Metaphysics Toolkit - California State University, Sacramento



Metaphysics Toolkit

Entity

An entity is existing thing, a “being.” This will be our term of art for an existent of any category. Our usual concept of a “thing” will designated by the term object.

Particular vs Universal

An entity is a particular if it is non-repeatable: there can be at most one of it. There cannot be more than one Bill Clinton.[i]

An entity is a universal if it is repeatable. It is capable of having instances, occurrences, examples, tokens, members, cases…, whatever. There can be more than one square, more than one blue triangle, more than one mouse, etc.

Concrete vs Abstract

Concrete

It is difficult to say exactly what makes an entity concrete. Concreteness involves (i) a certain internal integrity, (ii) a certain external independence from other entities, (iii) a certain complexity and fullness of properties, and (iv) being fundamental in a certain way, rather than derivative from, or dependent upon, other entities.

It is even more difficult to say (i) what kind of integrity it must have in order to be concrete, or in what ways an entity must be (ii) independent, (iii) thick with properties, or (iv) underived. Obvious examples of concrete entities are tables and chairs, mice, rocks, etc.[ii]

Intuitively, concrete entities are the basic entities in the world. The other entities are dependent upon the concrete ones, or we can have access to the other entities only by means of the concrete ones.[iii] However, this is not to say that all starting points are equally good; that would beg the question toward anti-realism.

“Concrete” should be distinguished from “physical”. Particular souls and particular angels would be concrete entities, but they wouldn’t be physical. However, the concept of being concrete also seems to involve the concept of being able to interact causally. So non-physical concrete particulars can (somehow) produce effects.

Abstract

Etymologically, something is “abstract” if it is the product of abstraction. Abstraction is the mental process by which we consider an entity apart from its existence in a complex context, compresent with other entities. The mind abstracts from the complex context and considers just that one entity in the complex, for example the squareness of a box. Generally we abstract properties and relations, and kinds. So, we look at this particular square box in this particular situation, but we consider only its squareness, or only its blueness, boxness, etc. The entities we consider, the squareness or blueness or boxness, we consider by virtue of our capacity for extracting that particular entity from the complex in which we find it.

While this is the root meaning of abstraction, it tempts us into the confusion of thinking that, since abstraction is mental process, the object of the process must also be mental – something that exists only if we think it. Maybe abstract entities are just mental entities; but maybe they’re not. Maybe they exist independently of anyone’s thinking of them or actually performing the abstractive mental operation. Other examples of abstract entities are: “being four inches long,” “being red,” “being spherical,” “having whiskers,” “hitting at least 40 home runs in each of two seasons,” “attracting bodies with mass according to the inverse square law,” “being north of,” “being a Chihuahua,” “belonging to the genus Homo,” Maybe abstraction is just the way we come to have a cognitive connection with these entities, which can exist just fine without us. We shouldn’t beg the question either way. Abstract objects are generally thought of as “causally inert”: they cannot produce effects, nor are they affected themselves by anything.

It may not be the case that all particulars are concrete and all abstract entities are universal.

On the other hand, maybe all four possibilities are exemplified:

Concrete particulars: a particular table; a particular man; a particular Chihuahua

Abstract particulars: the squareness of that particular table; Barry Bonds’ hitting 40 home runs in two successive seasons; singular propositions; singular states of affairs, numbers

Concrete Universals: Perhaps an entity could be scattered in space or time in such a way that the entity could be conceived of as entirely present in each element of the scatter, yet the whole be fundamental and independent.[iv]

Abstract Universals: the genus Homo; the set of all human beings;

PARTICULARS

An entity is a particular if it has a kind of unity and is non-repeatable. The unity involved can be existence at only one place, or only at one time, or only at one place at a time. It may also involve existence in a particular order of entities (An example of existence in an order of entities would be numbers themselves), or of a particular degree (the particular intensity of that light).

Categories of Particulars

Objects

An object is the possessor of properties and relations. “That of which everything else is said.”[v] A person is a kind of object that has certain mental as well as physical properties.

Parts

Many concrete particulars can be said to have parts which are themselves particulars. Thus, a Chihuahua has a tail; a car has a steering wheel; a table has a top; a year has a month; a square mile has a square foot; a baseball game has a second inning; a second inning has a top and a bottom.

The minimum requirements for being a (proper) part are:

1. If a is part of b, then b is not part of a. (Asymmetry)

2. If a is part of b, and b is part of c, then a is part of c. (Transitivity)

3. If a is part of b, then there is a part of b which has no common part with a. (Supplementivity)

Parthood is a “concrete” relation, in the sense that a part is of the same degree of concreteness as the whole entity itself. Thus the parts of concrete particulars are concrete; the parts of abstract particulars are abstract.

Disputed question: Whether the parts of an entity are necessary to it; that is,

Necessarily, if a is part of b, then if b( results from the substitution of a by (the nonidentical) a(, then b is not identical to b(.

(Did you get your old car back after the brake job?)

The principle that the parts of an entity are necessary to it, and so nothing can survive a change in parts, is called “Mereological Essentialism.”

Wholes

At least some aggregates of particulars are themselves particulars. Some aggregates of ten years are decades (although some aren’t); some aggregates of space are spaces (though some aren’t); some aggregates of events are events (the event known as the Battle of Midway is part of the event known as World War II). A widely-accepted principle of wholes is that if a and b have all and only the same parts, then a = b.[vi]

Disputed Question: Whether for any two entities (of the same category of entity) there is some whole which they comprise.

The claim that there is is known as the “Sum-” or “Fusion-“ Principle. For some entities the Fusion Principle seems obviously true: there is some sum of water composed of the last raindrop to fall on Sacramento in 1998 and the first raindrop to fall on Toledo in 1999.[vii] It may be a scattered sum, but that’s not always bar to particular-hood or “whole”-hood. After all, while most of the physical expanse of the United States is an integral whole, some of it is stuck onto the northwest corner of Canada, and some more of it is scattered in the middle of the Pacific Ocean. However, is there a physical object consisting of your left ear and the Eiffel Tower? Is there a time consisting of the years 1066 and 1492?

Property-Instances

A property-instance is the presence of a property in an object. Thus the whiteness of this piece of paper is an instance of the universal property of “being (this shade of) white.”

Disputed Question: Whether property-instances are in fact genuine particulars. Or are all instances of a property identical?

A universal property just is the kind of entity that can exist in many different times and many different places. Is there some coherence and clarity to be gained by accepting “property instances” as abstract particulars not to be identified with the abstract universal of which they are instances?

Relation-instances

A relation-instance is like a property instance. Similar considerations apply here as to properties. So the particular fathering relation between George Bush and George W. Bush is an instance of the relation of ‘fathering’, or ‘being the father of’.

Places

A particular place can be identified by what objects are present, or what properties or relations are instantiated at them. (But then, if you identify objects by their location, or distinguish two objects by their different locations, you are involved in a notorious circularity.) Thus one may wish to claim that an individual place can be identifed by the existence of an order of places. If space, or space-time, has an intrinsic ordering, then places will be identified by their place in that ordering independent of what objects or properties, if any, are present at them.

Times

A particular time can be identified by what events occur, at that time. (But then if you identify events by the time of their happening, or distinguish two events by their different times, you are involved in another notorious circularity.) Or, again, an individual time can be identified by the existence of an order of times, and the same considerations that applied to space or space-time apply here. There is an obvious intrinsic order to times: McTaggart’s B-series.

Disputed question: Whether there is another, independent order of times – McTaggart’s A-series.

States of Affairs

A particular state of affairs is George W. Bush’s being President. Another particular state of affairs is Al Gore’s being President. These two states of affairs are alike in that they both exist. The difference between them is that one obtains, and the other does not. Some states of affairs obtain necessarily, others obtain contingently. Still others necessarily do not obtain. That is, states of affairs have “modal” features (see below). States of affairs also have constituents, some of which are concrete entities. George W. Bush’s being President has George W. Bush (who is a concrete object) as a constituent, as well as the property of being President (which is an abstract property). But states of affairs are themselves abstract particulars.

Facts

The abstractness of states of affairs raises a problem. It seems that George W. Bush’s being President can be both an effect and a cause. (George W. Bush’s being President causes NY Times columnist Paul Krugman to be depressed.) So we need a concrete entity related to this state of affairs that can be involved in causal relations. Such an entity is the fact that George W. Bush is President. So the question arises: how is the state of affairs related to the fact? It seems that for the fact to exist is for the state of affairs to obtain. On one metaphysical view facts, not objects, are the furniture of reality: “The World is the totality of facts, not of things.”[viii]

Events

An event is a kind of fact. They are the terms of causal relations: In A causes B, A and B are events. A particular event is a particular object’s coming to have a property.

Disputed Question: Are all events, all instances of coming-to-have-a-property, changes? Coming to have the flu is a change in you. Is coming to have a Bachelor’s degree a change (though not a physical change) in you? Getting a new nickname is an event, but do you change when you get one? Some philosophers follow P. T. Geach in distinguishing real change from what he calls “Cambridge Change”. A Cambridge Change results when some sentence changes truth value, and thus a predicate true of an object at one time becomes false of that object at another time. Do you change when it is no longer true of you that you are the fourth person in the checkout line? Does a CD player change during a Sale, when it ceases to be $400 and becomes $300? Does the number Five change when it ceases to be the number of my brother Bill’s children? Or does change require coming to have an intrinsic property? Which properties are intrinsic?

Propositions

A proposition is what corresponds to a sentence. Indeed, propositions are taken by some metaphysicians to be the meanings of sentences. So the meaning of the sentence “John loves Mary” or its Latin equivalent “Iohannes Mariam amat” is the proposition that John loves Mary. Propositions are what are true or false. Propositions seem obviously related to states of affairs: the proposition

that George W. Bush is President

seems connected to the state of affairs

George W. Bush’s being President.

But what precisely is the connection? Some differences: propositions differ from states of affairs in being more finely distinguished and individuated. Thus, the sentence “Cassius Clay won the 1960 Olympic Gold Medal in boxing” and the sentence “Muhammad Ali won the 1960 Olympic Gold Medal in boxing.” Both sentences are true, and they are true because they report or denote the same obtaining state of affairs, namely that one particular person won that one particular medal. However, it is useful to be able to say that they express different propositions. It is especially useful to have propositions as well as states of affairs as part of an account of our beliefs and other intentional states. If we had only states of affairs in our metaphysical toolkit we could not explain what it is to believe that Cassius Clay won the medal and not believe that Muhammad Ali did (because we don’t know that they are the same person). What makes the propositions true, it would seem, is the obtaining of that state of affairs. Or, to put it another way, the fact that Cassius Clay/Muhammad Ali won the 1960 Olympic Gold Medal in boxing is the ‘truthmaker’ for the proposition.

Another apparent difference between states of affairs, facts, and propositions has to do with the capacity for propositions to fit into logical structures. For instance there is a conjunctive proposition “Theseus loves Hippolyta and Lysander loves Hermia.” But it is not required that there be any corresponding conjunctive state of affairs or fact. There is also the negative proposition “Demetrius does not love Helena,” but there may be no corresponding negative fact. (Is it a fact about a can of beer that it is not in your refrigerator?)

Some other differences between states of affairs and facts, on one hand, and propositions on the other include the fact that propositions come in pairs and facts/states of affairs don’t. There can be only countably many propositions, but there may be uncountably many states of affairs. Propositions are logical entities but states of affairs aren’t.

Sentences and Utterances

Sentences are abstract linguistic structures formed according to grammatical rules. The same abstract structure can obviously be instantiated in writing, in speech, etc. Take the sentence

All pigs are pink.

As a piece of linguistic structure it can be used to express a proposition, but it doesn’t do so just by itself. In this time and place, where “pig” and “pink” mean what they mean, it expresses the proposition that all pigs are pink.[ix]

An “utterance” is a concrete instance of a sentence produced at a particular time, place, and context. Again, utterances can be written, spoken, wigwagged, smoke-signalled, etc.

Utterances, and the sentences they embody bear the ”Type/Token” relation. A sentence “token” is the particular ink stain on your copy of this Toolkit, or the (particular sound-event if spoken, etc.). The sentence “type” is the abstract structure itself.

Universals

As indicated before, a universal is repeatable. It can have instances, occurrences, members, examples, or what-have-you.

Properties and Relations See “Property-instances” above.

We shall speak this way: ‘Instances’ of properties and relations are as we described them above. An example of an ‘instance’ of whiteness is the whiteness of the particular piece of paper you are reading now. However, we shall say that properties and relations also have ‘occurrences’: Occurrences are the objects which have the properties. So an example of an occurrence of whiteness is this particular piece of paper itself. An occurrence of the father/son relation is the pair, George Bush and George W. Bush

Kinds the breed Chihuahua; the Genus Homo

Although being an instance of a given kind may require the possession of a given set of properties (the essence), a kind is not itself a property. A kind has instances but no occurrences (to think of it another way, with kinds instances are identical to occurrences). Instances of the kind Chihuahua are Chihuahuas, not properties of Chihuahuas. Kinds are also called “Sortal universals” or “Sorts”.

Sets the set of square things; the set of Chihuahuas; the set of members of the Genus Homo

A set is an abstract entity even if the elements of the set are themselves concrete. A set has its members necessarily: If there had been no Chihuahuas, the set of Chihuahuas would either not have existed or would have been identical to the “null set”. If there had been other Chihuahuas than there actually have been (or even one more or one less) the set would be a different entity than it is.

Although all and only the members of the set of Chihuahuas are instances of the kind Chihuahua, the set and the kind are different entities. The kind would have been the same kind even if different Chihuahuas had existed (or perhaps had no Chihuahuas existed at all). The members of the kind are not essential to it, as they are to the set.

Disputed Question: For every property or relation there corresponds a set: the set of individuals having that property or being in that relation. Can we do without properties, countenancing only sets instead? This has some conveniences. The metaphysics of sets is pretty well understood, the metaphysics of properties less so.[x] Reducing properties to sets would have some inconveniences too: “being a wife of Pope John Paul II” becomes the same property as “being a unicorn.”

Types Mozart’s 21st Piano Concerto, Metaphysics: The Big Questions, the 1964 Mustang

The Type/Token relation is a kind of Kind/Instance relation. What seems to distinguish types from other kinds is that some attributions which are applicable to the tokens are also applicable to their types as well. A concerto is instantiated in particular performances. We do not say of the kind Chihuahua that it is a dog or an animal. However, we do say of Mozart’s Concerto that it is a piece of music just as we say of any token of it (any performance of it) that it is a piece of music. The 1964 Mustang, the type, is a car; any token of it is a car.[xi]

MODALITIES

The modalities we will be concerned with are Necessity, Possibility (which includes Probability), Contingency, and Impossibility. These concepts are all interdefinable: Where E is any entity,

E is possible =df Not-E is not necessary

E is impossible =df Not-E is necessary

E is contingent =df E is not necessary and not-E is possible

So we need discuss just the concept of Necessity.

Necessity (and all the other modalities) comes in different grades and in different sorts. We shall take necessity to be a property; what it is a property of depends on what grade and what sort of necessity it is.

Grades of Necessity

(In ascending order of disputedness)

1. The first, most minimal and least controversial, grade of necessity is a property of sentences. Since sentences express propositions, and propositions are the bearers of truth and falsity, this necessity is related, to but different from, necessary truth or falsity, which is the third grade of necessity. Consider the sentence:

“All pink pigs are pink.”

The sentence is a grammatical structure, unless it expresses a proposition it is neither true nor false. But this sentence has this important property: by virtue of its grammatical structure it cannot express a proposition that is false. Call this sort Necessity1, or “Tautology-hood”.

2. The second grade of necessity is a property of concepts: Conceptual Necessity. Conceptual necessity is the sort of necessity at work in sentences like “Bachelors are unmarried,” and “The rook moves any number of spaces along the rank or file.” A chess piece has its powers necessarily, but the necessity is confined to the fact that, say, “moving any number of spaces along the rank or file” is a component of the concept of a “rook”. And the concept itself is not necessary but contingent: it’s obviously possible that the rook might have had different powers. We would simply then have had a different concept of the piece than the one we do.

“Budget bills originate in Congress” is also a conceptual necessity. But had the Constitution been different it would have been false.

Disputed question: To what extent to our concepts – which are products of our natural and cultural history after all – ‘track’ realities which are not human artifacts: realities of mathematics, science, and metaphysics itself. If they track them fairly closely, then the philosophical enterprise of “conceptual analysis” could yield genuine knowledge of the world.

Call this sort Necessity2. It’s also known as “Analyticity”.

3. The third grade of necessity is a property of propositions themselves. A necessary proposition is one that cannot be false. What does this mean?

The proposition that George W. Bush is President is true; it is not false.

But surely it could be false. You and I could be mistaken – unlikely, but not impossible.

Or it could have been false. Al Gore might have won the 2000 election, in which case it would have been false.

So this proposition is true; but it is not necessarily true.[xii]

The proposition that all pink pigs are pink is true; but this proposition also cannot be false. We can neither be mistaken about it, nor could pigs that are pink not be pink.[xiii]

Call this sort Necessity3, or Logical Necessity.

What’s the difference between being necessarily3 true and just being true? The distinction can be clarified by appeal to Leibniz’s notion of a “possible world.”

Possible Worlds

The idea is that the actual world is just one of infinitely many possibilities that might have been actual instead.[xiv] In some of those possibilities, for instance, Gore won the election; in some of them Cassius Clay did not win the 1960 Olympic Gold Medal in boxing. So, to say that a proposition P is true is to say:

P is true in the actual world, but it could have been false.

And this is to say that in other possible worlds P is false. In some worlds, for instance, the proposition that George W. Bush is President is false. In some of those worlds, the proposition that Al Gore is President is true. In others the proposition that Osama bin Laden is President is true. In others the United States has a king, a Gauleiter, a commissar, or it doesn’t exist at all: the Chihuahuas rule the earth.

So, to say that a proposition is necessarily3 true is to say that it is true in all possible worlds. All logically necessary truths are true in all possible worlds.

4. The fourth sort of necessity is a property of properties, relations, and kinds. A necessary property is one which an entity has in all possible worlds.[xv] Thus triangles have the property of possessing internal angles equal to 180( in all possible worlds; the number Four has the property of being even in all possible worlds; red is a color in all possible worlds.[xvi] Call such a property “necessary4,” or “necessary de re,”[xvii] or “essential.”[xviii]

What seems to be required to understand necessity4 is the conception of an entity’s existing in other possible worlds besides the actual world. This concept of “Trans-world Identification” seems to abound with problems. Suppose Bush and Gore existing in those other worlds where the outcome of the 2000 election is reversed. What other properties does Bush lack in those other worlds? Could he not be married to Laura? (Remember that to say no is to say that his being married Laura is necessary4, and thus that the proposition that George W. married Laura is necessary3 – which clearly it is not.) Can George W. not be a Republican? Not be from Texas? Could he be a 17th century French peasant in another possible world? Could he not be human? Could he be an alligator in another possible world, or is being human necessary4 (essential) to him? Could he have been the former dictator of Iraq? Could he have been named “Saddam Hussein”? Could Saddam, in turn, have been born in Texas, etc. How would we then tell who is George and and who is Saddam in that world? Does that even make any sense? Have we fallen into metaphysical quicksand?[xix]

5. The last and most controversial grade of necessity is a property of particular objects, abstract or concrete. Most particular objects seem to exist contingently; at least, for objects which actually do exist, we can easily imagine them not existing.[xx] Are we then capable of conceiving – really conceiving – what it would be like if some objects which do not actually exist did? How could we do that?

Possibilia

If I try to conceive of a person who, if they had existed, might have been President instead of George W., what do I succeed in doing? Is it a he or a she? Fat or thin? Republican or Democrat? To be able to have this conception, and have it be about that person, the person must somehow be an object. But he is not an actual object: if he were, he would be in the actual world.

Four alternative theories:

( i ) The actually existing objects are the only possible ones, so we can’t have that thought.[xxi] This has the unattractive consequences of making all the actually existing objects necessary beings, which conflicts with the intuition of contingency and constitutes a very powerful metaphysical thesis in its own right. What arguments could we have for such a strong claim?

( ii ) Although other objects might have existed that don’t, we cannot say so or think so.

The idea that we can conceive of an object that might have existed but doesn’t is a conceptual illusion, similar to perceptual illusions like the drawings M. C. Escher.

A weaker version of this thesis is that it is only of actually existing objects that we can assert existence. We can neither assert nor deny propositions about nonexistent objects, because if the objects don’t exist then neither do propositions about them. While it may be true that other objects might have existed, these are not truths we can state – or think.

This conflicts with an intuition that any proposition is expressible or graspable. After all, we can identify which proposition the sentence “There is a Santa Claus” would express if there was a Santa. Isn’t it kind of odd to say that we can identify the proposition, distinguish it from all the others, yet not grasp it?[xxii]

( iii ) Different objects exist in different possible worlds. So objects that are merely possible, from the point of view of the actual world, exist in other possible worlds. If those worlds had been the actual world, they would be actual.

The trouble with this alternative is understanding how we talk think about those individuals or refer to them. “Reference”, or “intentionality” – whatever relation it is by which our thoughts and utterances about particular objects succeed in being about those objects – requires some sort of actual connection with the objects. That connection may be causal, physical, perceptual, historical, whatever.[xxiii] Theories on the matter differ. But the theories agree that we have to have some kind of tie. But none of the ties proposed can operate from another possible world to this one. Causality can’t; it operates just within worlds, not across them. There’s no physical tie between possible worlds. And we have no “Julesverne-o-scope” (as Keith Donnellan has called it) allowing us to perceive other possible worlds. Nor are there any historical connections between possible worlds.

( iv ) All possible objects exist in all possible worlds. However, only some of those objects are “actual.” This thesis may be combined with the another thesis: that objects which are not actual in our world are “actual” in other possible worlds.[xxiv] However, nonactual existence in the actual world is existence enough to enable us to refer and have thoughts about possible objects.

The trouble with this alternative is to make sense of this distinction between existence and actuality. On the theories of reference mentioned before some actual tie is needed, so it’s not clear what constitutes “enough” non-actual existence to enable us still to refer to possibilia.[xxv] How many possible Presidents fit on the head of a pin?

Is there a Necessary5 Being? Anselm’s Ontological argument concluded that there was. The tradition since Kant’s criticism of Anselm’s Argument held that the notion of necessity does not apply to objects: there is no such property as “necessity5”. However, some recent metaphysicians are prepared to countenance such a property. Indeed, according to them there are lots of necessary beings: if numbers are objects they would be necessary, furthermore any sets of necessary beings would also be necessary beings. If Mereological Essentialism is true, then parts of necessary beings would also be necessary beings. If the Fusion Principle is true, wholes with necessary beings as parts are necessary beings. (Mathematics as a branch of theology!)

Sorts of Necessity

1. Logical necessity. This sort of necessity is generally expressed by sentences which are necessary1 (tautologous) or necessary 2 (analytic). Such sentences are true in virtue of their logical form’. This is evident for tautologies, but analytic sentences show this necessity if you substitute (say) an analysis of the concept of a rook for the term ‘rook’.[xxvi] However, since propositions also have logical form, propositions can also have this sort of necessity. Thus Logical Necessity is a form of necessity de dicto.

Disputed question: Is the necessity involved in mathematical truths logical necessity or not? That is, is mathematical truth just a form of analyticity or tautology-hood, or is it something more?

2. Metaphysical Necessity (sometimes unhelpfully called “Broadly Logical Necessity”) Metaphysical necessity is what is common to all possible worlds, whether it is the truth of certain propositions are the compresence of certain properties and kinds. There appear to be necessary truths whose necessity is not explicable in terms of logical form. Some examples are: (i) “Everything green has some spatial property,” “Dogs are mammals,” “Lightning is static electricity.” It is surely not part of the concept of ‘dog’ that dogs are mammals, since the concept of dog was around for a long time before we knew what they were. Likewise for theoretical reductions like “Lightning is static electricity.” Thus these may not be detectable by means of formal logic, but only by metaphysical argument. This sort of necessity seems to span the de dicto/de re distinction.[xxvii] That is, there seem to be de dicto necessities that are not merely logical, but metaphysical.[xxviii]

3. Physical or Natural Necessity. What is common to all possible worlds sharing the same physical laws as the actual world. Physical necessity is a narrower notion than logical necessity. It may not be physically possible for to swim the Atlantic, but its metaphysically possible. There’s a possible world with physical laws different from our own and in which George W. Bush exists, where human beings do have the physical capacity to swim the Atlantic.

4. Deontic Necessity. “Obligatoriness” What is common to all possible worlds sharing the same moral realities (Facts? True propositions? Good- or right-making properties?) as the actual world. Again, narrower than physical necessity.

5. Epistemic Necessity. “Knowledge” What is ideally undefeated and justified true belief in all possible worlds sharing the same epistemic principles as the actual world.

6. Miscellaneous Other Kinds. “When two motor vehicles arrive at a four-way stop simultaneously, the motor vehicle on the left must yield right of way.” “The soup spoon must be placed outside the salad fork.” “A player ejected from the game must leave the dugout.” These do not ordinarily give rise to metaphysical issues.

Descriptive vs. Revisionary Metaphysics

Descriptive metaphysics is the task of developing a coherent account of our prevailing conceptual scheme. Revisionary metaphysics is any attempt to displace some category of entity in that conceptual scheme or to introduce some new category.

Wilfrid Sellars has called the prevailing conceptual scheme the “Manifest Image”: that picture of the world and of our place in it that guides or ordinary lives and grounds our common sense.

Manifest Image categories are the most basic ones we have. There are objects. Those objects have the properties (which they have even when we’re not looking). They undergo changes in properties while remaining the same. Properties like color, weight, smell are properties of the objects themselves.

Space is three dimensional, a neutral container of objects. Time passes; that is, past, present, and future are metaphysically signficant distinctions. Some objects are persons; they have certain mental properties; that is, they have minds. Mental properties are different in important ways from ordinary physical ones.

There are also events. Some events are not just happenings, but are actions.

Revisionary metaphysics is motivated by dissatisfaction with the Manifest Image, either because some category seems incoherent (change, for instance), or because categories within the Manifest appear to conflict with one another (identity and change), or because it is imcompatible with a scientific account of the world. In such cases there must be some “commensuration” between the questionable categories and favored ones.

Strategies of Commensuration:

Semantical Reduction:

A semantical reduction of Realm B to Realm A takes the form of showing that all the terms or concepts by which we designate categories of entities in B really refer to categories of entities in A. Roughly, “When we talk about B’s we’re really just talking about A’s.”

An example of semantical reduction is the ethical non-cognitivist’s proposal to reduce ethical assertions to assertions about desires and preferences. This theory proposes a systematic meaning and reference relation between sentences employing moral vocabulary with sentences about individual or community preferences. When I utter a moral sentence like “Stealing is wrong,” I am not attributing a “moral property” – wrongness – to an action-kind, stealing. Rather I am saying, “I disapprove of stealing.” Or perhaps, “My community disapproves of stealing.” The more radical version of this semantical reduction is the ethical emotivist’s view that “Stealing is wrong” does not state my disapproval of stealing, but rather expresses it. On the latter view, it is like saying (in a minatory, disapproving tone of voice), “Oooooooh! Stealing!”

Semantical reduction does not by itself have metaphysical consequences. There may still very well be real moral properties, and it may be in virtue of those properties that actions are right and wrong. It’s just that we never refer to those properties or appeal to them in our moral thinking.

Theoretical Reduction:

A theoretical reduction of realm B to realm A is found most obviously in science, but it is not limited to that. It occurs when the laws of a theory of realm B can be shown to be consequences of laws of a theory of realm A which is somehow more “basic.” Thus certain biological laws (auto exhaust is toxic) may be entirely explained as consequences of chemical laws (hemoglobin forms a bond with a chemical in the exhaust, CO, more readily than it forms a bond with O2, atmospheric oxygen). The chemical laws may in turn be explained as consequences of laws in physics (relative chemical attraction has to do with the atomic structure of the atoms composing the chemicals). The most striking example of theoretical reduction in the history of science is the unification of the laws of celestial motion (Kepler’s laws of planetary motion) with the laws of terrestrial motion (Galileo’s laws for free-fall and projectile motion) by Newtonian physics. The laws of realm B are still “true”; but their truth-makers are the corresponding laws in realm A.

A successful theoretical reduction does lend support to the claim that the laws of theory B are special cases of laws in theory A. This in turn lends support to the claim that the facts of realm B are, at base, facts of realm A. That is, the fact that we can operate quite successfully on the methodological assumption that everything about biology can be explained in chemical theory lends credence to the claim that biological properties are really identical to chemical properties, biological facts nothing but chemical facts. However, moving from a methodological theoretical reduction to an ontological reduction is an added step. If enough theoretical reduction holds together that step may become one that it is increasingly irrational not to take.

Ontological Reduction:

An ontological reduction of realm B to realm A involves showing that (and how) entities of realm B are nothing but (logical constructions out of) entities of realm A. This is usually the goal of a successful theoretical reduction. An ontological reduction is consistent with continuing to maintain that sentences that appear to referr to realm B properties being meaningful and true. Realm B sentences may be useful abbreviations of Realm A sentences, and realm B laws pragmatically useful. (Sailors still take the laws of ptolemaic astronomy to be true when they use celestial navigation.)

Supervenience:

The difficulty with reductions of all kinds is that they have to be complete in order to be true. No residue of realm B must be left unreduced or unaccounted-for. And no other explanations may intrude. Thus, if some biological properties are partially deducible to chemical properties plus, say, the behavior of the organism involved, the reduction is not complete. Therefore the reduction fails unless we have a complete chemical reduction of the behavior too.

Furthermore, an ontological reduction must be metaphysically necessary. If heat is nothing but molecular motion, then heat is necessarily4 or essentially molecular motion. This is a very strong metaphysical thesis, and thus requires a lot in the way of theoretical justification.

Since reduction is such a strong result, a strategy short of ontological reduction becomes attractive. It becomes tempting to say that realm B supervenes upon realm A. This amounts to something like the following:

For a property in realm B, say F, to be instantiated, it is necessary that some set of properties in realm A, say G and H , be instantiated. That is, something has property F if and only if something (not necessarily the same thing) has properties G and H.

But there is no suggestion that having F reduces to, or is merely a matter of having, properties G and H.

Supervenience is the presence of a determining relation between the properties in the two realms. Having realm B properties is determined by having realm A properties.

There can be no difference in objects with respect to their realm B properties without a difference in their realm A properties.

The notion of supervenience actually began in ethics, with the thesis that moral properties somehow supervene on natural properties. An action’s “being wrong” supervenes on its, say, causing unnecessary pain to the innocent, or on its having worse consequences than alternative actions possible in the situation. The predicate “being wrong” doesn’t mean “causing unnecessary pain.” Nor is an action’s being wrong to be explained as a consequence of some complete theoretical account of how such actions cause unnecessary pain,[xxix] nor is being wrong nothing over and above causing unnecessary pain. Rather, an action’s causing needless pain determines that that action is wrong. An action’s being wrong is determined by its causing needless pain. Nothing could differ in B realm properties, its moral properties without also differing in A realm properties (its unnecessary-pain-related properties) too.[xxx]

Given the manifest difficulty involved in theoretical reduction of mental events and states to physical events and states, it is tempting for a physicalist to claim supervenience instead. No mental event can occur without the determining physical event’s also occurring. No two people can differ in their possession of mental states without there also being a difference in the physical states they also possess. Since a supervenience thesis asserts a relation between properties of both realms, it presupposes the existence of properties in both realms. Many philosophers put the best face on this by considering supervenience theories a form of physicalism. The designation seems more stipulation than discovery.

Elimination:

Consequently physicalists are tempted by another tactic: arguing that in some cases, reductive efforts lack any point. That is, it is best to treat the properties of realm B not as candidates for ontological reduction but as fictions. If the phenomena associated with demonic possession are successfully explained as psychotic or hysterical symptoms we are not disposed to claim some theoretical identity like “demons = psychotic symptoms.” Rather we are inclined to claim that there are no demons.[xxxi] On such a view the properties of realm B are discarded and replaced with realm A properties. Eliminativism, like all attempts to dismiss manifest image concepts as fictions, owes us an explanation of our tendencies to make this error. They must, say, dismiss our ordinary mental concepts as “folk psychology.” They must then either show how our forms of life which presuppose those concepts can be conducted after the substitution, or show that they are fundamentally misguided somehow.

-----------------------

[i] Obviously there are lots of Bill Clintons. But there cannot be more than one Bill Clinton identical to Bill Clinton. Those other Bill Clintons are not identical to Bill Clinton; they’re identical to themselves.

[ii] But is a cloud a concrete entity? Does it have the requisite integrity and independence of other things, including other clouds? Or is a cloud simply an aggregate of more fundamental concrete entities – water droplets or water molecules?

[iii] Perhaps it’s best to think of “concrete” and “abstract” not as opposites but as ends of a continuum, with degrees of concreteness or abstractness in between. Or better, what you consider concrete and what abstract is a matter of where you begin — of what entities you take as basic. For instance, you can get pretty far in interpreting Plato with the idea that, in his ontology, the concrete basic particulars are the Forms. Material objects would be dependent on them as abstract particulars are dependent on concrete ones – by reflection or abstraction, as it were, from the true reality.

[iv] This was the view of British Hegelians F. H. Bradley and Bernard Bosanquet. An individual person is a concrete universal, identical in all his actions. Likewise a nation like the United States is identical despite being composed of completely different indviduals in 1800 and in 2000.

[v] As Aristotle says in the “Categories” (2a11).

[vi] This will be disputed by Substantialists. They think that non-identical objects can consist of the same parts. The timbers out of which the Pilgrim ship Mayflower was built now compose a barn in Kent. Call this barn Shmayflower. Is Mayflower = Shmayflower? A substantialist will say no. One has properties which the other lacks: Mayflower is essentially a ship, and Shmayflower is not; Shmayflower came into existence after 1620, Mayflower before; etc., etc.) In Sacramento, 16th Street and State Route 160 consist of the same length of asphalt. Yet they are different roads: SR 160 has existed only for about 70 years, where 16th St. has existed for over a century.

[vii] It’s not a raindrop, but is it a whole composed of water? Is it, as it were, a “piece” of water?

[viii] Lugwig Wittgenstein. Tractatus Logico-Philosophicus, 1.1.

[ix]Othello orders Desdemona, “Swear thou art honest.” (Othello, Act IV, Scene II) An illogical demand if ever there was one! She will swear to it either way – honestly if she is, dishonestly if she isn’t. But Othello was not demanding what the sentence seems to us to demand. In 17th Century England, another time and place, “honest” meant chaste. The sentence just expresses a different proposition now.

[x] Is the property of having a heart the same or different from the property of having kidneys? The sets are the same. We could perhaps distinguish the two properties by saying that the having of a heart might have diverged from the having a kidney. It’s just a fact that they didn’t: just as if all and only snowboarders were Republicans. If that were true it wouldn’t make “being a snowboarder” the same property as “being a Republican.” But that can’t be the reason: “having four sides” and “having four angles” could never diverge in that way. Yet we are not prepared to say that they’re the same property.

[xi] The type and the token are not univocally pieces of music or cars. The expressions are analogous somehow. But even the correct analogous use of the same term mark a difference between types and other universals. This may have something to do with the fact that types are usually not natural entities but artifacts intended to be repeatable.

[xii] It is contingently true: The proposition is not necessary, and its “negation” (that George W. Bush is not President) is possible.

[xiii] The last clause is ambiguous: It may very well be possible that some pigs which actually are pink might not have been pink. What is impossible is that any pigs could be actually pink and not pink. (Using symbolic logic, the distinction is between (x[(Pig(x) & (Pink(x)) & ((Pink(x))], which might very well be true, and ((x[(Pig(x) & (Pink(x) & ( Pink(x))] which is necessarily false.

[xiv] Why is the actual world the actual world? Leibniz, who held the Principle of Sufficient Reason, claimed that this cannot be just chance or a brute fact. This world is the actual world because it is the best of all possible worlds.

[xv] That is, the entity has this property in all possible worlds in which it exists: see Necessity5.

[xvi] Again, all possible worlds in which those entities exist.

[xvii] The first three grades of necessity are “necessities de dicto”: the necessity is characterized in terms of truth. The last two grades are “necessities de re”: the necessity has to do with the entities themselves, not with the truth or falsity of what is said about them.

[xviii] “Essential” is a term employed in Aristotelian metaphysics. Essential properties are those having to do with the essence of an entity. So, one disputed question has to do with the extent to which countenancing necessity4 commits one to an Aristotelian-style ontology. Recall that what we call Modern Philosophy – that movement from Descartes through Locke to Kant – was precisely a rejection of Aristotelian logic and metaphysics in favor of the new sciences. So this can appear to be a reactionary turn to premodern ways of thinking.

[xix] No. See Saul Kripke, “Naming and Necessity,” Lecture I, in Semantics of Natural Language, 2nd Edition, edited by Donald Davidson & Gilbert Harman, D. Reidel (Dordrecht, 1972), 267-274.

[xx] Had Pappy Bush not met Barbara, George W. Bush would never have come into existence.

[xxi] We may think that we’re having that thought, but we’re not. There is no such thought. Some witches may have believed that they really were casting spells. But they weren’t.

[xxii] But it may not be so odd. Our intuition that we can identify the proposition may itself be an illusion.

[xxiii] Our ability to refer to actually existing objects in other possible worlds (to wonder whether Bush would have become President, say, in a world where he didn’t give up drinking) depends upon our ability to refer to those objects in the actual world.

[xxiv] Or perhaps it would be better to say that those objects would have been actual, if that possible world had been the actual world.

[xxv] Bertrand Russell’s dismissal of one such theory, by German philosopher Alexius Meinong, still ranks as one of the best put-downs of one philosopher by another in the 20th century:

“In such theories…there is a failure of that feeling for reality which ought to be preserved even in the most abstract studies. Logic, I should maintain, must no more admit a unicorn than zoology can….” Introduction to Mathematical Philosophy (1919).

You have to love the way Brits can really get the boot in without ever descending to personal abuse or trash-talking.

[xxvi] Or ‘unmarried male’ for ‘bachelor’ in “All bachelors are unmarried.”

[xxvii] “Everygreen thing has a spatial property” would seem to be de dicto, where the mammalian nature of dogs might be a de re construal.

[xxviii] Or we could, perhaps makea distinction between the proposition that every green thing has a spatial property, which is de dicto, and the state of affairs every green thing’s having a spatial property which necessary de re.

[xxix] Along with some complicated theoretical account of how causing unnecessary pain is a wrong-making property of actions generally.

[xxx] Obviously there are other candidates for the wrong-making property than causing unnecessary pain. The “metaphysics of morals” is a neglected topic.

[xxxi] A neighbor boy told me one Christmas, “My dad is Santa Claus!” Little Carl apparently held a theoretical reductionist position on the metaphysics of Santa Claus rather than the usual eliminativist one.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download