General Ontology
9
General Ontology
Existence, Identity, and Reductionism
What “Lions are real” means is that some particular property . . . the property of being a lion . . . does in fact belong to something.
G. E. Moore
Each thing is what it is and not another thing.
Joseph Butler
1—Introduction
In chapter eight, we saw that two main tasks of general ontology are to explain what existence itself is and to describe general features that are true of all entities whatsoever. In this chapter, we will engage in these tasks by first looking at issues in the theory of being or existence. What is it to exist versus not exist? Second, we will investigate one feature that is true of all entities, namely, the fact that everything is identical to itself and different from all other entities. What is identity, and what role does an understanding of it play in philosophy? Finally, we will examine some aspects of an area of metaphysics closely related to existence and identity: replacement and reductionism. What does it mean to reduce one thing to another; for example, to reduce color to wavelengths, heat to the movement of molecules, minds to brains?
2—The Nature of Existence
Suppose we have before us a real, live horse named Fury. Now think about something that is not real, say the winged horse Pegasus. There are many things true of Fury: he is black, a horse, and so on. But the most fundamental thing true of Fury is that Fury exists. If we compare Fury with Pegasus, clearly there is something different between them: Fury exists and Pegasus does not. This difference is as real as the difference in color between a brown and blue billiard ball. How are we to account for the difference between Fury and Pegasus? This and related questions are what a theory of existence tries to answer. However, before we look more deeply at theories of existence, we need to glance at a prior question: Is being a genus?
2.1 Is Being a Genus?
If being is a genus—that is, a univocal notion that applies in the same way to all things that have being—then this means that whatever existence itself turns out to be, everything that exists will have existence or being in the same sense. Being is a univocal notion that means the same thing for all entities whatsoever. On the other hand, if being is not a genus, then what it is for one thing to exist, say the number two, may be entirely different from what it is for a carbon atom to exist. It is most natural to take being as a genus, that is, to hold that a general theory of existence will give us a univocal notion of being that is equally true of all things that do, in fact, exist. When we contemplate all the things that do and do not exist, we seem to have a uniform notion we are using that characterizes the former group but not the latter.
However, not all philosophers agree that being is a genus. To understand their main argument against such a view, consider a heap of sand. Each grain of sand exists. So does the heap itself taken as a whole with the grains as parts. Those who deny that being is a genus would say that existence itself, possessed by the grains and the whole heap, is very different in the two cases. The grains have independent existence (they could exist outside the heap), but the heap itself has dependent existence.
But should we explain the difference between the heap and the grains as a difference in two distinct kinds of existence? Probably not. To see this we need to make a distinction between what it is to exist (i.e., existence in and of itself), on the one hand, and the different kinds of things that have existence, on the other. We need not explain the heap and sand example by saying that there are two different kinds of existence involved (when existence is considered in and of itself). Rather, we can say that the heap and the grains both exist in the same sense (and we will look at different views about what existence itself is shortly), even though they are two different kinds of things that have existence: The heap is a dependent thing that has existence and the grains are independent things that have existence. In short, the grains and the heap do not have different kinds of existence; instead, they are different kinds of things that have existence.
2.2 Theories of Existence
Given that being is a genus, we are in a position to try to find out what existence or being is, in other words, to formulate a general theory of existence applicable to all entities whatever. In this section, we will look at theories of existence in two steps: (1) general characteristics a theory of existence ought to have; (2) different accounts of what existence is.
2.2.1 Five Characteristics of an Adequate Theory of Existence
There are five different traits that a good theory of existence ought to have. For one thing, it needs to be consistent with and explain what actually does and does not exist. Second, it needs to be consistent with and explain what could have existed but either does not exist or is not believed to exist (perhaps falsely) by the person advocating a given view of existence. For example, even though unicorns do not, in fact, exist, they could have existed. God could have made a world where Pegasus exists in the same sense that Fury exists. Physicalists do not believe the soul is real, but even if they are right (see chaps. 11–14), surely existence is such that souls could have existed.
Third, a theory of existence must allow for the fact that existence itself exists. To put the point differently, it must not be self-refuting. For example, if someone claims that to exist is the same thing as being inside space and time (existence itself is being spatiotemporally located), then, on at least some views, space and time would not themselves exist since they are not inside space and time. Whatever existence amounts to, one thing is clear: it makes a real difference in the world, and it must itself exist in order to make such a difference. If existence itself does not exist, then nothing else could exist in virtue of having existence.
Fourth, a theory of existence must not violate the fundamental laws of logic: the laws of identity (P is identical to P), noncontradiction (P cannot be both true and false at the same time in the same sense), and excluded middle (P must be either true or false). Contradictory states of affairs, for example, a square circle or its simultaneously raining in Grandview and not raining in Grandview, do not exist. Moreover, something must either exist or not exist, and nothing can exist and not exist at the same time.
Fifth, a theory of existence must allow for the existence of acts of knowing. Since a theory of existence is a theory, it will depend for its rational acceptability on knowledge that people have. Now an act of knowing something is, among other things, a conscious act of a person. Thus any theory of existence that denies the existence of conscious persons who know things is false. It has been said, not unfairly, that some versions of physicalism suffer from this problem (see chaps. 11–12).
2.2.2 Different Theories of What Existence Itself Is
Someone might think that Christian theology has a fairly straightforward answer to the question “What is existence?”: to exist is to be created by God or to be God. Now it is certainly true that everything other than God has been created. But this answer will still not do. Why? Because it is circular. Suppose we want to know what it means to say “X exists,” and we answer, “X is created by God.” If we now ask what this statement means, we will answer, “For God to give X existence.” Thus “X exists” will amount to “X is given existence by God,” and we have made no progress. We want to know exactly what God does when he creates something. To answer this, we need a theory of existence.
Several theories of existence have been offered. Here are some that are inadequate. To exist is to (a) be located in space and time; (b) be physical; (c) be causally efficacious, that is, to be capable of being an efficient cause (that by means of which an effect is produced) or of being acted on by an efficient cause; (d) be an event or a bundle of events; (e) be perceived or be a perceiver; (f) be a property; (g) be a property of properties (e.g., to be a second-order property of first-order properties).
We will not apply each of the five features of a theory of existence to each view of existence, but it would be a good exercise to do. However, a few remarks should give you a feel for how to use the five features listed above to evaluate different views of existence. Feature three (existence itself exists) would seem to rule out (a) and (c). Space and time are not themselves located in space and time; when an efficient cause produces an effect, the cause and effect enter into a causal relation and that relation itself is not causally efficacious, yet it exists.
Feature two (what could have but does not exist or is not believed to exist) counts against (e) (which was held by George Berkeley [1685–1753]). Surely, mountains and dinosaurs could exist even though no one is looking at them; and even though God constantly looks at mountains even if humans do not, it is not God’s gaze that gives mountains being, but his creative act of speaking them into existence and continually holding them in being. Moreover, one could be “seeing” a mountain in a hallucination, and it would not follow that a real, extra-mental mountain existed. Feature two also counts against (b) (as well as others on the list) because disembodied existence is surely possible, even if it is not real, and (b) rules out the very possibility of disembodied existence since it would not be physical existence. It also makes God’s existence metaphysically impossible, and this is quite a strong claim.
Feature one (what actually does and does not exist) eliminates (b) and (d) if minds, God, values, and abstract objects exist, since they are neither events, groups of events, nor physical entities. It also rules out (g). If (g) is true, then existence is a second-order property of properties. That is, existence is something that truly applies only to properties; for example, to being red, hardness, triangularity, humanness. But we all know that, in addition to properties, various individuals exist as well, such as a specific red ball. If existence is a feature only of properties, then individuals could not have existence; since they do, in fact, have existence, (g) must be false.
The point here is to illustrate ways that features of a theory of existence can be used to evaluate alternative proposals for existence itself. So far, nothing has been said about (f): existence is a property. Throughout the history of philosophy, many thinkers (e.g., Plato and Descartes) have held that existence is a property in the same sense that redness or being square is a property. For a ball to be red is for it to have the property of redness; for it to exist is for it to have the property of existence. Now at first glance, something seems right about this proposal. We do say, quite appropriately, that the ball has redness and that it has reality as well. In some sense, then, things can be said to have or not have existence.
However, something seems wrong with this proposal too. Existence is simply not a normal property like redness. This point was made by Immanuel Kant (1724–1804) in Critique of Pure Reason (A 600/B 628):
By whatever and by however many predicates [properties] we may think a thing—even if we completely determine it—we do not make the least addition to the thing when we further declare the thing is. Otherwise, it would not be exactly the same thing that exists, but something more than we had thought in the concept; and we could not, therefore, say that the exact object of my concept exists.
Kant’s point can be understood in this way. When you think of a ball, it adds to your conception of the thing to be told that it is red. But it does not add to your conception of the ball to be told that it exists. Put differently, saying that a ball is red tells us something about the character of the ball. But saying that it is real says that the ball, with all of its properties, does in fact exist. Thus existence does not relate to the ball like being red does.
Kant seems right here; nevertheless, saying that the ball is real does add something because there is, in fact, a real difference between existence and nonexistence. Can we shed more light on what this difference is? Perhaps so. Consider the statement “Tigers exist.” This would appear to assert the following: (1) The property of being a tiger (2) belongs to something (an individual tiger, say Tony). In chapter ten we will look at this “belonging” relation, but for now, it should be noted that it has gone by several names: exemplification, predication, and instancing. The claim that tigers exist is the claim that the essence of being a tiger (the what of being a tiger) is actually exemplified by or belongs to something (the that or fact of an individual tiger existing).
Note that when “Tigers exist” is broken down into the two aspects above, (1) refers to the essence or nature of being a tiger and (2) expresses reality or existence. From this we learn two things. First, there is a difference between a thing’s essence and its existence. Knowledge of what a tiger is does not tell us that tigers exist. There is a fundamental difference between essence (whatness) and existence (thatness).
Second, existence is not a property that belongs, but is the belonging of a property. Existence is the entering into the predication or exemplification relation, and in general, the following characterization of existence seems to fit the five features of a theory of existence: existence is either the belonging of some property or the being belonged to by a property or, more simply, the entering into the nexus of exemplification. In the case of Tony the tiger, the fact that the property of being a tiger belongs to something, and that something has this property belonging to it is what confers existence. Existence is the entering into the predication or exemplification relation.
How does this view square with the five features of a theory of existence? It would seem to account for everything that does or could exist and that does not or could not exist. Things that exist have properties. When something such as Zeus fails to exist, there is no object Zeus that actually has properties. Since unicorns could have existed, this means that the property of being a unicorn could have belonged to something. It would also account for existence itself existing because the belonging-to (exemplification, predication) relation is itself exemplified (a nonfictional, real tiger named Tony and the property of being a tiger both enter into this belonging relation) and the belonging-to relation exemplifies other features (e.g., it has the property of being a relation that belongs to it). Finally, this view of existence does not violate the fundamental laws of logic, nor does it rule out the existence of acts of knowing.
In sum, we have learned three things from our brief discussion of existence: (A) There is a genuine difference between existing and not existing. (B) This difference is not a normal property like the property of being red. (C) Existence is not part of the essence or “whatness” of ordinary entities; that is, for ordinary entities there is a difference between essence and existence.
2.2.3 Final Observations About Existence
There are three further remarks to be made about existence. First, our characterization of existence allows us to specify some other notions: coming-to-be and perishing. Since we already have an idea of what existence itself is, and since coming-to-be and perishing involve gaining and losing existence, then these latter notions can be understood in terms of our general theory of existence itself. The symbol = Df may be read as “is defined as.”
E comes into being = Df There is at least one property which is such that E has that property, and there is no property which is such that E had that property.
E perishes = Df There was at least one property which was such that E had it, and there is no longer any property which is such that E has it.
When something comes into existence, there must be at least one property that belongs to that thing. For example, when a human being comes-to-be, then the property of being human belongs to that individual at that moment. When something ceases-to-be, it no longer has any properties whatever. Coming-to-be and perishing should be kept distinct from what philosophers call alteration. An example of an alteration is an apple’s going from sweet to sour. Alterations are types of change. Before change is possible, two things must be true: (1) the thing that is changing must exist, and (2) the thing that changes must exist at the beginning, during the process, and at the end of change. In the example above, the apple exists and continues to exist while it is sweet, during the time it changes to being sour, and while it is sour. An alteration is a case in which a thing changes in the properties it has; it is not a case in which something changes with respect to existence itself. Alterations presuppose and, therefore, cannot be the same thing as a change in existence itself.
A second remark about existence is this: nothingness is just that—nothing. Nothingness has no properties whatever. Things that do not exist have no properties. For example, the flying horse Pegasus has no properties, and that is what his nonexistence amounts to. You may think that he has the property of being a winged horse. But that is not true. Our concept of Pegasus (which is in our minds when we are thinking of Pegasus) is a concept of something that would have the property of being a winged horse if it existed. But Pegasus does not exist, and he, along with all other cases of nothing, has no properties. It should also be said, however, that one’s concept of Pegasus exists, it is in the mind of the person knowing the concept, and the concept itself does not have the property of being a winged horse.
Many philosophers also hold that negative properties do not exist either. For example, an apple may positively possess the property of being red, sweet, and round, but, if negative properties exist, then the apple would also have the properties of not-being-green, not-being-square, and indeed, the apple would have an infinite amount of negative properties (e.g., of not-being an elephant). However, it is more natural to say that the apple fails to have the property of being green, instead of claiming that it has the property of being not-green. In general, when it is not the case that x is F (the apple is not green), we can explain this as a case where x fails to have F, rather than asserting the positive existence of negative properties, for example, as a case where x has not-F.
Finally, something should be said about a position in the history of philosophy called the modes of being view. According to this view, being is a general category, and existence is just one kind or mode of being (see figure 9.1).
Things that actually exist (e.g., lions) have existence. Fictional objects like Pegasus do not have existence, nor do they not exist. Rather, they have being and thus are real in a lesser way than those entities that actually exist. Existence is just one kind of being, and nonexistent fictional objects have being even though they do not exist. On this view, there are degrees of reality, and the type of being that Pegasus has is halfway between existing and not-existing.
Figure 9.1. Existence as a mode of being
In our view, this is an extravagant position. There is no good reason to believe it, and it violates the fundamental laws of logic applied to the being of things. Something either does or does not have being, and everything either has or does not have being. On the modes of being view, Pegasus does and does not have being.
Moreover, the modes of being view has an inadequate treatment of coming-to-be and perishing. Coming-to-be and perishing are all-or-nothing affairs. They are not like walking into a library where you start outside, move to 10 percent inside, then 50 percent, and finally you get all the way in. Coming-to-be and perishing are not gradual processes like that. Why? Existing or not existing is a matter of having or not having at least one property and, for any property, something either does or does not have that property. When something comes-to-be, it does not get a partial foothold in existence, and slide into being until it is fully real. In situations where something grows or diminishes in intensity (e.g., a sound gets louder or softer), these are cases of alteration, not cases of gradual coming-to-be and perishing. It would seem, then, that the modes of being view is inadequate.
3—The Nature of Identity
Existence is a fundamental metaphysical issue. So is the nature of identity. When philosophers talk about the problem of identity, usually one of these four issues is in mind:
Questions 1 and 2 are basic, metaphysical questions. Question 2 focuses on sameness through change, and we will study it when we look at the metaphysics of substance (chap. 10) and personal identity (chap. 16). Question 1 is the most basic metaphysical question about identity, and it will be the subject of investigation shortly. Question 3 is basically an epistemological question, not a metaphysical one. Often, question 3 is confused with questions 1 and 2, as we will see in our treatment of personal identity in chapter sixteen. Finally, question 4 is a matter of the philosophy of language. Its main concern is not identity itself, but identity statements—linguistic expressions that assert identity. We will look at this question later in the chapter. Let us begin our study of identity by looking at the issue surfaced in question 1.
3.1 The General Nature of Identity Itself
Suppose you wanted to know whether J. P. Moreland is identical to (is the same thing as) Eileen Spiek’s youngest son. If “they” are identical, then, in reality, there is only one person: J. P. Moreland, who is (identical to) Eileen Spiek’s youngest son. If they are not identical, then there are two people, not one. There is a general law of identity known as Leibniz’s law of the indiscernibility of identicals:
(x)(y)[(x = y) → (P)(Px ←—→ Py)]
This principle states that for any x (e.g., that person who is J. P. Moreland) and for any y (that person who happens to be Eileen Spiek’s youngest son), if “they” are identical to each other (“they” are, in reality, the very same entity), then for any property P (being 5’8”, being human), P will be true of x (J. P. Moreland) if and only if (represented by ‘ ) P is true of y (Eileen Spiek’s youngest son). In general, everything is what it is and not something else. Everything is identical to itself and thus shares all properties in common with itself. This implies a test for nonidentity or difference: if we can find one thing true of x that is not true of y or vice versa, then x is not identical to y.
We can gain further insights about identity if we think about how it applies to events. Let us assume what is called the property exemplification view of events: an event is the coming, continued possession or going of a property by a substance at or through a time. A leaf losing greenness at noon would be an example of an event in which a substance (the leaf) loses a property (greenness) at a time (noon). If E, S, P, and t refer to an event, a substance, a property, and a time, respectively, then Leibniz’s law of the indiscernibility of identicals, applied to events, becomes this:
(E1 = E2) → [(S1 = S2) & (P1 = P2) & (t1 = t2)]
If E1 and E2 are identical, then the substances, properties, and times constituting “those” events will be identical as well. If the latter are not identical, then the events are different. For example, a case where S1 is different from S2 would be where two apples become red simultaneously; a case where P1 is different from P2 would be where a specific apple became red and sweet at the same time; a case where t1 and t2 are different would be when an apple became sweet at noon, turned sour overnight, and became sweet again the next day.
There is another law of identity that is most likely a false principle: Leibniz’s law of the identity of indiscernibles:
(x)(y)[(P)(Px ←—→ Py) → (x = y)]
In ordinary, nonsymbolic language, this says that, for all x and y, if x and y have all and only the same properties, then they are identical to each other. This principle is false because there is more to a thing than its properties. For example, we could have two red and round disks that had the very same color, shape, size, and so forth. They could share all and only the same properties but still be two disks and not one because an individual thing like a disk is not exhausted by its properties. We will say more about this in chapter ten, but for now, you should remember that “two” things that are identical will share all “their” properties, because “they” are the same thing. But two things can share all their properties and not be the same thing if the two things have more to them than properties.
The correct principle—Leibniz’s law of the indiscernibility of identicals—implies another insight about identity:
(x)(y)[(x = y) → □(x = y)]
In ordinary language, this says that for all x and y, if x is identical to y, then necessarily, x is identical to y. There is no possible world where that thing which is x is not identical to that thing which is y. For example, a cat may happen to be yellow and twenty pounds, but it does not just happen to be identical to itself. It is necessarily identical to itself. The fact that something is identical to itself is a necessary feature of everything. Suppose that person who is J. P. Moreland and that person who is Eileen Spiek’s youngest son are different but both are 5’8”. Then, in the actual world, they do not differ in height. However, if it is just possible for them to differ in height—if there is a possible world where one is 5’8” and the other is 6’, then they are not identical.
Again, if disembodied existence is metaphysically possible—if there is a possible world with disembodied existence—then a person cannot be identical to his or her body because there is no possible world where the person’s body exists and is disembodied. Since it is possible for a person to exist disembodied, but it is not possible for a body to exist disembodied, then a person is not identical to his or her body. Why? Because something is true of the person (the possibility of disembodied existence) that is not true of his or her body.
The identity relation is a relation that everything has to itself and to nothing else. This relation should be kept distinct from three other notions with which it is sometimes confused: cause-effect, coextensionality, and inseparability. With regard to cause-effect, if A causes B, then A is not identical to B. Smoke causes fire as its effect, but smoke is not identical to fire. Further, two things can be coextensional; coextensionality means one obtains if and only if the other obtains. For example, the property of being triangular is coextensive with the property of being trilateral. One obtains if and only if the other obtains; no object has one without the other. If the two properties were identical, then whatever is true of being triangular would be true of being trilateral, and conversely. But the two properties are not identical because the property of triangularity has something true of it, namely, having angles, that is not true of the property of being trilateral.
Finally, identity differs from inseparability. Two entities can be parts of some whole and be inseparable either from each other or that whole, yet still not be identical. For example, the individual white instance of color in a sugar cube cannot be separated from the individual square instance of shape in that cube and still exist, like the leg and back of a chair can be separated from each other or from the chair taken as a whole. But the instance of whiteness in the cube is an instance of color and its instance of shape is not; thus they are not identical. Again, a person’s emotions cannot be separated from the person or from the person’s beliefs and placed in different locations of the room like the hand and leg of his body can. But one’s various emotions and beliefs are all distinct and not identical to other emotions or beliefs.
In sum, the identity-difference distinction is not the same as the cause-effect, the coextensionality, or the inseparable-separable distinctions. Before we look at identity statements, we should ponder some points made by the great medieval philosopher Francisco Suárez (1548–1617) in his work titled On the Various Kinds of Distinctions. Suárez discussed certain distinctions that shed light on identity and identity statements. First, he described real distinction. Two entities differ by means of a real distinction just in case they can be separated and still exist. For example, the different legs of a chair bear a real distinction to each other. Independence of existence is the key here.
Second, there is what Suárez called the distinction of reason. If “two” things differ by means of a distinction of reason, then “they” are identical. For Suárez, there are two kinds of distinctions of reason. One is the distinction of reasoning reason, and it arises solely because we use the same word twice in sentences like “Peter is Peter.” Here, there is no distinction that exists in reality, and it is merely the process of thought or language that gives rise to a distinction in which the same, identical thing is named (or thought of) twice.
Another distinction of reason is the distinction of reasoned reason. An example would be, “The red object is the sweet object,” said of a red, sweet apple or “The Evening Star is the Morning Star,” where each description refers to the planet Venus. When the distinction of reasoned reason is present, the “objects” referred to are identical (the apple or Venus), but the concepts or terms used to refer to the object (the red versus the sweet object, the Evening versus the Morning Star) do not exhaust the object in question and they express different nonidentical aspects of the same, identical object.
A third type of distinction for Suárez was called the modal distinction. If A is modally distinct from B, then (1) one of them is a constituent of the other and (2) one of them, say A, could exist without the other, B, but not vice versa. If A and B are the independent and dependent entities, respectively, then B is a mode of A; B is an inseparable aspect of A; B is dependent on A. For example, consider the property of redness. When redness is exemplified by or had by an apple, then at least these three entities are involved: the property of redness, the apple, and the-having-of-redness-by-the-apple. This latter entity is an instance of redness and is modally distinct from redness. Redness is modified by its being exemplified by the apple. Redness could exist without its being possessed by this apple, but the exemplification-of-redness-by-the-apple; that is, this specific instance of redness in this specific apple could not exist without redness. Redness is not identical to but is modally distinct from this instance of redness.
3.2 Identity Statements
The identity relation itself is independent of language users. The sun would be identical to itself if no language user would have existed. Nevertheless, we use identity statements to express claims of identity (e.g., “Color is identical to a wavelength of light”), so it is important to mention identity statements to see if we can learn something about them and about identity itself.
The ancients noticed that there was a brightest star in the evening sky, and they called this star the Evening Star or Hesperus. They also noticed that there was a brightest star in the morning sky, and they called this star the Morning Star or Phosphorus. However, the empirical discovery was made that the “two” stars were identical and were, in fact, the planet Venus. This discovery was contingent, that is, dependent on what the facts turned out to be. It could have turned out that the Evening Star and the Morning Star, understood as whatever turns out to be the brightest star in the evening and morning, respectively, were two different heavenly bodies. Now consider the following two statements:
There are a number of ways to understand identity statements like these, but two accounts are very prominent in philosophy. The first is called the traditional or objectual account of identity statements. On this view, an identity statement asserts that the thing referred to by the first term has a certain characteristic true of it, namely, it is identical to itself.
Now, this certainly does account for much of what we say when we use identity statements. However, the philosopher Gottlob Frege (1848–1925) raised a problem with the traditional account and offered a different understanding of identity statements. Frege argued that on the traditional view, statements (1) and (2) assert the same thing—the thing referred to by the first term (i.e., the planet Venus) is identical to itself. But this cannot be right because (1) and (2) do not assert the same thing. How do we know this? Statement (1) is not very informative. It tells us very little and seems to be necessarily true by definition. We know (1) is true before we make any empirical study of the heavens. However, (2) is very informative; the truth of (2) was an empirical discovery, and thus while an object’s identity to itself is necessary, (2) seems to be a contingent truth (it could have been false, and its truth depends on the discovery of certain empirical facts). It is not true by definition. Given that a thing’s identity to itself runs throughout all possible worlds in which it exists, how is one to distinguish (1) from (2), and more generally, how can there be contingent identity statements if identity is necessary?
Frege sought to answer these questions and, in the process, offered a different view of identity statements called the metalinguistic account (meta here means “about,” and metalinguistic means “about language”). On this view, an identity statement like (1) or (2) does not merely say something about the entity Venus referred to by using the words Hesperus and Phosphorus; more importantly, it mentions the words themselves; that is, it says something about the words Hesperus and Phosphorus. Identity statements are statements about language, and they assert that a certain relation holds between the two referring expressions used in the statement: they are coreferring expressions, or in other words, they each name the same thing.
Each account has something to be said for it, and it is beyond our purpose to weigh the strengths and weaknesses of each. However, one thing carries real importance. The identity relation itself is fairly straightforward and clear, but identity statements themselves are more ambiguous, and whenever you are looking at an identity statement itself, you should try to understand just exactly what the person using that statement is trying to say. For example, when someone says “Thank God that Beckwith is himself today,” we shouldn’t take this to mean that he literally was not identical to himself yesterday; rather, someone is claiming that he was acting out of character.
We have looked at two different theories of identity statements. Here are three different kinds of identity statements:
Meaning identity statements: These occur when the two referring expressions are synonyms. “A bachelor is an unmarried male” and “A motor car is an automobile” are examples. These can be found in dictionaries.
Referential or name identity statements: These occur when two proper names (names of individuals like “Tom Jones”) or two natural kind terms (terms that name naturally occurring kinds of things like “H20” or “the lion”) each function merely to tag or refer to its object of reference in every circumstance in which it could exist. “Mt. Everest is Chomolungma” and “Water is H20” are examples. “Water is H20” amounts to the statement, “The stuff out there that is actually water is identical to the stuff out there that is actually H20.” These statements are not found in dictionaries. Instead, they express necessary truths about the world.
Contingent identity statements: While an entity’s identity to itself is necessary, some identity statements are contingent, that is, they are true in some possible worlds and false in others. One context in which contingent identity statements occur is when two descriptions of something happen to be fulfilled by the same thing but could have been fulfilled by different things. Suppose Aunt Sally’s favorite color is blue. Then “The color of the sky is Aunt Sally’s favorite color” expresses an identity. But this identity statement, though true, is contingent. God could have made the sky green and Aunt Sally could still favor blue, or the sky could be blue but Aunt Sally could have preferred pink. Either way, “The color of the sky is Aunt Sally’s favorite color” would be false. These kinds of identity statements can be understood as saying this: “Whatever turns out to be the color of the sky is identical to whatever turns out to be Aunt Sally’s favorite color.” Identity statements can be contingent, but the fact that something is identical to itself is necessary.
The nature of identity and identity statements will be of crucial importance for looking at the mind-body problems in chapters eleven through fourteen. In the final section of this chapter, the notions of existence and of identity will be applied to an important metaphysical topic: replacement and reductionism.
4—Replacement and Reductionism
As human beings, we are interested in what is real and what is not. Important intellectual debates rage about the existence of God, the soul, values, and life after death, just to mention a few. Philosophically speaking, there are two important ways to deny the existence of some entity X: eliminate X and replace belief in it with belief in something else Y, or continue to believe in X but reduce X to Y.
To understand this, consider for a moment the history of science; scientific theories do not always last forever. Theory change happens in science, and theories come and go in two central ways: replacement and reductionism. Sometimes a change from one theory S to a new theory T involves an abandonment of S altogether and an elimination of the entities postulated to exist by S. An example of this kind of change, called replacement, was the change from phlogiston to oxygen chemistry. When a metal is heated in the atmosphere it weighs more after heating than before. In the eighteenth century, scientists explained this by saying that heating drives out an impurity in the metal known as phlogiston, which has negative weight.
In the late 1700s, A. L. Lavoisier’s oxygen theory replaced phlogiston. Oxygen was seen as an entity, different from phlogiston, with positive weight and that attached to a metal when it was heated in the air. The shift from phlogiston to oxygen theory was a theory change of replacement. Lavoisier’s model of oxygen was not seen as a better picture of phlogiston than that of rival scientists. Phlogiston was eliminated and replaced altogether; people no longer believed that phlogiston existed.
On the other hand, sometimes a theory change involves reduction, that is, the continued belief in some entity X, but X is no longer thought to be what theory S said it was; rather, it is now “reduced to” being Y, which the new theory claims it is. For example, scientists used to believe that when a body grows hot this is because it receives heat, understood as a subtle, weightless, invisible fluid known as caloric. Today, scientists believe that there is no such thing as caloric; instead, heat has been “reduced to” the vibration of molecules, for example, the average kinetic energy of a gas. Heat was not eliminated and replaced in this theory change. Scientists still believe in the existence of heat. Instead, heat was first thought to be X (caloric), but now it has been “reduced to” being Y (the vibration of molecules).
The idea of replacement is fairly clear. But what does it mean to reduce one thing to another? What is reductionism? Unfortunately, the notion of reduction in philosophy has a wide variety of meanings and no common usage is employed. However, here are some of the different senses of reductionism:
1. Linguistic Reduction. This occurs when language that uses one vocabulary or set of terms is replaced with language that uses another vocabulary or set of terms. For example:
a. The average family has 2.5 children.
a’. Add the children and divide that number by the number of families and you get 2.5.
In this case, sentence (a) uses the term “the average family,” and this is replaced by sentence (a’ ), where the term is replaced by a formula for calculating (with the terms add, divide, and number). Sentence (a) is the reduced sentence, and (a’ ) is the reducing sentence. In some cases of linguistic reduction, the term in the reduced sentence (e.g., “the average family”) is no longer taken to designate a real entity in the world because it is left out of the reducing sentence. Sentence (a) could lead someone to think that there was a family somewhere—the average family—that actually has 2.5 children. But through linguistic reduction we see there is no such entity. However, in other cases of linguistic reduction, the mere fact that truths said in one vocabulary can be said in another vocabulary does not, by itself, mean that the entities designated in the reduced sentence do not exist. For example:
b. There are three triangular things in the room.
b’. There are three trilateral things in the room.
c. A solid body floats on a liquid if its specific gravity is less than that of the liquid.
c’. A solid body floats on a liquid if the quotient of its weight and volume is less than the corresponding quotient for the liquid.
d. Red is a color.
d’. Red things are colored things.
In the example of (b) and (b’ ), just because the truth value of (b) and (b’ ) are the same (i.e., b is true if and only if b’ is true), it does not follow that the thing expressed by the term triangular (the property of being triangular) can be reduced to (made identical to) the thing expressed by the term trilateral (the property of being trilateral).
In (c) and (c’ ), if we think of specific gravity as a real entity in the world, then we would think of it as a disposition or tendency to sink, a sort of heaviness possessed by a body. This is expressed in (c). Statement (c’ ) treats (c) in the same way statement (a’ ) treated (a). But just because we can replace (c) with (c’ ) and therefore eliminate the term specific gravity, it does not follow that specific gravity itself, understood realistically as a tendency to sink or a heaviness, does not exist.
Finally, in (d) and (d’ ), even if (d) could be linguistically reduced to (d’ ) (and in chap. 10 we will see that it cannot), it does not follow that the properties of redness and coloredness do not exist. Statement (d’ ) eliminates the terms red and color that seem to refer to these properties and uses only terms that refer to individual particulars (individual red things and colored things, like red balls and flags). In sum, linguistic reduction may or may not mean that the entity referred to in the reduced sentence does not exist.
2. Strong Ontological Reduction. This occurs when some entity x is reduced to (i.e., identified with) some entity y. In this case, x exists and is nothing but y; in other words, x is identical to y. Some have said (erroneously, in our view) that statements such as the following express strong ontological reductions: “Redness is a wavelength,” “Heat is the vibration of molecules,” “Pain is a certain brain state.” In each case one entity (redness, heat, pain) is reduced to another entity (a wavelength, a vibration, a brain state) in that the former entity exists and is nothing more nor less than the latter entity.
3. Weak Ontological Reduction. Here entity x is reduced to entity y in that x is caused by or explained by or dependent on y. Y is a sufficient condition for x. For example, wetness is reduced to the molecular structure of a group of water molecules in that this structure is what causes and explains wetness. In a strong ontological reduction, the wetness would be treated as being identical to the molecular structure. In a weak ontological reduction, the wetness is supervenient or emergent on that molecular structure. The molecular structure of a group of water molecules is different from and not identical to the wetness of that water. In this example, emergence means that the structure is a sufficient condition for that wetness to emerge. Supervenience means that, given the structure, the wetness will supervene on (be caused or explained by or dependent on) that structure.
In sum, debates about the reality of some entity or other include attempts to replace that entity with another one or to reduce that entity to another one in different ways. It should be clear that such replacement and reduction attempts must be evaluated on a case-by-case basis. And when such evaluations are made, a broad philosophical understanding of the nature of existence itself, identity and identity statements will be relevant to such assessments.
There is one more lesson to be learned from this chapter. If we define naturalism (see chap. 10) as the view that everything that exists is in space and time, and if physicalism is taken to be a version of naturalism, saying that the only entities that exist are merely physical entities that are in some way accessible to the senses and to scientific investigation used in chemistry and physics, then physicalism is false.
So understood, physicalism implies that everything that exists can be, at least in principle, exhaustively described using the language of chemistry and physics. This is so because the only entities that exist are completely physical (chemical and physical) entities. However, there are two things about all entities whatsoever, including so-called physical entities like chairs, rocks, and protons, that are not physical: what it is that accounts for the entity’s existence (the having of properties by the entity in question) and the entity’s being identical to itself. These are metaphysical facts about all existents, including all physical existents.
Thus, while there may be entities that are at least physical in that they can be partially described in the language of physics and chemistry, there are no entities that are only physical, that is, that can be exhaustively described using only chemistry and physics. There will always be more to a physical object than its chemical and physical aspects. These will be the more fundamental aspects of existing and self-identity. In this way we see that science is only a type of knowledge of even physical things like electrons. Metaphysics gives us a more fundamental knowledge of the metaphysical, nonmaterial aspects of those physical things.
Chapter Summary
Being is a genus, and there is therefore a general notion of existing applicable to all entities. A good theory of existence should account for what does/does not and could/could not exist, it should imply that existence itself exists, it should not violate the fundamental laws of logic, and it should imply that acts of knowing exist. Several theories of existence fail to pass these tests. However, one view seems to succeed. Existence is either the belonging of some property or the being belonged to by a property. This understanding of existence was then used to define coming-to-be and perishing, which are distinct from alteration. Nothingness has no properties and, therefore, does not exist. Negative properties do not exist either; rather, things can fail to have a property. Finally, the modes of being view says that fictional objects have being but not existence. This position was rejected.
Leibniz’s law of the indiscernibility of identicals is a true law expressing the identity relation, but the identity of indiscernibles is false. The identity relation should not be confused with any other relation, for example, cause-effect, coextensionality, or inseparability. Suárez’s three distinctions are helpful in understanding identity and identity statements. The debate between the objectual and metalinguistic accounts of identity statements illustrates how ambiguous those statements can be. Three main types of identity statements are meaning, referential, and contingent identity statements.
Debates about the existence of some entity often involve replacement or reductionism. Three main types of reduction are linguistic, strong ontological, and weak ontological reduction. Physical entities like a carbon atom are at least but not only physical because the having of properties (existence) and the identity relation are not physical entities.
Checklist of Basic Terms and Concepts
alteration
cause-effect
change
coextensionality
coming-to-be
contingent identity statement
difference
distinction of reason
distinction of reasoned reason
distinction of reasoning reason
emergence
essence
events
exemplification
existence
fictional objects
genus
identity
identity statements
inseparability
instancing
Leibniz’s law of the identity of indiscernibles
Leibniz’s law of the indiscernibility of identicals
linguistic reduction
meaning identity statements
metalinguistic account of identity statements
modal distinction
modes of being
negative properties
nothingness
traditional (or objectual) account of identity statements
perishing
predication
property exemplification view of events
real distinction
reduced sentence
reducing sentence
referential (or name) identity statements
replacement
strong ontological reduction
supervenience
weak ontological reduction