ONTOLOGY

Ontology is the most general science or study of Being, Existence, or Reality. An informal use of the term signifies what, in general terms, a philosopher considers the world to contain. Thus it is said that Descartes proposed a dualist ontology, or that there were no gods in d'Holbach's ontology. But in its more formal meaning, ontology is the aspect of metaphysics aiming to characterize Reality by identifying all its essential categories and setting forth the relations among them.

Being Qua Being

Existence, as the most comprehensive category of all, should embrace members with the least in common. Nevertheless, Western philosophy long sought some substantive common content present in anything just in virtue of its existence. The history of these attempts to identify the common character of being qua being is not encouraging.

In The Sophist, Plato's Eleatic Stranger proposes that a role in the world's causal network is the necessary and sufficient condition for existence, that "Power is the mark of Being." This idea has had some currency in the twentieth century, particularly in the work of David Lewis (1986) and D. M. Armstrong (1978, 1989, 1997). This Eleatic principle is an attractive test for reality in the natural world, for whatever is real in nature should be able to make a difference. It might be necessary to weaken the requirement and admit a passive space-time that provides the arena within which the active beings exert themselves. Even so, the Eleatic principle seems to be at best a contingent aspect of the world because there seems to be no impossibility involved in the idea of a completely inert being. It also begs the question against abstract entities such as numbers, or geometric points, or sets, which, if they exist, lie outside the causal nexus.

For Samuel Alexander (1920), to be is to be the exclusive occupant of a volume of space-time. This rules out not only abstract entities, but even a field theory of the natural world, for force fields occupy regions of space-time, yet do not exclude one another.

J. M. E. McTaggart (1921–1927) argued that the mark of being is to stand in a determining correspondence with all of one's infinite parts. A determining correspondence ensures that from a sufficient description of anything, a sufficient description of any of its parts can be derived. This requirement implies that space, the natural world, and most of the contents of minds are unreal. From this consequence the conclusion to be drawn is that McTaggart's proposed mark of being is excessively demanding.

The problem of a substantive content for being qua being is reflected in the idiosyncratic behavior of the verb "to exist." Consider singular negatives: "Aristotle does not speak Spanish" is true because the predicate "does not speak Spanish" applies to the item referred to by the subject term. But "Pegasus does not exist" cannot be true because its predicate applies to the item referred to by the subject term. If the subject term refers to anything, that item exists, which would make the whole statement false.

Kant famously declared that existence is not a property, and this view has become widely accepted. The modern logic that descends from Gottlob Frege and the Principia Mathematica (1910–1913) of Alfred North Whitehead and Bertrand Russell replaces all expressions using "exists" with others using "There are." Thus, "Lions exist" becomes "There are lions," while "Dragons do not exist" becomes "There are no dragons."

In technical terms, this process replaces any existence claim with one using a quantifier ranging over a domain (the world), so that to exist becomes a matter not of possessing the special property existence, but of possessing some other, ordinary, properties. The determination to restate all claims to existence or nonexistence with "There are …" and "There are no …" is expressed in W. V. Quine's dictum: "To be is to be the value of a variable."

If existence is not a property, it cannot be a perfection. This undercuts those versions of the ontological argument for the existence of God that rely on existence being among the perfections. A recent response has been to argue that, even if existence is not a property, necessary existence is (Plantinga 1974, 1975; van Inwagen 1993).

Reality and Actuality

Is existence all there is, or should we recognize categories even broader that that of Being? In Plato, and even earlier, is to be found the distinction between Reality (What is) and Appearance (What is not nothing, yet only seems to Be). Aristotle distinguishes the fully existent (Being), from that which is still in formation (Becoming). These distinctions are perhaps best seen as advocating different grades of reality within the one category of Being.

Aristotle also distinguishes the fully Real (Act) from that which may be (Potency). This distinction is the forerunner of a strong strand in ontology that recognizes possible worlds in addition to the actual world, the one we inhabit. In the Neoplatonists, and again in Alexius Meinong, the realm of the existent is augmented by that of the subsistent, which encompasses what does not exist although it might have done so, such as golden mountains.

A full-scale ontology of this kind, in which the realm of Essence is wider than that of Existence, was presented by James K. Feibleman in 1951. In the work of Richard Sylvan (1980), this is extended even further. In Sylvan's system, the individual variables range over not only the actual and the possible, but the impossible as well.

possible worlds

Gottfried Wilhelm Leibniz was the first to make systematic use of the idea that all the possibilities can be regarded as forming worlds—each a complete internally consistent realm that may combine some elements matching the actual world with others in which it differs. The actual world is one of the possible worlds, distinguished from all others by the fact that none of its elements is merely possible. If one is able to refer to possible worlds, it is easy to define necessary beings, otherwise so difficult to characterize, as those present in all possible worlds (see below).

modal realism

Possible worlds make available explanations of causal powers, of counterfactual conditionals, of unexercised dispositions, and of real uninstantiated properties. Such advantages led David Lewis (1986) to embrace modal realism, which affirms the literal reality of all possible worlds.

Other philosophers, while appreciating these advantages, have balked at the apparently infinite expansion of the ontology that this requires. This has led to accounts of ersatz possible worlds: Rudolf Carnap and others have proposed that a possible world is a maximally consistent set of sentences. Armstrong and others have developed Wittgenstein's idea that a possible world is a nonactual recombination of the elements of this world. Peter Lopston (2001) advances a reductive realism, which expands the kind of property assigned in the actual world to include might-have-had features. The success of these approaches is subject to continuing controversy.

many worlds in quantum theory

The notion that the world we live in is not the only one has also been canvassed recently in the interpretation of some otherwise baffling paradoxes in quantum physics. On these accounts the world is not a single unified entity, but one subject to continual bifurcation, a process that generates an ever-increasing number of worlds. Many-world views of this kind are in an important way different from modal realism: all these quantum worlds are supposed to be actual but mutually inaccessible.

The Categories of Being

The principal task of ontology is to furnish an inventory of the categories, the most general divisions of Reality. The most important of these are:

substances

An individual or particular substance is an object, a thing in its own right. Common everyday things, such as bricks and bedsteads, provide a model for the category of substance. Substances are required to have several basic features, although it is not clear that these features are compatible with one another.

Particularity and individuality

A substance is both a particular and an individual; not just some duck or other, but this very duck. An object is of the kind it is (a duck) on account of its properties. But if these properties are universals, shared by many particulars, they cannot themselves confer particularity. Some philosophers, most notoriously Locke, proposed a constituent of substances that would perform this role, a substratum that would confer both particularity and individuality. A substratum would be a bare particular, an item inherently particular and individual, yet without any other feature. It is difficult to see how such bare particulars could be distinguished from one another, but if bare particulars are all exactly alike, how could any one of them individualize its own substance? More generally, bare particulars conflict with Aristotle's dictum that the minimum of being, the least thing there can be, is a "this-such," a particular having a property.

Another proposal is that substances are individuated by their location. Locations—space-time points and regions—are themselves unique particulars; if they can have primitive particularity, that raises the question why other particulars require a substratum or other particularizer. There are other difficulties with location also: Location will not individuate force fields or other physical entities that do not monopolize their space. It fails also for any items of an immaterial kind.

Either individuality—and hence particularity—are primitive, or there are bare particulars, or each substance has a special property, known as haecceity or thisness, which can bestow particularity and individuality on its bearer. For a discussion see chapter fifteen of John Heil's From an Ontological Point of View (2003).

Indivisibility

Individual substances must be distinguished from compounds, so a single substance must be indivisible, in the sense that it has no parts that are themselves substances. This disqualifies ordinary things as individual substances. This simplicity requirement is much emphasized in Aquinas's doctrine of God. It leads in Leibniz to the monadology, and in Roger Joseph Boscovich to the doctrine of material points.

Persistence

Substances are distinguished from their properties by a capacity to persist, that is, to retain their identity through at least some changes. A fire truck can change in color, yet remain the fire truck it always has been. The ordinary compound substances of everyday life have some persistence, but cannot survive all changes. A fire truck dismantled and scrapped is no longer a fire truck. Complete persistence belongs only to the fundamental substances.

Independence

Any substance could be the only thing in existence. If this independence is interpreted causally, no ordinary object is a substance, for they are all brought into being, and hence depend for their existence on their causes. Space-time and its fields might qualify, yet even these depend, in theistic systems, on the creative activity of God. So in Thomism, God is the substance par excellence, but the natural world includes created substances, dependent on God, but otherwise existent in their own right. Spinoza, insisting on absolute independence, concluded that there can be only one substance, the all-embracing totality, God-or-Nature.

If one takes the independence of substances in a logical, rather than a causal sense, a substance is anything that, in principle, could stand alone. This was David Hume's requirement, and anything meeting it is a Humean substance. For compounds, the requirement is that the thing, including all its parts, could exist alone. This requirement is much less rigorous than causal independence and requires no persistence.

No-substance theories

There have been attempts to dispense with substances. Russell has proposed that an ordinary concrete object is no more than a bundle of all its properties. There is always an issue over what it is that binds the bundle. Moreover, as the properties are universals, this theory implies that no two things can be exactly alike.

In Donald Williams's version of the bundle theory (1966), the properties are particular instances or tropes (see below). This avoids any problem with the possibility of there being two exactly resembling objects, but it requires that all members of the bundle be "compresent"—all at the same place in space-time. There are difficulties in treating a space-time location as just one further trope in the bundle, but if it is given special treatment it becomes a substantializing substratum.

Russell also advocated an event ontology as a no-substance view. He used "event" for the occurrence of a property at a place and a time; such events are not happenings, but states of affairs (see below). He proposed that ordinary substances, and their more fundamental parts, are sequences of clusters of such events.

The basic elements in these ontologies may not be simple or indivisible, and they lack persistence. Nevertheless, these states of affairs or events are Humean substances. Indeed, unless there is nothing at all, something must be a Humean substance, and in that sense, any no substance theory must fail.

properties and relations

Properties are the intrinsic features or characteristics of things, which belong to them considered singly. Relations, involving two or more terms, are the ways in which things stand to one another. In many respects, properties and relations can be treated together.

Properties as universals

Properties are usually thought of as universals that can characterize indefinitely many instances. There is but one Eiffel Tower, but the tower's height, weight, and iron constitution are features it has in common with many other things. The Problem of Universals is the problem of explaining how any one real entity could possibly exist, fully and completely, in many different instances. This problem has attracted three different proposed solutions: nominalism, conceptualism, and realism. Nominalism and conceptualism both deny that properties are genuinely universal. According to nominalism, the only element common to all iron things is that they can all be described using the predicate "iron," or all are members of the class iron things, or all resemble some typical iron objects. According to conceptualism, the universal element consists in an impulse of our minds to group several things together. These reductive theories have had adherents since the time of Plato and were particularly prevalent among the British Empiricists and their descendents. Nominalism and conceptualism were explicitly challenged by Russell in Problems of Philosophy (1912). The most thorough case against such views is presented in D. M. Armstrong, Universals and Scientific Realism (1978).

Realism regarding universals is at least as old as Plato. His theory of Forms presents a thoroughgoing realism that accords to genuine properties both a real existence, in a realm of their own, and a status superior to any this-worldly instantiations of them there may be. The Forms exist ante rem —that is, whether or not they are instantiated. The traditional account of Aristotle ascribes to him a modified realism, according to which properties are real, and universal, but can exist only in rebus, as the properties of concrete instances. Here one encounters again his view that the least that is "apt for being" is a this-such, a union of particular with universal.

Realism has always faced two principal objections. First, that it is uneconomical, especially in its Platonist form. The question of economy is a current issue in the philosophy of science, as it at least appears that our best physical and chemical theories involve uninstantiated properties. The second objection is that it can provide no coherent account of the link between a property and the substance that bears it, the inherence relation. Inherence cannot be a normal relation, for then it is just one further universal standing in need of an inherence link to its terms, the substance and the original property. But if it is not a relation in the ordinary sense, what is it? The problem with inherence lends support to versions of realism in which properties are particulars.

Properties as particulars

Even if the property iron is universal, the particular case of being iron that occurs in the Eiffel Tower belongs to the tower alone and is as particular as the tower itself. Trope theory, as developed first by Donald Williams, treats the instance not as a dependent entity arising from the instantiation of a universal, but as a Humean substance in its own right.

When this approach is coupled with a bundle or compresence account of ordinary many-featured substances, the problem of any inherence relation disappears. There is a further significant economy, for there is no need for a separate category of substance. These possibilities are explored further in Keith Campbell's book Abstract Particulars (1990).

Relations

When Russell reanimated the realism debate he accorded to relations a status fully equal to that of inherent properties. Indeed, it was his reflections on the role of relations in the foundations of mathematics and of logic that led him to his realism. Armstrong's realism takes the same form.

There is, nevertheless, a long tradition that accords primacy to the intrinsic properties. Aristotle held that relations are "the least of the things that are"; Hobbes and others held that the existence of relations depends on a mental act of comparison; and Leibniz's view was that every relation has its foundation in an intrinsic feature of one or both of its terms. This reductive program is expounded in Campbell (1990).

Relations do seem to be dependent in the sense that they must have substances as their terms, and these substances must have intrinsic properties. So unless there are intrinsic properties there can be no relations, but not vice versa. Bundle theories of ordinary things concern only the intrinsic properties. To include relations in the bundles leads to problems over where to assign the relations, and this in turn induces a tendency towards a monism such as Francis Herbert Bradley's, in which ordinary substances are absorbed into a single all-embracing totality.

Powers

Some properties, such as square, seem to belong to how an object is. Others, such as being a solvent, seem to refer to what an object can do. This is the distinction between categorical and dispositional properties. One line of thought takes up the Eleatic principle, and identifies real properties as those that confer on their bearer a disposition to act or to be acted upon. Such dispositions are powers ; a metaphysic of powers is set forth in George Molnar's Powers (2003) and in Brian Ellis's Scientific Essentialism (2001).

complexes

Substance and property are basic categories. In combination, they can provide a richer ontology.

States of affairs

A basic state of affairs consists in a particular having a property, or in two (or more) particulars standing in a relation. A single property inhering in a single particular is a minimal "this-such." Wittgenstein's Tractatus Logico-Philosophicus (1921) presented an ontology in which the world is composed of minimal relational states of affairs: those that actually obtain being facts, those merely possible being the remaining states of affairs. These themes—that the basic categories only ever occur in combination, and that these combinations constitute reality—are taken up in D. M. Armstrong's A World of States of Affairs (1997).

Events and processes

A state of affairs is static. To account for the dynamic aspects of the world requires an account of change. This can be done by using sequences of states of affairs: stability consists in successive states of affairs closely resembling one another, whereas change consists in the states of affairs at one time being replaced by others systematically different. An event is a single change, involving a pair of states of affairs; a process is a more complex series of events.

Whitehead, in Process and Reality (1929) accorded priority to the dynamic; all apparently persisting substances are actually slowly evolving processes. The status of space-time is still controversial. It may be a Humean substance; however, some accounts of matter assign it a place as a process, a sequence of complex, changing relations between particulars.

Abstract Objects

Human thought, particularly in mathematics and logic, seems to involve entities that have no apparent place in the natural spatiotemporal world, and no causal role. To admit such items challenges the principle of economy, yet successful reductions are difficult to accomplish.

numbers and sets

Because all numbers can be represented in set theory, there is no need to admit both sets and numbers. Russell had proposed to eliminate sets in favor of propositional functions, but this proved impossible for more than a fragment of mathematics (Goodman and Quine 1947, Quine 1969). Because the variables of set theory have sets as their values, and to be is to be the value of a variable, we are committed to their reality—this is Platonism about sets and numbers. The most important work in attempting to avoid Platonism is Hartry Field's (1980, 1989).

geometrical objects

Unlike anything in the natural world, the objects of geometry—Euclidian cubes, for example—are thought of as perfect, changeless, timeless, and without any physical causal powers. Moreover, there are geometries, and corresponding geometrical objects, with many more dimensions than this world has. A geometrical space can be divided and subdivided into an infinity of different shapes of different sizes. Platonism in geometry thus involves an infinite expansion in ontology.

One approach to this issue is to consider geometrical objects as abstracted objects, that is, objects taken from a context. On this view, every cube is just a particular spatial fragment of space-time and every triangle a fragment of one of space-time's spatial surfaces. One problem with this is that not all shapes will be available. If our space-time is nowhere perfectly Euclidean, there will be no real Euclidean cubes. We can treat these nonexistent objects as imaginary variations on the actually existing ones, and geometries that quantify over such things, as not literally true.

logic

The philosophy of logic makes reference to propositions, operators, functions, and inferences. These are abstract entities, related to reasoning in much the same way as numbers are related to counting and measuring. The problems and prospects of a reductive treatment of them are also parallel.

Necessary Beings

Ordinary things are usually held to exist contingently; that is, they do exist, but might not have. Had our world's initial conditions or laws of nature been different, there would have been a different group of contingent beings. But some things seem to be immune from the vagaries of cause and chance; being outside the causal net, they cannot be brought into being and cannot be destroyed. These are "necessary beings." If Platonism is correct regarding any of the abstract objects, these will be necessary beings, even, paradoxically, the null class.

For Aristotle, anything that exists through an infinite time is necessary because he held that over infinite time every possibility would at some point be actualized. For Plotinus, any divine being would be outside time, and as such could not change, could not cease to exist, and thus would be A necessary being. For Aquinas, God's necessity derives from his simplicity: God's essence and his existence are identical; in this way he is a kind of being that must exist. For Spinoza, every genuine substance is causa sui, containing within itself the sufficient explanation for its own being, and thus it can guarantee its own existence under all possible conditions.

Duns Scotus, then Descartes, linked necessary being with logic: A necessary being is one, the denial of whose existence would be self-contradictory. "Real"—i.e., existing—"beans do not exist" is a self-contradiction, but only trivially because existence has been inserted into the definition of the subject. This does not make beans necessary beings. If existence is not inserted into the subject term's definition, it is doubtful whether any denial of existence would be a self-contradiction. The best discussion of necessary being is in Alvin Plantinga (1974, 1975).

See also Metaphysics.

Bibliography

Adams, R. M. "Primitive Thisness and Primitive Identity." The Journal of Philosophy 76 (1979): 5–26.

Alexander, Samuel. Space, Time, and Deity: The Gifford Lectures at Glasgow. 2 vols. London: Macmillan, 1920.

Armstrong, D. M. A Combinatorial Theory of Possibility. Cambridge, U.K.: Cambridge University Press, 1989.

Armstrong, D. M. Universals and Scientific Realism. 2 vols. Cambridge, U.K.: Cambridge University Press, 1978.

Armstrong, D. M. A World of States of Affairs. Cambridge, U.K.: Cambridge University Press, 1997.

Campbell, Keith. Abstract Particulars. Oxford: Blackwell, 1990.

Ellis, Brian. Scientific Essentialism. Cambridge, U.K.: Cambridge University Press, 2001.

Feibleman, James K. Ontology. Baltimore: Johns Hopkins Press, 1951.

Field, Hartry. Realism, Mathematics and Modality. Oxford: Blackwell, 1989.

Field, Hartry. Science without Numbers: A Defence of Nominalism. Oxford: Blackwell, 1980.

Forbes, Graeme. The Metaphysics of Modality. Oxford: Clarendon Press, 1985.

Goodman, N., and W. V. Quine. "Steps Towards a Constructive Nominalism." Journal of Symbolic Logic 12 (1947): 105–122.

Heil, John. From an Ontological Point of View. Oxford: Clarendon Press, 2003.

Lewis, David K. On the Plurality of Worlds. Oxford: Blackwell, 1986.

Loptson, Peter. Reality: Fundamental Topics in Metaphysics. Toronto: University of Toronto Press, 2001.

McTaggart, J. M. E. The Nature of Existence. 2 vols. Cambridge, U.K.: Cambridge University Press, 1921–1927.

Molnar, George. Powers: A Study in Metaphysics. New York: Oxford University Press, 2003.

Plantinga, Alvin. God, Freedom, and Evil. London: Allen and Unwin, 1975.

Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon Press, 1974.

Quine, W. V. From a Logical Point of View: 9 Logico-Philosophical Essays. 2nd ed. Cambridge, MA: Harvard University Press, 1961.

Quine, W. V. Set Theory and Its Logic. Cambridge, MA: Belknap Press of Harvard University, 1969.

Russell, Bertrand. The Analysis of Matter. London: Kegan Paul, Trench and Trubner, 1927. Event ontology.

Russell, Bertrand. An Inquiry into Meaning and Truth. London: Allen and Unwin, 1940. Substances as bundles of universals.

Russell, Bertrand. The Problems of Philosophy. London: Oxford University Press, 1912. Realism about universals

Sylvan, Richard. Exploring Meinong's Jungle and Beyond. Canberra: Australian National University, 1980.

Van Inwagen, Peter. Metaphysics. Boulder, CO: Westview Press, 1993.

Whitehead, A. N. Process and Reality: An Essay in Cosmology. Cambridge, U.K.: Cambridge University Press, 1929.

Williams, Donald C. Principles of Empirical Realism: Philosophical Essays. Springfield, IL: Thomas, 1966.

Wittgenstein, Ludwig. Tractatus Logico-Philosophicus. London: Routledge and Kegan Paul, 1921.

Keith Campbell (2005)