If the arguments I give in sections 2 - 4 of this paper are correct, they may lead to vindication of a position that has been called «blob theory»: there are no n-adic properties. The point of this paper is to show that blob theory may be the correct account of reality, due to hitherto undiscussed problems that I will point out to do with property possession in bundle theory and substance theory,Foot note 10_29 and that are so serious that I do not see a way to avoid blob theory.Foot note 10_30
In this paragraph I will give a brief synopsis of how I argue that problems to do with the theories of property possession lead to blob theory. As I will discuss in later sections, a property must be a property of something. There cannot be properties that are not possessed by something that they are properties of, for if there were, they would be properties that are not properties of anything (they would be properties that are not properties). That is how I come to the conclusion that properties do not exist, for the following reasons. If it could be argued that there are no currently available coherent theories of property possession, then the best account of reality we have is one where particulars do not have properties. If that is the case, then if there are any properties, they can only be properties that are not properties of any particular. But if there cannot be any such properties, as just mentioned, and as will be argued below, then there cannot be any properties whatsoever. Therefore, by focusing on problems to do with property possession, I am able to come to the conclusion that properties do not exist.Foot note 10_31
Before discussing problems with property possession, in this introduction I will give some introductory comments about blob theory (subsection 1.1), I will discuss the non-commonsensical nature of blob theory (subsection 1.2), and I will discuss property possession in theories of objects as given to us by metaphysicians (subsection 1.3).
1.1 Blob Theory
Blob theory follows from what some have called «extreme nominalism»:Foot note 10_32 n-adic properties do not exist. According to blob theory, reality is unstructured (it is an unstructured blob). Moreland lucidly describes blob theory:
Among other things, EN [extreme nominalism] is what has been called a blob theory regarding concrete particulars. A blob theory of ordinary concrete particulars is consistent with a mereological analysis of those particulars as wholes constituted by separable parts; but a blob theory renders concrete particulars structureless entities with no internal differentiation of properties and relations within those concrete particulars. In this sense, EN treats concrete particulars as simples and thereby fails to acknowledge that the redness, circularity, size and other features of [for example, a red ball] are real entities that are neither identical to each other nor to [the ball] as a whole.Foot note 10_33
(Extreme nominalism is not identical to Quinnean nominalism. The primary difference between the two is that extreme nominalism is identical to blob theory, and blob theory implies that, as I will discuss in subsection 1.2, nearly all our statements are false, whereas this is not the case at all in Quinnean nominalism.)
In sections 2 - 4, I discuss hitherto unnoticed problems to do with substance and bundle theories. I do not discuss blob theory in detail. Rather, my goal is only to argue for it by revealing hitherto unnoticed problems to do with property possession in analytic metaphysics.
1.2 Blob Theory in Opposition to Common Sense and Macroscopic Perception
Maddy writes that it is «quite clear» that blob theory does not describe our world:
...[A] purely physical world... lacking all... properties... is a world entirely without individuating structure: The Blob. Even if such a world is possible, it seems quite clear that it is not our world, and thus, that it would be of no particular interest to the physicalist.Foot note 10_34
In opposition to Maddy, I will present novel arguments which indicate that (i) metaphysicians apparently have not presented an account of property possession that is coherent, and for that reason, (ii) the position that there are no properties may be the best theory of reality we currently have. I will also argue in the conclusion, and in opposition to Maddy, that blob theory may be of utmost interest since it may be supported by theories of philosophic atomism.
Extremely counterintuitive theories of reality, arriving at conclusions similar blob theory have been discussed or argued for by many philosophers from Parmenides up to the present. For example, in an excellent paper, O'Leary-Hawthorne and Cortens discuss philosophical positions similar blob theory:
In Unger's earlier works, he also has argued for positions that arrive at conclusions much like blob theory:
If the reader is dubious about the project undertaken in this paper, due to the obvious disagreement that blob theory has with ordinary experience, I suggest that the reader still consider the arguments of this paper, for reasons I discuss next.
Serious consideration of accounts of reality that are entirely against our ordinary macroscopic experience of reality (perhaps even to the point of resembling a philosophy of the blob) is not uncommon in quantum physics, especially in theories of quantum gravity (string theory, M-theory, loop quantum gravity, etc.). (Quantum gravity theorists are not to be equated with quantum theorists in general, but rather they are to be thought of as a very specific subgroup of quantum theorists). Many theories of quantum physics, especially quantum gravity, are very much at odds with our commonsense understanding of reality, and our macroscopic ordinary perception of reality, including being in direct opposition with the rarely questioned and entirely commonsensical classical concepts of distance, time, space, motion, and so forth.Foot note 10_37 On this issue, Petitot and Smith write:
Things are apparently so counterintuitive at the most fundamental quantum domain of the Planck scale that many quantum gravity theorists are now telling us that at the fundamental level nature is «timeless» and «spaceless». Consider what Greene, a leading quantum gravity theorist, tells us about noncommutative geometry, the mathematics that in the future might be found to describe the smallest level of reality that physicists study: «On scales as small as the Planck length a new kind of geometry must emerge, one that aligns with the new physics of string theory. This new geometrical framework is called quantum geometry.»Foot note 10_39 Greene continues:
If reality is propertyless, then it is unstructured (undifferentiated) since, to give just one example, without part-whole relations (which are polyadic properties) there are no parts and wholes, and without parts and wholes, there is only one entity.Foot note 10_41,Foot note 10_42 The account of reality at the quantum domain given by Greene more closely resembles an unstructured reality more than it resembles a reality with structure. For this reason, the philosophy of blob theory may (notice I wrote «may») be a description of reality more in accord with the fundamental quantum domain than the metaphysical realistFoot note 10_43 descriptions of reality are. Greene's passage of course does not establish that quantum reality = propertyless reality. Greene's passage only perhaps suggests that a description of quantum reality may be more like a propertyless reality than it is like a structured reality. This may be an important point for readers who are dubious about blob theory, since despite the obvious conflict with commonsense, ordinary perception, and our macroscopic understanding of reality that blob theory involves, there may nevertheless be reasons for the reader to consider the philosophy of the blob, since, at the very least, blob theory might be an account of reality that is aligned more with the fundamental level of reality than any metaphysical realist account is.
I will not discuss quantum theories in this paper further than the comments just given, other than one more comment in the next paragraph. The goal of this paper is only to argue that blob theory is a serious theory that deserves consideration due to the novel arguments I present in sections 2 -- 4 against the established theories of property possession. In this paper, I only discuss problems with property possession, and I make no attempt to describe how or why humans experience an ordinary, common sense reality of properties if blob theory is the correct account.Foot note 10_44 I do not discuss why or how, if reality is propertyless, humans have experiences of properties, and have the experience of a world composed of properties grouped in a way where the properties give rise to ordinary objects. This paper is not about any issues except hitherto unnoticed problems to do with property possession in theories of ordinary objects.
The disagreement that blob theory has with common sense perception may appear to be a large issue to pass over, but I do not consider it so, for the following reason. Since quantum mechanics is entirely unable to explain why humans experience the commonsense macroscopic world of color patches and solid surfaces as they do, but is taken seriously, this appears to be evidence that fundamental reality is in opposition to commonsensical macroscopic reality, and thus in addition to quantum physics, blob theory should also be taken seriously. Since quantum experimental and empirical evidence leads to doubt about the correctness of our commonsense macroscopic perception of reality, it appears that if quantum physics is on track, we should be at ease when our philosophical reasoning leads to the position that our ordinary, commonsensical, macroscopic view of reality is incorrect. And it appears that we should be leery of theories that imply our ordinary, commonsensical, macroscopic view of reality is correct. This appears to imply that model of reality put forth by blob theory can be trusted more than the ordinary, commonsensical, macroscopic view of reality, due to the fact that blob theory may be more in line with an anti-commonsensical view of reality that denies ordinary perception and our macroscopic view of reality. Quantum theorists also ubiquitously avoid discussion of how or why ordinary common sense reality appears the way it does in light of the highly counterintuitive quantum theories.Foot note 10_45 Quantum theories are taken seriously even though they have enormous disagreement with commonsense, ordinary, macroscopic reality. Since quantum theories show that reality at the fundamental level might be largely unstructured, then I see no reason why blob theory, which arrives at a similar picture of reality, cannot also be taken seriously.Foot note 10_46 If my arguments in this article are correct, then blob theory is not only more in accord with the most progressive areas of quantum reality (namely quantum gravity), it is also more coherent than the metaphysical realist theories since my arguments in sections 2 -- 4 apparently show metaphysical realism to be incoherent.
1.3 Substances that are Bundles, and Non-Bundle Substances
Before moving to my arguments, I will briefly discuss theories of ordinary objects, according to the metaphysical realist account. In this subsection, I will discuss that according to metaphysical realism, objects are substances or bundles: perduring or enduring particulars that exemplify properties, or bundles of compresent properties. These are the two available analytic metaphysical realist theories of ordinary objects that are considered by the contemporary philosophers who describe and discuss in great detail the nature of property possession by ordinary objects. (I do not discuss Quinnean nominalism since Quinnean nominalism involves some sort of property possession -- all particulars possess the polyadic property, set membership -- and for that reason, the problems I discuss to do with bundle theory and substance theory apparently also apply to Quinnean nominalism.)
When one looks out into the world, as it is given to phenomenal consciousness, one is presented with attributes, and they come in groupings which the metaphysical realist typically calls «substances», «bundles», «ordinary objects», or «things». For example, the lion (substance) is a grouping of attributes: felinity, four-leggedness, goldenness, sublimity, ferocity, etc. The metaphysical realist's task is to figure out how it is the case that properties can amass, conglomerate, bundle, or group to coherently (non-contradictorily) give rise to objects. In other words, the metaphysical realist is very concerned with the question: If we know that properties exist, then how do properties interrelate, interconnect, hold together, or tie together, to give rise to objects, where objects in turn give rise to nature? This comprises a major principle of the metaphysical realist: To understand reality, one must understand how reality is composed of objects, where objects can be considered groupings, collections, assemblages, or sets of attributes that are coherently interrelated and connected either to each other (this is called the bundle theory, which I discuss below), or to an item that holds the properties together (this is the non-bundle substance theory, which I discuss below).
The metaphysical realist tells us that the difference between attributes and ordinary objects which are not attributes, is: ordinary objects have (possess) attributes (the lion has ferocity, the electron has structurelessness,Foot note 10_47 a galaxy might have the property of being a quasar); and an object is not had (possessed) by anything else (this will be discussed in more detail in later sections).Foot note 10_48
There are two major theories of objects: bundle theory and substance theory. According to the bundle theory, properties are tied to one another. According to substance theory, properties are tied not to one another, but rather are tied to an entity that is a non-property. Many metaphysical realists use the word «substance» to denote either the substances of substance theory, or the bundles of the bundle theory of objects.Foot note 10_49 Bundles are often called «substances that are bundles of properties»; and substances are often merely called «substances», but I will call them «non-bundle substances». The non-property item that properties tie to in non-bundle substance theory will also be discussed in detail in sections below; for now, it can be called a thin particular, or an internally bare particular.Foot note 10_50 Armstrong discusses what is meant by «internally bare particular», or «thin particular»:
(I will discuss in section 2 that many metaphysical realists hold that properties of the non-bundle substance are not tied to a bare non-property (internal bare particular or thin particular), but rather are tied to what they call a «thick particular». I will give an argument showing that this position is apparently incorrect, where my argument is based on the issue that the definition of «thick particular» has not been outlined specifically by metaphysical realists. Often the issues of bare or thick particulars are not addressed in the literature, and rather than specifying if the thin or thick particulars are of concern when talking about property possession, philosophers will merely write that «particulars exemplify properties» («x has F») (e.g., an electron has charge, Jones is present, God is omnipresent, etc.) without discussing the issues that those who are specialists in the philosophy of property possession bring up to do with bare or thick particulars.)
Properties are held to the thin particular in the case of substance theory, or to one another in the case of bundle theory, by special ties, often referred to as «predicating ties». The tie in substance theory (where properties do not tie to one another, but tie to the internally bare particular) is called the exemplification tie; and in bundle theory (where properties tie to one another) the tie is called the compresence tie (or it is sometimes called the compresence relation). These ties are discussed more in sections below.
Bundle theories, and non-bundle substance theories, are the only theories given to us by metaphysical realists. Loux discusses how there are very few theories of ordinary objects offered to us by metaphysical realists:
What I want to point out about these passages, is how he discusses that there so «few options» for theories of property possession by ordinary objects, and that the options only consist of bundles or non-bundle substances. In the above passage, Loux mentions bundle theory, Aristotelianism, and substratum theory (the latter two are non-bundle substance theories). In a passage I cite next, Loux mentions the remaining major branch of metaphysical realism, which is the platonistic account of property possession by ordinary objects (platonists are either bundle theoristsFoot note 10_53 or non-bundle substance theorists).
In sections 2 - 4, I discuss hitherto unnoticed problems to do with substance and bundle theories. I do not discuss blob theory in detail. Rather, my goal is only to argue for it by revealing hitherto unnoticed problems to do with property possession in analytic metaphysics.
2. Any Non-Bundle Substance Involves an Internally Bare Particular
In this section I will find that, contrary to what many metaphysicians typically tell us, non-bundle substances can only be considered in terms of a bare particular: the properties of a non-bundle substance attach to a non-property: an internally bare particular.
Since not all non-bundle substance theorists hold that there are such bare and thin particulars, in this section I present an argument for the position that any non-bundle substance can only involve an internally bare thin particular. (I will discuss in more detail what is meant by «internally bare particular» in section 3. But in section 3, I will not be primarily concerned with the problems to do with bare particulars that are typically discussed in the literature, such as: How can an entity be (internally) propertyless?) If this reasoning is correct, I will discuss in the next section that there are hitherto unnoticed problems for the tying of (exemplifying of) properties to internally bare particulars. But it is a long while until I arrive at that conclusion; many issues need to be discussed before that.
In showing that any non-bundle substance can only be considered in terms of bare particular, I will discuss the exemplification tie (subsection 2.1), thin and thick particulars (subsection 2.2), the nature of properties (subsection 2.2), and property possession (subsection 2.3). As I discuss these issues, my arguments that any non-bundle substance can only be described in terms of a bare particular will be revealed. After this, in section 3, I will be able to discuss the specific problems to do with property possession in non-bundle substance theory.
2.1 The Exemplification Tie
In this section I will discuss the non-relational tie of exemplification that holds a property and particulars together in the non-bundle substance account of property possession by ordinary objects. (In this subsection of this section, I will refer to the «particular» rather than the «bare particular», or «thin particular», as the item that exemplifies properties on the non-bundle account of substances; and only after I establish that only bare particulars can be the items that exemplify properties on the non-bundle account of substance will I refer to all particulars as «bare particulars».) On the non-bundle substance account, properties do not directly attach to the particular. Rather, I will discuss in this subsection that metaphysical realists tell us there is a tie, called the exemplification tie, acting as an intermediary between properties and particulars. In other words, properties do not directly attach to particulars, but rather the intermediary of the exemplification tie directly attaches to the properties and particulars. Loux lucidly explains these issues (and he discusses a few other issues to do with the details involved in the exemplification tie being an intermediary between particular and properties, which I will discuss after Loux's passage):
(Note that Loux mentions that the exemplification tie is to be considered ontologically primitive. This issue is not important to the reasoning of this paper since I do not make any attempt to analyze the exemplification tie (I only inquire in 3.2 as to whether or not the tie has properties), but it is interesting to note that, at least to my knowledge, specifically why the tie is to be considered primitive has never been argued for, but rather it has merely been assumed by many philosophers that the tie is not to be analyzed. This assumption arose after Bradley's regress revealed problems to do with the theories of property possession. Placing an unanalyzable tie into the metaphysics of property possession removes the regress, but this, it could be objected, is merely an ad hoc solution, perhaps covering up deeper problems with the currently accepted theories of property possession. I discuss such deeper problems in much more in three recent papers (2003, 2004a, 2004b). But more importantly, my reasoning in this paper about problems with theories of property possession follow whether the exemplification tie is primitive or not.)
On the non-bundle substance account, properties are considered to not involve a relational or direct attachment to the particulars that exemplify them. If they did, a Bradley-esque regress would ensue. Rather, properties, and the particulars that exemplify them, according to non-bundle substance theories, are typically considered to be mediated by a non-relational tie, called the exemplification tie, and for that reason, Bradley's regress is avoided.
There are two types of attaching that metaphysical realists are concerned with when they discuss the property possession of substances: (i) intermediary attachment, which is the sort of attachment between property and particular by way of the non-relational intermediary tie of exemplification, and (ii) unmediated attachment, which is the sort of attachment between the exemplification tie and property, and the exemplification tie and the particular. The non-relational exemplification tie stands between property and particular, mediating the particular and its properties, and for that reason, the attachment between property and particular is an intermediary attachment. Unmediated attachment is a kind of attachment that entities are involved in where an intermediary is not involved. Let «unmediated attachment» denote the way that the exemplification tie attaches to both the property and particular. Unmediated attachment is an attachment that is not a relation that the exemplification tie has with the property and particular. Unmediated attachment does not involve any sort of entity (intermediary) that is between the exemplification tie and the property, or that is between the exemplification tie and the particular. Rather, properties and particulars each involve an unmediated attachment to the non-relational exemplification tie, and the exemplification tie, in turn, involves an unmediated attachment to the particular, and to the properties that the particular exemplifies.
2.2 Properties are Ways
In the next two subsections I discuss properties. Specifically, I will inquire as to what properties are, and what, exactly, properties tie to (via the exemplification tie) when they are exemplified by a particular. In other words, I will be concerned with the issue of specifically what it is about a non-bundle substance that the exemplification tie involves an unmediated attachment with in its unmediated attaching to a particular.
A property is a way any substance (bundled or non-bundled substance) is: a property is what a substance is like. In discussing substances (Heil uses the word «objects» instead of the word «substances»), Heil discusses the differences between properties and the objects that have them:
Armstrong also discusses how properties are ways objects (substances) are:
If a property is a way that a thing is, then this brings the property into very intimate connection with the thing, but without destroying the distinction between them.Foot note 10_57 (Emphasis added.)
I have italicized the last sentence of Armstrong's citation to stress that, aside from the exemplification tie, there are two distinct entities (in the broadest sense of the word «entity») that must be involved when a particular exemplifies a property: there is (1) the property, and there is (2) some other entity that the property is a property of. If it were not the case that property possession involved two distinct entities -- property, and the particular that the property connects to (via the exemplification tie) -- a property would not be a way a particular is. A property not connected to another entity (via the exemplification tie) is an impossible entity:Foot note 10_58 it is a way that is not a way a particular is.Foot note 10_59 For there to be properties, there must be two entities (in the broadest sense of «entity») that are non-identical, where one is tied to (linked to,Foot note 10_60 borne by) the other. This issue, which is integral in my arguments in this article, has been ignored by philosophers -- especially substance theorists who assert that property and particular are not necessarily distinct.
2.3 What Specifically Do Properties Attach to (Via the Exemplification Tie)?
On the non-bundle substance account, since, as discussed in the last subsection, there must be two distinct items involved in property possession (a property tied to a particular that is entirely distinct from the property), there must be a specific entity that the exemplification tie is involved in an unmediated attachment with. In this subsection, I inquire as to what properties of a non-bundle substance are connected to (via the exemplification tie). I consider this a rather simple inquiry to make, for one reason: any of the first-orderFoot note 10_61 properties of a non-bundled substance must attach to some entity, but if they attached to each other, the substance would be a bundle, thus first-order properties must attach to a non-property. (This issue, which is integral in my theorization in this article, has been ignored by metaphysical realists.) If a substance is not a bundle, there must be something about the non-bundle substance that is not a property, and it is that «something» that the first order properties attach to (via the exemplification tie).
Many metaphysicians who are non-bundle substance theorists tell us that non-bundle substances do not involve a bare particular. They tell us that we «cannot get below the concept of a concrete particular».Foot note 10_62 On the Armstrongian account of non-bundle substances,Foot note 10_63 some Aristotelian accounts of non-bundle substances, and platonistic accounts of non-bundle substances, properties are widely held to be properties of thick particulars,Foot note 10_64 as when we say: «the lion (thick particular) is sublime (property)»,Foot note 10_65 since «lion» may appear to refer to a complex of properties, and not just to the non-property entity that properties tie to. («...is...» denotes the exemplification of the property.) Loux discusses this:
It is fair to say that most non-bundle substance theorists who are not bare particular theorists are not entirely clear on specifically which non-property entity it is in non-bundled substance theory that properties are connected to (via the intermediary tie of exemplification). Consider a passage from Armstrong's widely discussed book, A World of States of Affairs:
Armstrong continues over the next several pages without telling us, or even hinting to us, what specifically it is about the thick particular that any first-order property of a thick particular attaches to (via exemplification). Consequently, no definition of what specifically the exemplification tie attaches to via unmediated attachment is found, and it appears that regardless of Armstrong's motivations, the thick particular can only involve a thin particular and the exemplification tie being involved in an unmediated attachment.
The reasoning above, if correct, indicates that on the non-bundle account of substance, there must be an entity that is not a property, that has no properties in itself, and which all the first-order properties of the non-bundle substance tie to. Non-bundle substance theorists who are not bare particular theorists however not only fail to tell us what that entity is, but they also inform us that there is not an internally bare non-property involved with any substance, such as the sort of internally bare particular Loux discussed in an above passage. Non-bundle substance theorists who are not bare particular theorists, such as Davis and Armstrong, do not include a bare particular, and thus their account appears to be incorrect, if my reasoning above is correct.
A synopsis of the argumentation I have given in this section is as follows. The first-order properties of any non-bundle substance must be tied to an internally bare particular. The internally bare particular cannot be composed of properties, for if it were, the non-bundle substance would be a bundle. A property is a way some other entity is, which is to say that the property must be tied some other entity. In the case of a first-order property in non-bundle substance theory, the other entity that a first-order property is linked to apparently can only be a thin particular that is internally bare. Any non-bundle account of a substance is a bare particular account. For these reasons, I will hereafter only discuss bare particular accounts of non-bundle substance.Foot note 10_68
3. The Unmediated Attachment of Propertyless Entities
In this section I next discuss problems with the unmediated attachment of an internally bare particular and the exemplification tie. I will discuss that this unmediated attachment is an unmediated attachment of propertyless entities (in the broadest sense of the word «entity»), which leads to serious problems for non-bundle substance theory. Since the exemplification tie is considered to be unanalyzable, this unmediated attachment between the internally bare particular and the exemplification tie is typically also considered to be unanalyzable. Moreland writes that «[i]t is a primitive fact that properties are tied to [bare particulars] and this does not need to be grounded in some further capacity or property within them.»Foot note 10_69 In writing about this passage from Moreland, Davis has claimed that «[b]y embracing 'tied to' predication we have averted both the incoherence and infinite regress objections [that have been posed against bare particulars] in one stroke».Foot note 10_70 But if my arguments in this section are correct, there is an apparently fatal problem that is not addressed in the literature regarding the primitive unmediated attachment between the internally bare particular and the exemplification tie. Regardless of whether or not such an unmediated attachment is primitive or not, my argumentation will show that the mere issue of there being an unmediated attachment between the internally bare particular and the exemplification tie lead to apparently fatal problems for property possession in non-bundled substance theory and the bundle theory of substance.
I will first discuss what has been called the «internal nature» of bare particulars (3.1). Then I will discuss the whether or not the exemplification tie itself has properties (3.2). Those issues will enable me to discuss whether or not the exemplification tie can coherently be involved in an unmediated attachment with the internal nature of the bare particular (3.3). I will find that it cannot, and I will discuss that for that reason, the non-bundle account of property possession by ordinary objects is impossible.
3.1 Bare Particulars
Bare particular theorists commonly tell us that the bare particular is internally bare, not externally bare: it has no properties in itself, but it is tied to properties that are distinct from it. Moreland and Pickavance write:
Bare particular theorists discuss an internal and external nature of the bare particular. The external nature is not propertyless; the internal nature is propertyless. A non-bundle substance might be considered in terms of the following diagram, where a non-bundle substance consists of an internally bare particular, IS, which exemplifies the properties A, B, C, and D:
C Is A
A, B, C, D are first-order properties
IS is the internal nature of the bare particular
We can consider (i) the entire non-bundle substance, which is the bare particular along with the properties the bare particular has. This is the internal and external nature of the bare particular, which is, of course, not a propertyless item. But since the bare particular must be distinct from the first-order properties of the non-bundle substance, we can consider (ii) the bare particular's internal nature, represented by the encircled «IS» at the center of the above diagram. IS, in itself, is propertyless (where «propertylessness» somehow does not denote the property, propertylessness), and is entirely distinct from A, B, C, and D. Armstrong discusses (i) and (ii):
Suppose that a instantiates F, G, H,... They comprise the totality of a's (nonrelational) properties. Now form the conjunctive property F&G&H. ... Call this property N, where N is meant to be short for a's nature. a is N is true, and a's being N is a (rather complex) state of affairs. It is also the thick particular. The thick particular is a state of affairs. The properties of a thing are «contained within it» because they are constituents of the state of affairs...
The «internal» nature (IS), to use the word Moreland used in a passage above, is what I am concerned with. In this section, I am concerned specifically with the issue of the bare particular's being involved in an unmediated attachment with the exemplification tie.
IS is the literal possessor of first-order properties, and IS is tied to first-order properties of the non-bundled substance via the exemplification tie. Regardless of whether or not the exemplification tie has properties, I will discuss that the unmediated attachment of the exemplification tie with IS will involve an unmediated attachment of propertyless items. It is this sort of unmediated attachment that I will be concerned with near the end of this section. I will argue that this unmediated attachment leads to a fatal problem for property possession in non-bundle substance theory. Before discussing this unmediated attachment, I will discuss the issue of whether or not the exemplification tie has properties.
3.2 Does the Exemplification Tie have Properties?
If the exemplification tie were also propertyless, the unmediated attachment of IS and the exemplification tie would be an unmediated attachment of propertyless items, which is an unmediated attachment I will later describe as fatal for metaphysical realism. If this is the case, it may be better for the metaphysical realist to consider the exemplification tie to have properties.
If the exemplification tie did have properties, it would be a substance, for the following reasons. The exemplification tie is not a property of a substance (it is not a way that a substance is), and for that reason, it is unexemplified. Entities that are not exemplified, but which also have properties, are primary substances. If the exemplification tie is a primary substance, it would have an internally bare particular that its first-order properties tie to. I will call the exemplification tie's internally bare particular, IE. In being involved in an unmediated attachment with a property and with IS, for reasons I discuss in the remainder of this subsection, it is not the entirety of the exemplification tie (IE and all its non-relational properties) that is involved in an unmediated attachment with IS. Rather, it is merely IE that involves an unmediated attachment with IS, rather than the entire exemplification tie substance (IE + all non-relational properties IE has). I will next explain this.
In being involved in an unmediated attachment with IS, we can ask: What is it about the exemplification tie that is involved in an unmediated attachment with IS? As discussed, the way properties attach or contact other items is apparently only by their being tied to other items. Items that properties tie to (are exemplified by) are made a certain way by being tied to the properties. The apple is red because the apple's internally bare particular (call this IS-apple) is tied to the property, redness. In considering the case where the red apple is on the brown table. The redness does not tie to the brown table's internally bare particular (call this IS-table), since redness is only tied to the apple's IS, in this situation, and assuming the table has no red in it at all. Since an item contacting the property will be the way the property makes it, if the apple touches the table, it cannot be the properties of the apple or the table that are involved in the touching when the apple and table touch. If the properties of the apple and table were the items that are in contact with the table when the apple and table touch, then the brown (not-red) table could «touch» redness, and the table would be brown (not red) and red. If the table touches the apple, it cannot be any of the properties of the table or apple that are involved in the act of touching, and for that reason, it is only the inner natures of the bare particulars of the apple and table, IS-apple and IS-table, that touch in the case where a table touches an apple. (Notice that this is an unmediated attachment of propertyless entities, which is the sort of unmediated attachment I am concerned with in this section, and which I am going to show is apparently problematic for non-bundle substance theory.)
For reasons just given, it is apparently only the internal bare particularities of the table and apple (IS-apple and IS-table) that «touch» when objects, such as tables and apples, touch. For the same reasons, the unmediated attachment of the exemplification tie and IS is an unmediated attachment that does not involve any of the properties of the exemplification tie, or of the non-bundle substance, since properties are not the items that are involved in unmediated attaching when a particular has an unmediated attachment with a particular. Rather, it is only IS and IE are involved in an unmediated attachment in the case where IS is tied to a property.
3.3 The Unmediated Attachment of IS and IE
I will continue discussing the unmediated attachment of the exemplification tie and IS, which is an unmediated attachment between IS and IE. In linking the property to IS, if the exemplification tie is a primary substance, it is only IE that is involved in an unmediated attachment with property and with IS. This would be an unmediated attachment of propertyless entities;Foot note 10_73 an unmediated attachment between the propertyless inner natures of two bare particulars. It is this sort of an unmediated attachment, one between propertyless entities, that I will be concerned with for the remainder of this section.
If all my reasoning to this point is correct, property possession in non-bundle substance theory must involve an unmediated attachment of propertyless entities. There are two ways this can occur. (i) IS involves an unmediated attachment with IE (if the exemplification tie has properties and is a primary substance, as I discussed in the paragraph before this one). Or (ii) IS involves an unmediated attachment with a propertyless exemplification tie (if the exemplification tie is propertyless). For the remainder of this subsection, I will be interested in how, exactly, propertyless entities can be involved in an unmediated attachment.
Propertyless entities are typically considered to be unstructured. I take it that, although it is hard to discuss propertyless entities at all, it is coherent to maintain, at the very least, that propertyless items can be described as be totally unstructured. If propertyless items are unstructured, they are point-sized (the size of a point, spatially unextended), and simple,Foot note 10_74 lest their having any spatial extension or physical parts give them some sort of structure.
I will next discuss the unmediated attachment of propertyless items. Propertyless, point-sized entities (such as IS and IE) that are involved in an unmediated attachment, in doing so, must collocate (spatially overlap), in order to involve an unmediated attachment. If any point-sized entities (such as IS and IE) do not spatially collocate, then there would be a spatial distance between them, and any distance between them would forbid them from being involved in an unmediated attachment. (To reiterate, propertyless entities can only be involved in a non-relational unmediated attachment. The unmediated attachment between propertyless entities cannot be a relation, since if propertyless items were interrelated, they would not be propertyless, but would share a relation, which is a polyadic property.) Since propertyless entities can only be partless,Foot note 10_75 an unmediated attachment of propertyless entities involves the unmediated attaching of their «totalities» (their «entireties»). The issue I am concerned about is this: The unmediated attachment of the entirety of two propertyless, partless, unstructured, collocated, point-size, also gives rise to a point-sized, unstructured, partless, point-sized, propertyless entity. In other words, the two propertyless entities involved in an unmediated attachment, upon their unmediated attaching, where their «entireties» are involved in an unmediated attachment, result in an entity (in the broadest sense of «entity») that is also a propertyless, partless, point-sized, unstructured, point-size entity. The two items cannot attach to form a mereological entity -- a whole with two parts -- since propertyless items must be devoid of a part-whole relation since a relation is a polyadic property. The two propertyless entities appear to amalgamate and unify to the point of becoming identical. There is no difference between one of the propertyless entities before the unmediated attachment, and the unmediated attachment of both of the entities. For these reasons, if the exemplification tie and the bare particular are involved in an unmediated attachment, which requires that IS and IE be involved in an unmediated attachment (or which requires that IS and a propertyless exemplification tie be involved in an unmediated attachment, if the exemplification tie is propertyless), upon their being in an unmediated attachment, IS and IE cannot be distinct: they are a single entity that is partless, point-sized, propertyless, and unstructured. IS and IE become identical, which means that the exemplification tie and the bare particular become identical. Without a difference between the exemplification tie and a bare particular of the substance, there are no distinctions between properties and the particular that has the properties, and the statement, «the exemplification of a property», has no meaning.
4. Substances that are Bundles of Properties
Bundle theories offer an alternative description of ordinary objects, and, as discussed in the introduction, bundle theory is apparently the only other theory about the property possession of ordinary objects that we are currently offered by metaphysicians. In this section I will be concerned with the unmediated attachment of the compresence tie and the properties that the tie bundles. In addressing this issue, I will first need to discuss the compresence tie that bundles those properties and that the compresence tie is propertyless (subsection 4.1). After that I will discuss a few issues to do with the properties of a bundle (subsection 4.2), whereby I will be able to discuss that the bundle theory involves an unmediated attachment of propertyless entities (subsection 4.2).
4.1 The Compresence Tie is Propertyless
Properties also have properties:Foot note 10_76 first-order properties instantiate second-order properties. When considering that properties have properties, on the bundle account, since properties are property-bearers (i.e., they are items that are composed of sets of compresent properties), properties can be considered secondary bundles. On this account, properties (secondary bundles) can be considered as bundles that are compresence with other secondary bundles (properties); whereas things (primary bundles) can be considered bundles that are not compresent with other bundles.
Secondary bundles (properties) require a compresence tie to bundle the properties that make up the secondary bundle. I will call this compresence tie that bundles the properties of the secondary bundle, TP. TP would have to be propertyless lest a vicious regress ensue: If TP has properties, since TP is not bundled, TP would also be a bundle, and would itself involve a compresence tie, TP2 that bundles TP's properties. TP's compresence bundle, TP2, also requires a compresence bundle, TP3, where TP3 requires TP4, and so forth, and a regress that is vicious may ensue, for the following reasons. If any bundle (such as TP) is bundled by another bundle (such as TP2), and so on, at every stage of the regress, the bundle at one stage is held together by another compresence bundle at the next stage. Each bundle stage depends on the next bundle stage of the regress. At any stage of the regress, the bundle is reducible to (i) the compresence bundle (TPN) and (ii) the properties that the compresence bundle bundles. Any bundle is only a bundle because of the existence of a second bundle, where the second bundle is only bundled due to the existence of a third bundle, ad infinitum. If properties of any stage of the bundles regress are bundled by the next bundle in the regress, never in the regress is there a point where the properties that are bundled are not dependent on other bundles. At any stage, a compresence bundle involves infinite compresence bundles, where none of the bundles can be described as being a last bundling in the regress. It appears there may not be a point in the regress at all where bundling occurs since this regress appears to be an infinite regress that attempts to complete a task by an infinite sequence of steps, where the «completion» «at infinity» in fact never occurs. If the bundles regress is not completeable, there may be reason to wonder how a regress of compresence bundles is coherent. Each stage of the regress depends on the coherence of a compresence bundle at the next stage, ad infinitum. But if there is no last stage, there is no point in the regress that one can point to where that bundle at that stage is clearly bundled in some way. For this reason, TP must be propertyless.
4.2 The Unmediated Attachment of the Properties of a Bundle and Propertyless the Compresence Tie
The primary bundle, like the TP of the secondary bundle, also involves a compresence tie: the tie that ties the primary bundle's properties (i.e., the first-order properties of the primary bundle or what I am calling the secondary bundles). All bundle theories involve this tie. Loux writes:
The point I will be concerned about in this section is that if properties are bundled by a compresence tie, I will discuss that the properties and the compresence tie involve an unmediated attachment, and like the problems of non-bundle substances discussed in the last section, this unmediated attachment also must be an unmediated attachment of propertyless items, thus rendering bundles impossible since properties and the compresence tie would be indistinguishable.
Loux discusses compresence as a tying relation, but compresence must be a non-relational bundler (a non-relational tie), and not a property (relational, or polyadic, property), that is involved in an unmediated attachment to the properties it connects. If compresence were a polyadic property, the following vicious infinite regress would ensue: bundle B is F, where if the italicized «is» denotes a relational compresence, then F is compresent with B's compresence relation, ad infinitum.Foot note 10_78
Compresence is propertyless, or it has properties. If compresence is propertyless, the unmediated attachment the compresence tie to TP would be an unmediated attachment of propertyless entities, which is the same problem that was discussed in section 3. So that option apparently won't do, and we must consider that compresence itself has properties. If the compresence tie of the primary bundle has properties (such as the property of being a tie, the property of being located where bundle B is, or the property of bundling properties F, G, and H), compresence would itself be a bundle -- a special bundle that is responsible for bundling other propertiesFoot note 10_79 -- since compresence itself is not bundled (it is not an ordinary member of a bundle).Foot note 10_80 The compresence bundle would itself have a compresence tie, which I will call TC, that bundles the properties together. In order to avoid a vicious regress. As with TP, TC must be propertyless, in order to avoid a vicious regress: TC requires compresence tie, TC2, and so on, where no stage of the regress involves any bundling. For these reasons, like TP, TC must be a propertyless item.
For compresence to be involved in an unmediated attachment with the properties (secondary bundles) it bundles, TP and TC would have to be involved in an unmediated attachment, which is an unmediated attachment of propertyless items. However, such an unmediated attachment would render the compresence tie and a property it bundles identical, and thus there would no difference between properties and tie in bundle theory, and the statement «property F is a member of bundle B», would have no meaning.
If my preceding arguments are sound, then it follows that there are no n-adic properties. There are no relations (polyadic properties), such as part-whole relations, topological spatial relations, or temporal relations; and there are no monadic properties, such as the properties of color, spatial extendedness, mental properties, modal properties, and so on. Although the current age of analytic metaphysics surely endorses the stance that a propertyless reality preposterous, this has certainly not always been the case. For example, perhaps my arguments in this paper will bring back interest in theorization about a Parmenidean reality, or about an atomistic reality, such as of the sort of atomistic reality described by the ancient Buddhist or Greek atomists. If my argumentation is correct, it could provide evidence that reality involves only bare particulars and no properties. It appears that the theorization of this article leaves room for such a position since it is only properties and property possession that have been attacked in this paper, and not bare particulars. If there are only bare particulars, reality might be a propertyless reality of only bare particulars, which is to theorize that reality is atomistic (composed of philosophic atomsFoot note 10_81), where the atoms are propertyless and unstructured, and where their activities are apprehended by minds that, in interpreting these apprehensions, create colors and shapes and surfaces out of these activities apprehended. Armstrong apparently considers bare particulars to be atomic and spatially unextended:
If blob theory gives evidence of a propertyless atomic reality, then from what I can tell, this is a position is very much aligned with the ancient Buddhist or Greek atomists. A passage from the philosopher Barry Stroud helps to explain:
The ancient Greek Democritus apparently held a position something like the one being described here, where philosophic atoms are propertyless. But it is hard to understand how his atoms are propertyless and structureless since he maintained that they have a spatial size. It seems that any item with a spatial size has at least some structure, and therefore it has properties. It would be better to theorize that reality consists of spatially unextended, and therefore absolutely structureless, philosophic atoms. Currently, the position that philosophic atoms are point-sized is widely held.Foot note 10_85,Foot note 10_86 Such a reality would consist of numerically distinct, but metaphysically identical philosophic atoms, where at the fundamental level of the philosophic atoms, reality does not involve structure, and since atoms, being propertyless, are indistinct, one might as well say there is only one, even though a mind could count them up, if they could be perceived. If reality is composed of philosophic atoms, the apprehension of, and representation of, the activities of the atoms by the mind, and the consequent mental representations of those activities in the mind,Foot note 10_87 might account for our experience of time, and of structure, in reality, even though there is no time or structure.Foot note 10_88