorite. However, in view of the experiential evidence at t, we can reasonably judge that the description ‘the hair is indeterminately part of Tibbles’ is true, whereas ‘the hair is determinately part of Tibbles’ is false.

Now suppose that there is in front of us a tomato (what we are disposed to call a tomato). I say ‘the tomato is a fruit’ while you say ‘the tomato is a vegetable’. Which of the two descriptions is true? Presumably, the descriptions are based on two different conventional networks of concepts, and each description is true relative to the respective convention. Here which perspective to adopt to ascertain the truth of the descriptions is basically a matter of linguistic convenience.

Next, consider the existence of the tomato. An ordinary person would say that the tomato exists. In contrast, a compositional nihilist, for whom nothing composite exists, would say that the tomato does not exist, for it is composed of fundamental particles (whatever they are). A follower of Chinese Madhyamaka would prefer to say that the tomato is neither existent nor nonexistent. It is not existent because it does not exist independently from our conceptual contributions; it is not nonexistent because it is efficient and looks substantial. We then have these descriptions:

D3: The tomato is existent.

D4: The tomato is nonexistent.

D5: The tomato is neither existent nor nonexistent.

D3 is based on the conventional (commonsensical) perspective of viewing things, one that is assumed by common people. The nonrealist framework actually accommodates within itself this perspective insofar as we recognize its conventionality.18 D3 seems also supported by experiential evidence. By these lights, D3 can be said to be true. If we understand D4 in terms of compositional nihilism, it is to be taken as false. Even if the tomato is composed of some fundamental particles, as a functioning, cohesive whole it cannot be reduced to the particles or their mere collection: the functions that the tomato serves cannot all be served by the particles or the collection. Additionally, the particles, being imperceptible, can be more conceptually formed than the perceptible tomato. In any case, the tomato is no less real than the particles. Meanwhile, D5 can reasonably be said to be true on account of its nonrealistically signifying the tomato’s dependence on conceptualization and accommodating the conventional view that an efficient and substantial-looking thing is not nonexistent. Furthermore, we can interpret D5 as suggesting that the tomato is neither determinately existent nor determinately nonexistent, which would turn out to be correct. Thus, this example illustrates that there can be an objective fact of the matter as to which perspective to adopt, and the nonrealist perspective fares better than the nihilist one.

All this is to explain that while Putnam’s doctrine of conceptual relativity helps establish the present reason for the coherence and intelligibility of the OI thesis, we do not have to embrace the antirealist view that there is no mind-independent fact of the matter as to which conceptual perspective to adopt. However, I need to show that, despite the difference, it still makes sense to hold that none of the descriptions of things represents definitively the ways they are. For this purpose, we may consider D5 because of its intimacy with the nonrealist framework.

D5 does not nonrelatively represent the existential status of the tomato because its underlying determination is made relative to a specific perspective, one that dismisses the ideas of conclusive determinability and determinate truth. D5 does not exclusively represent the status because the fact that D3 can be said to be true more or less undermines D5’s claim to exclusive truth. (D3 and D5 are not true in precisely the same way, being resultant from two relativized determinations, so they are not contradictory.) D5 does not finally represent the status either. We saw above that a given thing does not itself legislate what specific concept must be used to apply to it. Then, it makes sense to say that the tomato is originally not existent or nonexistent, nor neither existent nor nonexistent, that it is we who, differentiating it from its contrary, apply to it such a concept as ‘existent’, ‘nonexistent’, or ‘neither existent nor nonexistent’ and further believe that the tomato must be so-and-so as implied by the concept. Consequently, the tomato is not finally representable by any such concept. From this discussion, we derive that D5 does not definitively represent the existential status of the tomato.

Given the explanation in Section 2, the indeterminate existence here revealed should belong to the tomato itself, and not merely concern how we use the concept ‘existence’. We can thus state the situation:

S4: The tomato is ontically indeterminate with respect to its existence.

The take-home moral is that no determination or description represents definitively the ways things are, yet it is not the case that any perspective works objectively as well as any other.

From above, we can consistently and meaningfully think that conceptual determinations of anything are only made relative to conceptual perspectives that are not privileged for definitively representing the thing. This spells the possibility that things are not conclusively determinable. In light of this reason, we can coherently make sense of the thesis that all things are ontically indeterminate with respect to the ways they are.

As a final note before leaving this section, some may say that to sustain the thesis, we must ascertain that there is indeterminacy in how things are fundamentally, in the basic building blocks of the world. If only the derivative, nonfundamental things, but not the fundamentals, are indeterminate, we cannot reasonably argue for the possibility that the world itself is indeterminate in all respects.19 Nevertheless, the present approach is not obliged to endorse the realist view that some, but not all, things are fundamental. It accords no ontologically privileged status to any things that may be considered fundamental, for such things are at least as conceptually formed as things that are deemed to be their derivatives. Moreover, even if we accept the view, it should be evident that the two reasons discussed in Sections 3 and 4 can be applied to any worldly thing, whether fundamental or otherwise. Therefore, I shall not pursue this issue further here.

5. Concluding Remarks

Using Chinese Madhyamaka thought as a basis, I have, in the foregoing, presented these two reasons for the coherence and intelligibility of the OI thesis:

  1. Assuming that the way things really are is not conceptually determinable, it makes sense to say that worldly things are not conclusively determinable.

  2. Because determinations of things are always relativized to (non-privileged) conceptual perspectives, any determination of things inevitably fails to be conclusive.

Some aspects of my arguments may not seem convincing to demonstrate the correctness of the thesis. However, I hope to have shown that the thesis is coherent and intelligible. Given the assumption that the world is simply all concrete things, we can reasonably contend that there is a real metaphysical possibility that the world itself is indeterminate in all respects.

Significantly, this possibility does not mean that we might be doomed to live in a world full of chaos and unsettledness. As seen above, this paper construes ‘indeterminateness’ in terms of conclusive indeterminability and provisional determinability. The idea of conclusive indeterminability suggests that we humbly acknowledge that the ways things are are not to be definitively ascertained once for all. Given provisional determinability, we can, as usual, engage in various determinations and continue to debate their plausibility and (relative) superiority. We acquire a fresh, albeit more complex, way of viewing the world, yet perhaps nothing of real value is lost.


Earlier versions of this paper were presented in several places. I would like to thank the audiences, especially Yasuo Deguchi, Shaoyong Ye, Duen-Min Deng, Kuan-min Huang, and Wei-chieh Lin, for helpful feedback. I am much indebted to Jessica Wilson and two anonymous referees for Ergo for their detailed and valuable comments.


  1. See Ho (2020). The sketch of OI in Section 2 draws on this paper considerably.
  2. For recent works that advocate metaphysical indeterminacy, see Williams (2008), Barnes and Williams (2011), Wilson (2013; 2017), Barnes (2014), Torza (2020), and Goswick (2021). In developing OI, I have incorporated a number of ideas from Jessica Wilson’s account of metaphysical indeterminacy. Some theorists also use the term ‘ontic indeterminacy’, but I restrict my use of the term to Madhyamaka-based OI. Both ontological and metaphysical indeterminacy are included under the heading ‘worldly indeterminacy’.
  3. Few advocates of worldly indeterminacy would dismiss the relevance of some kind of semantic indeterminacy. With the exception of quantum indeterminacy, cases of worldly indeterminacy normally involve how we classify things through linguistic conventions (Akiba 2014: 7). For example, advocates of worldly indeterminacy generally hold that mountains have indeterminate spatial boundaries. Yet this indeterminacy does not make sense to a linguistic community, if any, that does not demarcate and classify anything as a mountain. Presumably, most cases of indeterminacy are partially rooted in how we conventionally represent the world.
  4. Keil takes vagueness to arise from the mismatch between our representations and things themselves and offers this intriguing analogy: “Forget about representation for a moment and think of language as a tool: Sugar tongs are perfect tools for gripping sugar cubes. Now if you try to grip powdered sugar with sugar tongs, and it doesn’t work that well, then who is to blame, the sugar or the tongs?” (2013: 163). He implies that vagueness has no single ultimate source. As the idea of definitive unrepresentability runs somewhat parallel to that of unsayability, consider this analogy too. A theist in one of the Abrahamic traditions may claim the unsayability of God by citing two reasons: (1) God transcends the created world, of which language is a part; (2) language, fit for describing the world, is deficient in describing that which transcends it. These two equally sensible reasons concern, respectively, the nature of God and the nature of language, but together they explain the alleged ineffability of God. Similarly, definitive unrepresentability can be the result of the indeterminate nature of the world. I am grateful to an anonymous referee for pushing me to clarify this issue in some detail.
  5. Sengzhao (2016a: 338a2 [page 338, column a, line 2]) claims that “the myriad things and figures are all formed by the mind.” This claim is approvingly repeated by Jizang.
  6. This amorphous reality, which Dummett does not really endorse, is sliced up into distinct objects through application of language-related criteria of identity. My following discussion freely draws on an ontological reading of Sengzhao’s (2016b: 152a2–154c22) notion of ultimate truth/reality.
  7. According to Sengzhao (2016b: 159b20–21), for an entity to be effable (viz., directly and properly expressible) is for it to be endowed with a conceptually representable mode. Yet the real is devoid of such a mode. Alternatively, we can say that the subject-predicate structure of language expresses a semantic correlate that represents a division between a thing and its features, but the real is devoid of such a division.
  8. This account draws on a strategy offered by the fifth-century Indian grammarian-philosopher Bhartṛhari. The strategy was later adopted by the Madhyamaka philosopher Candrakīrti (c. 600‒650) and the Hindu philosopher Śaṅkara (c. 788–820). I am thankful to an anonymous referee for encouraging me to further clarify the account.
  9. Such preconceptual differences are ineffable and so differ from, and do not require the existence of, any metaphysical joints, which are describable. If one finds it hard to make sense of this notion of preconceptual differences, consider an analogous case: the notion of qualia as the felt, phenomenal qualities of experience. Were such qualities fully describable, Mary, the brilliant color scientist in Jackson’s (1982) knowledge argument, confined to a black-and-white chamber, would have known what it is like to see something red simply by learning all of the describable facts about color vision through black-and-white books and television screens. This suggests that the qualia of a given experience are in some sense indescribable. See Jakab (2000) for a related discussion. Since the qualia contain internal differences, we can analogously understand the notion of preconceptual differences. Incidentally, we must not take this analogy too far because the preconceptual differences, unlike qualia, are not mental.
  10. Putnam (1988: 107) once viewed his project as a third way between metaphysical realism and antirealism.
  11. When writing in my own voice, by “an objective fact of the matter” I mean a mind-independent (especially, independent of conscious minds) fact of the matter. Moreover, the fact of the matter, which is never determinate, pertains to the conscious level, though the objectivity involved points to the real or its preconceptual differences. The point I make in Section 4 is that, while no conceptual perspective is determinately the best one to adopt, whether a given determination is plausible may often depend on the way the world (and eventually the real) mind-independently is such that there can be an objective fact of the matter about which perspective to adopt.
  12. Many of these ways of viewing things are rooted in the grammatical structures of the languages we speak (say, Indo-European languages versus East Asian languages). Perhaps some language-based tendencies are so deeply embedded in the mind that they already function subconsciously; consequently, people of different linguistic communities may carve up the real in somewhat different ways.
  13. For Jizang’s more exotic examples, see Jizang (2016a: 897a14–29; 2016b: 81b5–8, 93c29–94a2). Jizang elsewhere (2016a: 894c14) writes: “A physical thing is originally neither existent nor nonexistent, yet [sentient beings] understand it as existent or nonexistent.” We see such an idea below.
  14. For criticisms of the doctrine, see Sosa (1993: 614–24) and Haack (1996: 301–8). Many of the criticisms are directed at the antirealist underpinnings of the doctrine. For instance, antirealism appears to imply that, before humans came on the scene, there were no stars, oceans, and trilobites, and this implication seems absurd. The nonrealist framework, by contrast, acknowledges that, before the rise of sentient beings, there was the real. Moreover, what we now know as stars and oceans might be said to have existed in a preconceptual, inchoate form. Despite my discussion of Putnam’s example, this paper is not concerned with the indeterminacy involved in the individuation of reality.
  15. My subsequent discussion does not assume the antirealist view that there is no objective, mind-independent fact of the matter as to the truth of any empirical description. Indeed, Wolterstorff (1987: 262–64) contends, against Putnam, that a metaphysical realist can hold that there are many, even endless, ways of individuating reality. So, even a realist might acknowledge that a mind-independent object is capable of multiple equally plausible determinations.
  16. Analytical metaphysicians have lately debated such meta-ontological issues as whether there is an objective fact of the matter about what objects exist. See, for example, Chalmers (2009).
  17. As indicated in Section 3, the preconceptual differences of the real account for the mind-independent factor in experience. We can then say that what perspective to adopt is in many cases not merely a matter of conceptual convenience.
  18. The Madhyamaka tradition advocates a doctrine of two truths, conventional and ultimate, such that even conventional ideas, ideas shared by common people, can be approved insofar as we recognize their non-ultimate conventionality. Here I exploit the doctrine for my own ends. Note that my discussion in this paragraph draws largely on Chinese Madhyamaka and may not seem persuasive to some readers. However, my purpose is merely to illustrate how there can be an objective fact of the matter as to which perspective to adopt.
  19. See Barnes (2014) for arguments in favor of fundamental metaphysical indeterminacy. Madhyamaka thinkers would refuse to accord any privileged status to the so-called fundamentals; see Arnold (2010) for relevant discussion on Indian Madhyamaka.


1 Akiba, Ken (2014). Introduction. In Ken Akiba and Ali Abasnezhad (Eds.), Vague Objects and Vague Identity: New Essays on Ontic Vagueness (1–21). Springer.

2 Arnold, Dan (2010). Nāgārjuna’s “Middle Way”: A Non-Eliminative Understanding of Selflessness. Revue Internationale de Philosophie, 64(3), 367–95.

3 Barnes, Elizabeth (2010). Arguments against Metaphysical Indeterminacy and Vagueness. Philosophy Compass, 5(11), 953–64.

4 Barnes, Elizabeth (2014). Fundamental Indeterminacy. Analytic Philosophy, 55(4), 339–62.

5 Barnes, Elizabeth and J. R. G. Williams (2011). A Theory of Metaphysical Indeterminacy. In Karen Bennett and Dean W. Zimmerman (Eds.), Oxford Studies in Metaphysics (Vol. 6, 103–48). Oxford University Press.

6 Chalmers, David (2009). Ontological Anti-realism. In David Chalmers, David Manley, and Ryan Wasserman (Eds.), Metametaphysics: New Essays on the Foundations of Ontology (77–129). Oxford University Press.

7 Dummett, Michael (1981). Frege: Philosophy of Language. Harvard University Press.

8 Goodman, Nelson (1978). Ways of Worldmaking. Hackett.

9 Goswick, Dana (2021). Odd Objects: LEM Violations and Indeterminacy. Erkenntnis, 86(6), 1615–33.

10 Haack, Susan (1996). Reflections on Relativism: From Momentous Tautology to Seductive Contradiction. Philosophical Perspectives, 10, 297–315.

11 Ho, Chien-hsing (2006). Saying the Unsayable. Philosophy East and West, 56(3), 409–27.

12 Ho, Chien-hsing (2020). Ontic Indeterminacy: Chinese Madhyamaka in the Contemporary Context. Australasian Journal of Philosophy, 98(3), 419–33.

13 Jackson, Frank (1982). Epiphenomenal Qualia. The Philosophical Quarterly, 32(127), 127–36.

14 Jakab, Zolán (2000). Ineffability of Qualia: A Straightforward Naturalistic Explanation. Consciousness and Cognition, 93(3), 329–51.

15 Jizang (2016a). A Treatise on the Profound Teaching of Vimalakīrti (Jingming xuanlun). In Taishō shinshū daizōkyō, Vol. 38, No. 1780. Taishō shinshū daizōkyō is cited from CBETA Chinese Electronic Tripiṭaka Collection, Version June 2016, Chinese Buddhist Electronic Text Association.

16 Jizang (2016b). The Meaning of the Two Truths (Erdi yi). In Taishō shinshū daizōkyō, Vol. 45, No. 1854.

17 Keil, Geert (2013). Introduction: Vagueness and Ontology. Metaphysica, 14(2), 149–64.

18 Putnam, Hilary (1981). Reason, Truth and History. Cambridge University Press.

19 Putnam, Hilary (1983). Vagueness and Alternative Logic. Erkenntnis, 19(1), 297–314.

20 Putnam, Hilary (1987). The Many Faces of Realism. Open Court.

21 Putnam, Hilary (1988). Representation and Reality. The MIT Press.

22 Putnam, Hilary (2004). Sosa on Internal Realism and Conceptual Relativity. In John Greco (Ed.), Ernest Sosa and His Critics (233–48). Blackwell.

23 Sengzhao (2016a). A Commentary on the Vimalakīrtinirdeśa Sūtra (Zhu Weimojiejing). In Taishō shinshū daizōkyō, Vol. 38, No. 1775.

24 Sengzhao (2016b). The Treatise of Sengzhao (Zhaolun). In Taishō shinshū daizōkyō, Vol. 45, No. 1858.

25 Shoemaker, Sydney and Richard Swinburne (1984). Personal Identity. Basil Blackwell.

26 Sosa, Ernest (1993). Putnam’s Pragmatic Realism. The Journal of Philosophy, 90(12), 605–26.

28 Torza, Alessandro (2020). Quantum Metaphysical Indeterminacy and Worldly Incompleteness. Synthese, 197(10), 4251–64.

29 Williams, J. R. G. (2008). Multiple Actualities and Ontically Vague Identity. The Philosophical Quarterly, 58(230), 134–54.

30 Wilson, Jessica (2013). A Determinable-Based Account of Metaphysical Indeterminacy. Inquiry, 56(4), 359–85.

31 Wilson, Jessica (2017). Are There Indeterminate States of Affairs? Yes. In Elizabeth Barnes (Ed.), Current Controversies in Metaphysics (105–19). Routledge.

32 Wolterstorff, Nicholas (1987). Are Concept-Users World-Makers? Philosophical Perspectives, 1, 233–67.