Ibn Sīnā's Solution to Kant's Challenging View of Existence
Jerome Shaffer (1962) claimed that Kant's view on existence leads to two problems—what I call "the problem of contradiction" and "the problem of insignificance." The former shows that existential propositions are analytic, while Kant asserted that they are synthetic. According to the latter, Kant's view implies that in the act of predicating either, the subjects do not have and cannot capture their exact extensions and/or no predicate could be a "real" predicate. After formulating the two problems, Ibn-Sīnā's views concerning essence, existence, and their relations are explained. Then, it is shown how the problems can find satisfactory solutions on the basis of Ibn-Sīnā's views.
Kant has explained his view on modality and modal concepts in general in different sections of the Critique of Pure Reason (Kant  2003) such as the "Metaphysical Deduction" (B100–102), "Schematism" (B184/A145), and the "Postulates" (A218–235/B265–287). However, he discusses the issue of existence in particular mainly under the topic of "The Impossibility of an Ontological Proof of the Existence of God." The proof—known as the "ontological argument"—was first presented by St. Anselm and later revised by Descartes, Leibniz, and contemporary philosophers like Norman Malcolm (1960) and Alvin Plantinga (1965). Kant's view on existence, however, seems to lead to some problems. In the present essay two of these problems are discussed—the problems that I call "the problem of contradiction" and "the problem of insignificance." The former shows that Kant's view of existence implies that existential propositions are analytic, while Kant asserted that they are synthetic. According to the latter, Kant's view implies that, in the act of predicating, either the subjects do not have and cannot capture their exact extensions and/or no predicate could be a "real" predicate.
To remove these problems, many philosophers have tried to interpret Kant on the basis of what Jonathan Bennett (1974, p. 228) calls "the Kant-Frege view of existence," according to which "exists" is merely a second-level predicate that can be ascribed only to concepts and not to individuals. Although I believe that this way of solving the problems is unsuccessful, I do not intend to discuss it here. In effect, I do not believe that the two above-mentioned "problems" are genuine problems that could not be resolved through Kant's own resources. As a matter of fact, my aim is to show that the charges of "contradiction" and "insignificance" are not genuine threats in Kant's view of existence and could be addressed while remaining within Kant's own text.1
However, this goal is reachable provided that Kant's view on existence is understood and interpreted on the basis of the opinions of his preceding philosophers—like Descartes, Gassendi, Aquinas, and, before them, Ibn Sīnā (Avicenna). In other words, my main thesis is to show that the solution to the "problems" lies in reading Kant on the basis of the views of such previous great figures, and not on the basis of the post-Fregean interpretation of Kant.2 It is demonstrated that this approach is not a superficial interpretation of Kant, a view that has somehow been imposed on him. [End Page 112] Rather, by quoting from Kant's own works, Kant's basic agreement with this approach is shown.
I start by formulating the two problems of "contradiction" and "insignificance." Then I will very briefly look at different modern approaches to the problems. After explaining the views of some great philosophers before Kant (in particular, Ibn Sīnā) concerning essence, existence, and their relations, it will be shown how the problems can find satisfactory solutions on the basis of these views.
The Problem of Contradiction
One of Kant's challenges to the ontological argument for the existence of God is based on making a distinction between "real" (which Kant also calls "determining") and non-real predicates. Kant believes that "[a]nything we please can be made to serve as a logical predicate; the subject can even be predicated of itself; for logic abstracts from all content" (A598/B626). Kant, of course, does not specify what a logical predicate is, although it seems that he uses "logical" in the sense of "grammatical." The real predicates, however, are logical predicates that, according to Kant, have some discriminative specifications:
[A] determining [i.e., a real] predicate is a predicate which is [i] added to the concept of the subject and [ii] enlarges it. Consequently, [iii] it must not be already contained in the concept.(Kant A598/B626; brackets added)
"Being" or "existence," according to Kant, does not satisfy these conditions. So,
[iv] "Being" is obviously not a real predicate; that is, it is not a concept of something which could be added to the concept of a thing.(A598/B626; brackets added)
On the other hand, Kant believes that
[v] analytic propositions add "nothing through the predicate to the concept of the subject, but merely [break] it up into those constituent concepts that have all along been thought in it, although confusedly"; whereas synthetic propositions "add to the concept of the subject a predicate which has not been in any wise thought in it."(Kant, B11)
Combining [iv] and [v] has this result:
[vi] Existential propositions, in which the predicate is "being" or "exists" are analytic.
Kant, however, makes a claim that is in clear contradiction to this:
[viii] [W]e admit, as every reasonable person must, that all existential propositions are synthetic.3(A597/B625)
Regarding this problem, Jerome Shaffer (1962, p. 310) sees it as astonishing that Kant's argument against the ontological argument "has stood up for so long and is still commended by philosophers." [End Page 113]
The Problem of Insignificance
Why does Kant (A598/B626) believe that "exists" as the predicate cannot be added to the concept of the subject? Kant's argument can be summarized as follows:
To explain the argument above, consider a phrase like "dark matter." The concept that we get when we use this phrase is the indicator of the nature and properties of its object, according to the best knowledge we have of dark matter, even though we might not have a detailed knowledge of all properties of dark matter. So, in saying that "Dark matter exists," we utilize the concept of "dark matter" to say something about its extension (i.e., dark matter). This statement, however, does not declare that dark matter has a property (i.e., existence) in addition to the properties that we had already thought of in the concept of the subject, that is, "dark matter." Accordingly, "existence" cannot be a concept of something that is added to the concept of the subject.
In other words, if "exists" were added to the concept of "dark matter" it would change (enlarge) it. In this case, the concept of the subject that we would get would be a concept like the concept of "dark matter plus existence," or "existing dark matter." This concept, then, would not represent what we originally intended as the referent of "dark matter," that is, dark matter whose properties do not contain something as "existence." Thus, if "exists" were added to the concept of the subject, the altered concept of the subject of existential propositions would represent either no extension or an object that we did not intend to refer to.
However, according to Kant, there are objects as the exact extensions of the subject of (existential and non-existential) propositions. By this Kant does not mean that subject-concepts designate or capture real objects in all their particularities. In "The Ideal of Pure Reason," for instance, Kant (A572/B600) states that whereas all actual objects are completely determined with respect to all possible predicates, concepts are, in theory, completely determinable. Again, in his later Logic Lectures, delivered in the 1780s and 1790s, Kant (1992)4 repeatedly suggests that concepts are abstract representations of objects and thus are always universal or general but never particular, and that the concepts themselves are universal but their use can be particular, that is, they can be applied to particular objects. Kant's idea is that a concept can always be further determined through adding new "real predicates" to it, but its level of determination will never match that of the individual object, for then it will not be a concept anymore. [End Page 114]
In spite of all this, the point is that, according to Kant, the subjects of all true propositions can designate and capture real objects as their true and exact extensions. This means that there are objects as the exact extensions of the subjects of existential propositions. So, the predicate "exists" cannot be added to the subject and change its concept to something else; otherwise, the subjects of existential propositions cannot designate their real extensions. That is to say, "exists" is not a real predicate, since "it is not a concept of something which could be added to the concept of a thing" (A598/B626).
If this is the case, however, as Shaffer argues (1962, pp. 309–310), Kant's view makes it impossible for there to be any real predication at all. For the argument above can be applied to all propositions. So, even if we have a real predicate, the subject of our proposition can never represent its object. For example, if I state "Gold is yellow," the predicate "yellow" would change the concept of the subject from "gold" to, say, "yellow gold." I would then be unable to refer to the exact object of my concept, that is, the one (the gold) that I originally conceived, and say that it is yellow. So, "[t]he argument which shows that 'exists' is not a 'real' predicate also shows that nothing could be one" (Shaffer 1962, p. 310). Therefore, in the act of predicating, either the subjects do not have and cannot capture their exact extensions (epistemic idealism) or nothing could be a real predicate (all propositions become analytic, tautologous, or pseudo-propositions). Let's call this problem the "problem of insignificance" (to be examined later).
In sum, it seems that Kant's view about existence leads to (1) a contradiction (the problem of contradiction) and (2) a dilemma according to which we have to choose either a kind of idealism or accept that all propositions are analytic, tautologies, or pseudo-propositions (the problem of insignificance).
Some Reactions to the Problems of Contradiction and Insignificance
Surprisingly, there are a remarkable number of philosophers who, ignoring the aforementioned problems, believe that Kant's objections against the ontological argument are conclusive.5 Most of them also see Kant as a pioneer philosopher in introducing what nowadays is known as the Fregean view of existence, according to which "existence" is not a property of individuals and "exists" is not a real predicate.
Few philosophers, however, have taken the problems above seriously, either in this form or another. If they have, either they have found Kant's view on existence untenable or they have tried to interpret Kant in such a way that resolves the problems. For example, Shaffer (1962) and J. Michael Young (1976) belong to the first group. Young (1976, p. 95) argues that Kant's view on existence leads to a paradoxical conclusion according to which existential judgments "are not really judgments at all," neither synthetic nor analytic, and hence they "are not capable of being true or false." So, "there is no way in which existential judgments could be understood as predicative or attributive" (Young 1976, p. 101).
Among those who believe that a proper interpretation of Kant can remove the above-mentioned problems we can mention George R. Vick (1970), Richard Campbell (1974), Edgar Morscher (1985/1986), and Nicholas Everitt (1995). Vick (1970, p. 357) [End Page 115] believes that Kant "regarded existence as something more than a merely logical or grammatical predicate." Kant, according to Campbell (1974, p. 99), considers "exists" as a genuine predicate, although it is a "relational predicate," and hence is not a determining predicate to operate for characterizing what a thing is. Morscher (1985/1986, p. 282) claims that Kant's doctrine is that "'existence' is not a first-order predicate … but it is a second-order predicate," which is attributed to concepts, not individuals.6 Everitt (1995, p. 400) believes that "Kant's objections to the ontological argument are entirely epistemological." So, Kant does not deny that there can be true analytical existential judgments; rather "he means that existential judgments as known by us are synthetic" (Everitt 1995, p. 404; italics in original).7
What can be said about these positions on Kant's view? First of all, it seems that the problems of contradiction and insignificance should be taken seriously. Without solving these problems Kant's view would remain untenable. Second, Kant's arguments can be taken at their face value; that is, they can be considered as making ontological, and not epistemological, claims. The next point is that Kant's views on existence are consistent and unproblematic. To show this, however, we do not need to appeal to some interpretations that are in conflict with the appearance of Kant's words. Nor is it required that we consider them from a Fregean point of view. In effect, it would be misleading if Kant's view on existence is interpreted on the basis of the post-Fregean logical approach according to which existence is merely a quantifier or is a second-level predicate.
Given Kant's own scholastic background through Baumgarten and Wolff, the most natural way, therefore, which seems to be the most reliable approach, too, is to view Kant's position on the issue of existence on the basis of its philosophical background, that is, the Cartesian and Thomist traditions, which were undoubtedly under the influence of Ibn Sīnā's views concerning existence, essence (whatness),8 and their relations. This point becomes clearer and more acceptable if it is noticed that Kant discusses the issue of existence under his criticisms against Descartes' reformulation of the ontological argument. These criticisms in some respects are indeed a repetition of Gassendi's objections,9 which are in effect echoes of Ibn Sīnā's ideas. So, to make a proper interpretation of Kant's position we need to recall the views of some great philosophers preceding Kant on this issue. By appealing to such a background we can interpret Kant's approach in such a way that both problems—of contradiction and of insignificance—find a satisfactory solution. I start with an explanation of some terms.
Epistemically, the "what" in questions like "What is this thing?" and "What is it like?" forms a basic way in which our minds try to make sense of the things around us and to identify what a thing is. The questions can be about all kinds of entities: individuals; properties; universals; material and immaterial, concrete and abstract mental entities; and so on. To answer the questions we use whatish (māhovi) concepts that [End Page 116] describe and represent the "whatness" (māhiyyah) of entities. These concepts also indicate the difference between an entity and other things when we intend to define the entity. This point shows that the concept of existence is not a whatish concept. We shall return to this point later.
Ontologically, however, "whatness" is that which is about some genuine aspects of entities themselves, not their concepts. When we say that "The Sun is a huge star," "Blue is a cold color," and "Minds are immaterial entities," we refer to that which shows what an entity itself is, that is, to the elements that constitute its nature, and not its concept. The whatness of an entity, therefore, comprises all of its properties and represents all of its aspects (its nature, essence, substance, properties, relations, or the combination of all these) except its existence.10 This means, by definition, that there is nothing of an entity that can be out of its whatness or its existence such that it belongs neither to the whatness nor the existence. Accordingly, by whatish properties I mean all properties (intrinsic, extrinsic, essential and accidental, primary and secondary, relational and non-relational, etc.)11 that represent and belong to the whatness of entities.
It is important not to confuse the epistemological and ontological applications of the notion of whatness. Henceforth, however, by whatness I shall mean its ontological, not epistemological, application, that is, the whatness of an entity, and not its whatish concepts, unless otherwise stated.
It seems that the notion of existence (wujūd) is at once familiar such that we have no trouble with using the verb "exists." Yet the problems start when we try to define "existence" and explain what it is like to exist. Even philosophers disagree whether it is the task of ontology to give an answer to this problem, let alone what the nature of the problem and its solutions are. For example, philosophers like Frege (1960, pp. 48–49), Russell (1971, pp. 232–233), and Quine (1969, p. 97) claim that it is to quantificational logic to which we must turn in order to understand the nature of existence. However, philosophers such as Salmon (1987, p. 53), Gibson (1998, p. 1), and Munitz (1974, p. xiv) argue that it "is the task of ontology, not of logic, to de velop a theory of existence."
At any rate, the scope of the present essay does not allow discussing different approaches to this issue. I only mention that providing a non-circular definition of existence is highly unlikely, if not impossible. So, analogous to the notion of knowledge in Timothy Williamson's (2000) interpretation, I assume existence as a primitive, fundamental, and unanalyzable notion. That is, there are not any more fundamental or simpler components and conceptual elements out of which existence may be said to be composed, or to which it may be reduced, or in terms of which it could be defined.
In effect, most great philosophers have taken such a position. Ibn Sīnā (Morewedge 1973, p. 15), for example, says that "existence is recognized by reason itself without the aid of definition or description. Since it has no definition it has neither genus nor differentia because nothing is more general than it," or "existence does not [End Page 117] have a description since nothing is better known than it." Likewise, Descartes (1984–1991, 2 : 418) wrote: "there is no need here for a definition [of existence], which would confuse rather than clarify the issue." Leibniz (Adams 1974, p. 221) also takes the same position: "Existent cannot be defined … in such a way, that is, that some clearer notion might be shown to us," and "Existence therefore is a noncomposite, or unanalyzable (irresolubilis) notion." However, like the notion of knowledge, we see that generally people have no difficulty with using the verb "exists." Everybody understands, more or less, the notion of existence.
The important point, however, is to make a distinction between the "concept of existence," which as a mental entity exists in the mind, and "genuine existence," which as the reality of existence exists externally. All later discussions of existence and whatness must be understood in light of not confusing the conceptual and external applications of the notions of existence and whatness.
The Unity of the Notion of Existence
One important discussion concerning the notion of existence is whether existence applies to all existents with a single, unequivocal meaning or with different meanings, with a kind of equivocation. "Univocalism," as the prevailing thesis (Berti 2001), was supported by such philosophers as Aristotle, Russell, Quine, White (1956), and van Inwagen (2001). For example, Aristotle in his Metaphysics IV (1003a30–1003b10) wrote:
There are many senses in which a thing may be said to "be," but they are related to one central point, one definite kind of thing, and are not homonymous.… So, too, there are many senses in which a thing is said to be, but all refer to one starting-point.… It is for this reason that we say even of non-being that it is non-being.(1995)
Also, Quine (1960, p. 131) wrote: "There are philosophers who stoutly maintain that 'exists' said of numbers, classes, and the like and 'exists' said of material objects are two usages of an ambiguous term 'exists'. What mainly baffles me is the stoutness of their maintenance. What can they possibly count as evidence?" Similarly, van Inwagen (2001, p. 17) argues that "the univocacy of number and the intimate connection between number and existence should convince us that there is at least very good reason to think that existence is univocal."
On the other hand, philosophers like Ryle (1945, 1949) and Rescher (1978) have defended "equivocalism." According to Ryle (1945, pp. 15–16), "It may be true that there exists a cathedral in Oxford, a three engined bomber, and a square number between 9 and 25.… The senses of 'exists' in which the three subjects are said to exist are different and their logical behaviours are different." Likewise, Rescher (1978, p. 64) argues that existence is an equivocal notion since there are different kinds of existents upon which existence is not applied uniformly:
A close analysis of the meaning of these several modes of existence12 accordingly shows them to be distinct and variegated. They do not share a common explication.… [There is] [n]o common meaning.… The several senses of "exists" are linked together—at best—by a set of family resemblances.… [E]xistence is, in the end, an emphatically equivocal conception. [End Page 118]
Rescher's argument, however, is inconclusive. To explain the issue, first of all, as we mentioned before, "the concept of existence" should be distinguished from "genuine existence." Concerning the former, what we already explained affirms what Gibson (1998, p. 1) asserts: "The concept of existence … is an absolutely elementary concept." That is, the meaning of "exists," that is, the concept of existence (qua existence) is single and similar even if it is used in substantially different existents, that is, in cases like "God exists," "Dust exists," or "Number 7 exists." In other words, in spite of all differences among the cases that the notion of existence is applied to, we intuitively believe that existence has a common and single meaning among all. This intuition is also supported by the fact that "nothingness" or "non-existence," which is the opposite of existence, has a single concept.
On the other hand, as Rescher rightly asserts, it is clear that genuine existences are not realized in a strictly uniform manner. This, however, does not imply that the elementary notion of existence is not common among all these different cases. The fact that there are different kinds and degrees of color (white, black, whiter, blackest, etc.) with significant differences does not mean that there is not a single common notion, even though a rough one, of color (likewise, "color" cannot be defined).13 This common notion requires that the instances all have a common element, but does not require that the instances should be entirely similar. For example, John Austin (1970a, p. 27) describes a situation in which a "word may possess connotations which are partly identical and partly different":
When we speak of "healthy exercise", the [word] "healthy" has a connotation which is only partly the same as that which it has in the phrase "a healthy body." … Hence healthinessa, when predicated of an exercise, means "productive or preservative of healthinessb", i.e. of healthiness in the sense in which it is predicated of bodies. Thus "healthinessb" and "healthinessa" have connotations which are partly identical and partly different.
The same situation also holds, Austin says (1970b, p. 71), for the word "exist." So, when we say that "God exists," "Man exists," "Redness exists," "Numbers exist," and "Fictional entities exist," we apply the terms "exist" and "existence" uniformly to all of these things, referring to the element of existence, that is common among them. This common element is what Kenny (1968, p. 92) describes as simply "that attribute which is common to mice and men, dust and angels." So, ontologically, existence is realized differently when it is considered as the "existence of …," or is combined with other notions such as contingency, necessity, potency, and the like. However, when we consider the notion of existence qua existence, we intuitively find that we apply it uniformly to all cases.14
The Existence/Whatness Distinction
The idea of the existence/whatness (traditionally referred to as existence/essence) distinction can be traced back as far as Aristotle15 (An. Po. II, chap. 7, 92b 5–15) and Boethius (Nolan 2001). Ibn Sīnā (1403 H.Q., 3 : 11–14, 61; 1405 H.Q., Al-Ilāhiyyāt, 2 : 346–347; and 1383, Ilāhiyyāt, chaps. 11, 22, 24–25), however, was the first to investigate this issue systematically,16 and it was later developed by other Muslim [End Page 119] thinkers, in particular by Mulla Sadra Shirazi ( 1981, 1 : 243–245).17 While for Aristotle the existence/whatness distinction is not of any great consequence, it is the basis of Ibn Sīnā's most significant metaphysical theories about modalities, universals, substances, God, creation, and other related issues. Indeed,
[t]he distinction between "quiddity" and "existence" is undoubtedly one of the most basic philosophical theses in Islamic thought. Without exaggeration the distinction may be said to constitute the first step in ontologico-metaphysical thinking among Muslims; it provides the very foundation on which is built up the whole structure of Muslim metaphysics.
Ibn Sīnā's distinction also played an important role in Western philosophy. Aquinas (1968) was apparently the first to consider this issue as a major philosophical problem and centered some of his own significant doctrines on this distinction. After him almost all great philosophers, like Duns Scotus, Ockham, Francisco Suárez, Spinoza, Descartes, Leibniz, and Kant, and also some contemporary philosophers have discussed the issue.18
The Conceptual Distinction between Existence and Whatness
To draw the line between whatness and existence is no more than to notice the difference between what a thing is and that it is. For example, one can define what a number or an angle is independent of knowing whether there are any numbers or angles in the world. The reason for this conceptual whatness/existence distinction, as was first argued by Ibn Sīnā (1403 H.Q.) and was later repeated by Kant (A7/B11), is simple and yet convincing: the concept of a whatness qua whatness never contains the concept of existence, so that by breaking up the concept of a whatness into those constituent concepts that have all along been thought in it, we never get the concept of existence. Ibn Sīnā (1403 H.Q., 3 : 13–14) uses the example of a triangle to illustrate this: we can talk about the properties of a triangle without even knowing whether there are in fact any triangles.
Also, the concept of existence does not contain the concept of whatness. Hence, we can know that a thing exists without knowing what kind of thing it is. For example, a scientist might see something through his instrument in his laboratory while he has no idea what it is. Or, most philosophers believe in the existence of "consciousness" whereas they have no clear idea about its whatness. It seems, therefore, to be very clear that we have different ideas and concepts of the notions of existence and whatness.19 After assuming a distinction between the concepts of existence and whatness,20 a question that naturally arises is: is there also a real distinction between the existence and the whatness of a contingent entity?
No Real Distinction between Whatness and Existence
It might be said that there is not only a conceptual but also a real distinction between the whatness and the existence of contingent entities. That is, the whatish concept refers to one objective (metaphysical) aspect, and the concept of existence refers to the other objective (metaphysical) aspect of entities. In other words, [End Page 120] in this scheme, whatness and existence are two distinct things, although, it might be claimed, neither of them can exist without the other (Vallicella 2002, pp. 71–72). Therefore, contingent beings are objective and real compositions of existence and whatness.
The idea of the real existence/whatness distinction seems to be quite radical. Although some contemporary philosophers are apparently advocating a modern version of this idea,21 the idea had indeed very few proponents among ancient and medieval philosophers. Miller (1986, p. 237) attributes the belief in real essence/existence distinction to Ibn Sīnā, Aquinas, and Giles of Rome, although he does not explain what he or those philosophers mean by the real distinction. If he means what I have meant here, then except for Giles of Rome, who explicitly believed in the real existence/whatness distinction,22 Ibn Sīnā (definitely)23 and Aquinas (arguably) did not believe in such an idea.
Ibn Sīnā in the different positions of his writings (e.g., 1405 H.Q., Natural Science, 1 : 27; 1403 H.Q., 1 : 43–44; 1404 [H.Q.] 143; 1383, 3 : 38–39) explicitly asserts that the distinction between whatness and existence is conceptual, not real, such that by considering their concepts we observe that they are two independent concepts so that none of them contains the other. Then, to explain the relation between existence and whatness, Ibn Sīnā has used the word urūḍ, which is a homonymic word in Arabic and Islamic logic and philosophy with different meanings.24
It seems, historically, that Ibn Rūshd (Averroes) was the first to confuse the different meanings of the word urūḍ. He mistakenly attributed to Ibn Sīnā a belief in a real distinction between whatness and existence such that existence as an ordinary accident, like whiteness, is added to the whatness,25 and severely attacked Ibn Sīnā on this basis (Ibn Rūshd 1998 pp. 336–338).
This misinterpretation of Ibn Sīnā's writings was a historical mistake, which was later repeated by some Western philosophers.26 Thomas Aquinas, for example, in his Commentary on the Metaphysics (Bk 4, Lsn 2, Sct. 556, p. 223 n. 556) (1961), explicitly attributes the view to Ibn Sīnā according to which in every contingent entity the existence (esse) of that thing and its essence differ. Therefore, both unity and existence signify something added to the whatness of a thing. To this Aquinas (Bk 4, Lsn 2, Sct. 558, p. 224 n. 558) responds that "even though a thing's existence is other than its essence, it should not be understood to be something added to its essence after the manner of an accident.…" So, on the grounds of such a misinterpretation, which can easily lead to a misunderstanding of Ibn Sīnā's entire philosophical system, even such distinguished contemporary scholars as Goichon and Gilson have accused Ibn Sīnā of upholding a "fantastic theory," according to which the philosopher, in Goichon's words (Rahman 1958, p. 3), "did not give existence a higher place than other accidents."27
The fact of the matter, however, is that Ibn Sīnā in his Al-Ta'lighāt admits that existence, like an accident, is added to whatness,28 but sharply distinguishes it from normal accidents. According to Ibn Sīnā, existence is such a unique and very "particular accident" that "its behavior is opposite to that all other accidents": [End Page 121]
The "existence" of all "accidents" in themselves is their "existence for their substrata," except only one "accident," which is "existence." This difference is due to the fact that all other "accidents" in order to become existent, need each a substratum [which is already existent by itself] while "existence" does not require any "existence" in order to become existent. Thus it is not proper to say that its "existence" [i.e., the "existence" of this particular "accident" called "existence"] in a substratum is its very "existence," meaning thereby that "existence" has "existence" [other than itself] in the same way as [an "accident" like] whiteness has "existence." [That which can properly be said about the "accident"-"existence"] is, on the contrary, that its "existence in a substratum" is the very "existence" of that substratum. As for every "accident" other than "existence," its "existence in a substratum" is the "existence" of that "accident." (quoted by Mulla-Sadra in the Mashā'ir, p. 34, § 83; translated by Izutsu
So, according to Ibn Sīnā, first, whatness and existence are conceptually distinct, and second, in an existent whatness, its existence should be considered as a very peculiar accident that has been added to it, that is, "become wedded to it from "outside" itself" (Nasr 1989, p. 414). In this case, as the last sentence of the quotation above indicates, the addition of existence to a whatness is nothing but the actualization of the whatness.
It seems that those philosophers, in both the East and the West, who have objected to Ibn Sīnā have ignored these points. Fakhr al-Din al-Rāzi (d. 1209), for example, claims that if existence as an accident is added to a whatness, then prior to this addition and without the "occurrence" of "existence," the whatness should "exist" in one form or another, which is absurd.29 However, Nasir al-Din al-Tūsi's decisive answer to Rāzi in defense of the Avicennian position explains the true meaning of the "accidentality" of existence:
Rāzi's argument is based on his conception that "quiddity" previous to its "existence" has [from the very beginning] a kind of subsistence in the external world, and that, then [at the second stage] "existence" comes to inhere in it. But this conception itself is mistaken. For the being [or subsistence] of a "quiddity" is its "existence." … Thus the qualification of "quiddity" by "existence" is an event that occurs only in the intellect. It is different from the qualification of a body by whiteness. For it is not the case that "quiddity" has a separate "existence" and the attribute that "occurs" to it and which is called "existence" has another [separate] "existence," in such a way that the two become united just as a recipient and a thing received become united. Nay, a "quiddity" when it is, that being itself is its "existence." (Nasir al-Din al-Tūsi, Commentary on Avicenna's Al-Ishārāt wa al-Tanbihāt; translated by Izutsu
The fact of the matter is that none of the great Muslim philosophers believe in the real distinction between existence and whatness. They believed that the whatness and existence of a contingent entity are identical in reality and distinguished only within our thought by means of reason.31 Human beings draw distinctions in thought that do not obtain in reality. In reality, the whatness of a contingent entity along with all properties just is its existence.32
Similarly, in the West, most great philosophers including Aquinas (1968), Suárez (1947), Descartes (1984–1991, 3 : 280), and Gassendi (Descartes 1984–1991, 2 : 224) [End Page 122] took the same position.33 Descartes explains that we regard a single thing in different abstract ways; we can regard a thing as existing, or we can abstract from its existence and attend to its other aspects. In other words, we can focus on the whatness of a thing without considering any kind of existence or non-existence. In doing so, we have distinguished the existence of a substance from its essence within our thought. Descartes (1984–1991, 3 : 280) is keen to emphasize that such a distinction is purely conceptual, and that they are "in no way distinct" outside thought. Gassendi (Descartes 1984–1991, 2 : 224) takes the same view very clearly: "Real separation [between essence and existence] is impossible no matter how much the mind may separate them or think of them apart from each other."
Existence and Whatish Properties
It seems obvious that existence is not a whatish property, a property pertaining to the whatness of a thing (in, say, the way that being red is a property of an apple).34 So, existence is never categorized under whatish categories, and existing things are not a kind of thing. Existence is not a genus, a differentia, an essence, or an accident. Nor is it substance or a bare particular of entities. If existence were one of these notions, then a thing or a part of a thing could be (exist) without its existence, which is absurd.
Also, existence is not some kind of ontological element additional to objects. What existence posits is not an accidental quality added to a somehow pre-existing whatness. Rather, it is a metaphysically necessary condition for the instantiation of any whatish properties in the following sense: it is not possible for a non-existent thing to instantiate any property because there is nothing to which, so to speak, a property can stick. In other words, X's existing, as an individual, is a logically prior condition of its instantiating any property whatsoever.
Kant confirms this thesis in his own way. In his statements of the refutation both in the Beweisgrund (1763, 2 : 76) and in the Critique (A599/B627), Kant completes his negative thesis that existence is not a real predicate with a positive doctrine about existence: existence is the absolute positing of the subject-concept with all the predicates that are thought in it. Similarly, Gassendi (Descartes 1984–1991, 2 : 224–225) asserts: "if a thing lacks existence, we do not say it is imperfect, or deprived of a perfection, but say instead that it is nothing at all." So, even if this highly controversial claim is accepted that existence is a universal property, a property of everything,35 it is clear that neither is existence a whatish property36 nor does it add anything to whatness as whatness, yet it lets whatness represent itself.
Let us summarize what I have shown so far:
I. Existence and whatness are conceptually distinct; that is, neither does the concept of existence contain the concept of whatness nor does the concept of whatness contain the concept of existence.
II. The concept of existence is not a whatish concept, and adds absolutely nothing to whatish concepts as whatish concepts.
III. The meaning of "exists," that is, the concept of existence, is single and similar in all cases. [End Page 123]
IV. Therefore, what make things conceptually distinct are whatish concepts.
V. The notion of existence qua existence is applied uniformly to all cases.
VI. There is not a real separation between the whatness and existence of entities whatsoever. Existence and whatness are therefore identical in reality, and externally are indistinguishable.
VII. Existence is not a whatish property, and adds absolutely nothing to whatness as whatness.
Now, using the conclusions above as Axioms,37 I shall try to show that there is no inconsistency in Kant's position, and the problems of both contradiction and insignificance can be solved satisfactorily. First I will quote Kant's claims in the form of principles, and then explain them.
Principle 1. All existential propositions are synthetic (A597/B625). Existential propositions are synthetic since their subject represents only a whatness to which existence is predicated. In this case, the subject does not contain the notion of existence, and hence the predicate is added to it. Axiom I above provides such a condition for existential propositions.
It seems that Hume ( 1989 : 66–67) disagrees here with Kant:
[S]ince we never remember any idea or impression without attributing existence to it, the idea of existence must either be deriv'd from a distinct impression … or must be the very same with the idea of the perception or object.… [But] the idea of existence is not deriv'd from any particular impression. The idea of existence, then, is the very same with the idea of what we conceive to be existent.38
This doctrine is very similar to what some philosophers have used to argue against the Ontological Argument. George Nakhnikian and Wesley Salmon (1957, p. 541), for example, claim that "Every conception involves the predicate 'exists'."
Shaffer (1962, p. 316) argues that Hume "confuses the existence of the conception with the existence of what is conceived, the realitas formalis of the idea with its realitas objectiva." If by "it" in "attributing existence to it" Hume refers to "any idea or impression," then he indeed confuses the "existence of ideas" with the "idea of existence."39
Moreover, Hume's view, if by "existent" he means something whose existence is "actual," does not explain a lot of cases in which we conceive non-existents, like fictional entities. It is true that, in these cases, anyone who claims that a fictional entity (actually) exists contradicts himself. However, it is also true that this does not prevent one from conceiving them as non-existent.
Furthermore, if a whatness contained existence, this would require that simple existential propositions, like "Man exists" and "Man does not exist," turn into tautologies and contradictions, respectively. For, the former would mean "The existent man (the man who exists) exists" and the latter would mean "The existent man (the man who exists) does not exist." However, as C. D. Broad (1953, pp. 182–183) asserts: [End Page 124]
But it is plain that Dragons do not exist is not self-contradictory. It is not only logically possible but is almost certainly true.… But it is plain that Cats exist is not a mere platitude. It is a substantial proposition which might very well be doubted by a person who had never seen a cat.
Hume's idea concerning existence, therefore, is not only strongly counterintuitive but is indeed false. Surprisingly, Hume elsewhere ( 1978, p. 164) rejects explicitly what has been claimed here by saying that "Whatever is may not be. No negation of a fact can involve a contradiction." In sum, existential propositions are synthetic since the predicate (existence) has not already been thought in the subject (a whatness), and hence is added to it.
Principle 2. "Being" … is not a concept of something which could be added to the concept of a thing (A598/B626; my emphasis). At first it seems that Principles 1 and 2 are flatly contradictory. However, by appealing to Axioms II, VI, and VII, Principle 2 can be interpreted in such a way that removes the apparent inconsistency, and so the problem of contradiction in Kant's position is dissolved. There are two ways to do this. According to the first, Kant in Principle 2 is saying something, not about objects but about the elements of the existential propositions, which can be assumed the concepts representing objects. In this case, by replacing "a concept of something" and "the concept of a thing" with "a whatish concept," it becomes clear that Kant in Principle 2 says precisely what Axiom II states: "being" or "existence" is not a whatish concept, and although it is added to whatish concepts in existential propositions, it is not doing so as a whatish concept.
According to the second interpretation, in Principle 2 Kant's discussion is not about the concepts of existence and whatness, but rather is about genuine existence and existents themselves. Kant is indeed emphasizing the core idea of the proponents of a merely conceptual whatness/existence distinction (Axioms VI and VII), according to which existence and whatness are identical in reality, and existence adds absolutely nothing to whatness as whatness. Hume ( 1989, pp. 66–67) takes a position very much like this when he says:40 "To reflect on anything simply, and to reflect on it as existent, are nothing different from each other." This point is stated more clearly in the following words of Kant (A600/B628):
By whatever and by however many predicates we may think a thing … we do not make the least addition to the thing when we further declare that this thing is.… If we think in a thing every feature of reality except one, the missing reality is not added by my saying that this defective thing exists.
In effect, as we explained earlier, Kant believes that existence is the absolute positing of the subject-concept with all the predicates that are thought in it. Thus, existence has the peculiarity of not adding a determination to the subject-concept but positing the latter as a whole, with whatever predicates it has in it, as having a real object outside. So, since the function of "exists" is to posit the subject-concept as a whole, it must not be the case that "exists" itself adds something to the subject-concept. [End Page 125]
Interestingly, the same apparent contradiction can be seen in Descartes' position. On the one hand, as we showed already, Descartes explicitly believes in the conceptual distinction between whatness and existence. On the other hand, he says (Descartes 1984–1991, 2 : 117): "Existence is contained in the idea or concept of every single thing, since we cannot conceive of anything except as existing." Closer attention, however, shows that there is no contradiction. By saying that we cannot conceive of anything except as existing, Descartes means that every clear and distinct idea implies that there is an existent to which this idea applies. In other words, for Descartes one does not have to build existence into an object of which we have a clear and distinct idea; existence of an object (and not the concept of existence) is assumed if we have a clear and distinct idea of the object. Therefore, Descartes in one place speaks about the relation (i.e., distinction) between two concepts, that is, the concept of existence and the concept of whatness; whereas, in another place, he investigates the relation between having a clear and distinct idea and having existence (and not the concept of existence). Whether it is valid to leap from having a clear and distinct idea to the real existence of its extension is another matter, with which Kant explicitly disagrees: "In whatever manner the understanding may have arrived at a concept, the existence of its object is never, by any process of analysis, discoverable within it" (A639/B667).
In sum, in existential propositions the concept of existence (the predicate) is added to the concept of the subject, but not as a whatish concept. Also, if the propositions are considered as representing what they apply to, the predicate (existence) adds nothing to the subject (whatness) since, externally, they are not two distinct things that are added to each other. Kant clearly confirms these points in the following (A234/B287; my italics):
The principles of modality are not, however, objectively synthetic. For the predicates of possibility, actuality,41 and necessity do not in the least enlarge the concept of which they are affirmed, adding something to the representation of the object. But since they are nonetheless synthetic, they are so subjectively only, that is, they add to the concept of a thing (of something real), of which otherwise they say nothing.42
Principle 3. If "exists" were added to the concept of the subject we could not say that the exact object of my concept exists (A600/B628). Principle 3 does not mean Kant denies that a reality corresponding to our concepts has always some features that are not represented by our concepts—contrary to what Bennett (1974, p. 230) sees as a problematic result of Kant's view. Also, as Allen W. Wood (1978, p. 105) rightly points out, Kant does not deny that stating "x exists" can provide some new information about x. Rather, Kant says that if the concept of existence (predicate) were added as a whatish concept to the concept of the subject, then the concept of the subject would change. However, since in reality there is not any duality and distinction between what the subject applies to and what the predicate applies to (Axioms VI and VII), and consequently no addition of the latter to the former, the reality then would not be the true extension of the concepts. Therefore, we could not say that the exact object of our concept exists. [End Page 126]
For example, in the existential proposition "Gold exists" there is only one extension for the subject and the predicate. In other words, in reality there is not any duality between "being gold" and its "existence." So, if it were assumed that the predicate adds something to the subject, then we would find a combined concept in our proposition (i.e., "gold + existence") that does not have any extension in the real world, and hence is not its true representative.
Concerning non-existential propositions in which whatish concepts (e.g., "gold" and "yellow") are added to each other, however, there is not such a problem. For, each whatish concept (the subject and the predicate) has its distinct extension such that each one picks out a distinct whatness. Therefore, adding a whatish concept (predicate) to the subject would represent having distinctively the referent of the concept, that is, a whatish property, by the object. So, contrary to the case of existence, adding the predicate to the subject would not imply that our concept does not have an exact referent (object), since such an adding occurs in reality too. Counterfactual situations can show the difference between whatish properties and existence, and that is why, contrary to existence, it is reasonable to assume that each whatish concept has its distinct referent.
Consider the following example. P1: "There is x such that (A) x is gold, (B) x's atomic number is 79, and (C) x is yellow," in which "there is x" means "x exists."43 In our world, A, B, C, and hence P1 are true. Nonetheless, we can consider seven counterfactual situations in which at least one of A, B, or C is false, and hence P1 is replaced by other propositions that can be true in other possible worlds. For example, consider P2: "There is x such that x is gold, x's atomic number is 79, and x is not yellow" (in formal language: A&B&¬C). Or P3: "There is x such that x is gold, x's atomic number is not 79, and x is not yellow"44 (or, in formal language, A&¬B&¬C). And so on.
However, if the predicate "exists" had the same application as non-existential predicates, then we would have one more proposition, D (i.e., "there is x" or "x exists"), and the number of counterfactual situations would become fifteen. However, from Kant's point of view, it would be entirely meaningless if it is said that we can assume a counterfactual situation in which "There is not x and yet x is gold, x's atomic number is 79, and x is yellow."45 The reason is that there is not any duality whatsoever between the extension of being x, as a yellow gold with atomic number 79, and having existence. Hence, the concept of "existence" refers to nothing else but to what the concept of "x" refers to, that is, x.
What has been claimed here can also be drawn indirectly from Kant's own view. We showed earlier that, according to Kant, the function of "exists" is to posit the subject-concept as a whole. So, "exists" does not add something to the subject-concept; otherwise not the original subject-concept but a new one will be posited. The ordinary (non-modal) predicates do not have this problem, because they do not posit the subject-concept absolutely, but only relatively by adding further determinations to it (in synthetic propositions). Shaffer's argument against Kant (the problem of insignificance) is therefore inconclusive since it is based on a false analogy between the predicate "exists" and other predicates. [End Page 127]
Principle 4. A determining predicate "is a predicate that is added to the concept of the subject and enlarges it" (A598/B626). By enlarging a concept Kant does not mean that the concept of the subject is revised or changed if the concept of the (real) predicate is added to it—as Shaffer (1962, pp. 309–310) has claimed. Although change is an inseparable aspect of physical entities, it seems, however, that concepts and, generally, mental entities are immutable. If I have concept A, and then I find something new concerning A, it is not the case that A somehow changes to, for instance, A*. Rather, I find a new concept A* while I still have the previous concept A. If it were not so, I would lose A, and it would be impossible that I have both concepts A and A* simultaneously. Whereas it is obvious that as long as I do not forget A, I have it in my mind, no matter how many new ideas I find concerning A. Concepts, therefore, are not transformed into other concepts.
I therefore take it for granted that by enlarging a concept Kant means replacing the old concept with a new one. For example, suppose by the concept of gold we mean "a yellow metal that is not oxidized." Then, if it is discovered that the atomic number of gold is 79, the previous concept is replaced by a new one in which "having atomic number 79" has been added to the previous one. On this line of reasoning, Kant's claim is that "exists" is not a determinate predicate because if we discover that gold exists, it does not produce a new (whatish) concept. In Kant's words:
Were we dealing with an object … through its existence it is thought as belonging to the context of experience as a whole. In being thus connected with the content of experience as a whole, the concept of the object is not, however, in the least enlarged.(B628–9; my emphasis in last line)
The predicate "exists," therefore, only shows that our initial concept has an extension, whereas whatish concepts if added to the subject produce new concepts, and hence they are determining predicates.
Principle 5. A real predicate is a predicate "which determines a thing," and "Being" is obviously not a real predicate (A598/B626). If, according to Principle 4, "exists" is not a determining predicate, and if a real predicate is a predicate "which determines a thing," then "exists" is not a real predicate. By this principle Kant, however, does not mean that "exists" is not a predicate at all. Rather, he only means that "exists" is not a predicate that represents the whatish properties of entities. In other words, since only whatish properties are the properties that determine what a thing is, a "real predicate," from Kant's point of view, can only be a predicate that represents the whatness of a thing.46 This interpretation is also supported by focusing on the nature of the notions of existence and whatness. If, as we explained already, the word "exists" is used with one meaning in all cases, then only whatish concepts can make things conceptually distinct, and hence only they can be determining predicates47 (Axioms III, IV, and V). Therefore, in Kant's terminology, only the predicates that refer to the whatnesses of things, and not their existences, can be real or determining predicates. [End Page 128]
We tried to show that the "problems of contradiction and insignificance" can find satisfactory solutions if Kant's views on existence are interpreted not on the basis of the post-Fregean approaches but on the basis of the same assumptions of his preceding philosophers, the assumptions whose background can be traced back to Descartes, Gassendi, Aquinas, and Ibn Sīnā and their thesis concerning essence, existence, and their relation.
Regarding the points that we explained, it is worth mentioning that Kant's main objection to the ontological argument is not based on denying his preceding views about existence. Rather, he is trying to show that since such opinions are plausible the ontological argument is invalid.
For example, one main assumption of the ontological argument is that some existential claims can be deduced from merely their conceptual contents. In other words, in some cases we can infer the existence of an entity by merely considering its whatness. Kant's, and before him Gassendi's, objection is that this assumption is in direct conflict with Axiom I, with which Descartes totally agrees. Gassendi explains this point nicely (Descartes 1984–1991, 2 : 225):
You [Descartes] say that existence is distinct from essence in the case of all other things, but not in the case of God. But how, may I ask, are we to distinguish the essence of Plato from his existence, except merely in our thought? … Surely in the case of God the distinction between essence and existence is of just this kind: the distinction occurs in our thought.… [F]rom the fact that you think of God as existing it does not follow that he exists.
I do not discuss whether Kant's objections to the ontological argument are valid or not. Nor do I claim that Kant's position on the ontological argument, in all its aspects, is entirely and explicitly based on assumptions that are common between him and the philosophers preceding him. Rather, what I am saying is that at least in the case of the issue of existence it seems that we face a dilemma: either we leave the problems of Kant's opinion concerning existence unsolved or we view Kant's position on the basis of its philosophical background; in this case, most controversial pieces of Kant's puzzling approach fall into place.48 It would be reasonable that the second horn of the dilemma is chosen.
1. See, in particular, what Kant has said in his more substantive account of modal concepts in sections of the Critique of Pure Reason such as the "Metaphysical Deduction" (B100–102), "Schematism" (B184/A145), and the "Postulates" (A218–235 / B265–287). [End Page 129]
2. It is worth mentioning that Kant's notion of existence as not a real predicate that could be contained in the real contents of the concepts of things reflects on the scholastic distinction between essence and existence also voiced by Heidegger (1988) in his Basic Problems of Phenomenology.
3. Kant gives no clear reason why all existential propositions are synthetic, and indeed begs the question in this claim against the ontologists, who believe the proposition "God exists," in which God means "the necessary being, whose existence is necessary," is analytic. Through the interpretation that will be explained later, however, it will be shown why all existential propositions are synthetic.
4. See the Cambridge edition, "Vienna Logic," p. 352; "Jäsche Logic," pp. 589, 597.
7. In a forthcoming paper I try to show that none of these interpretations is conclusive.
8. Since "essence" has found different meanings in philosophy, I prefer to use "whatness" ("māhiyyah" or "quiddity") instead of "essence."
9. See, e.g., Descartes 1984–1991, 2 : 224–226. Forgie (2007, p. 523) claims that Kant "is not echoing claims made over a century earlier by Gassendi." For Gassendi considers existence as a first-level predicate, whereas, according to Kant, existence is a second-level predicate. The scope of this essay does not allow discussing this issue in detail (for criticism of Forgie's view see Vallicella 2004). I only mention that the interpretation of Gassendi's and Kant's opinion on the basis of the post-Fregean logical notions is misleading, although some consequences of this interpretation are very similar to the consequences of the approach of this essay.
10. It has been argued that some notions such as necessity, unity, and particularity are indeed the properties of existence, not whatness (if property is used in its liberal sense, according to which a property is anything that somehow can be attributed to entities—what Lewis [1986, pp. 59–61] calls "abundant properties"). In other words, they describe the state or the mode of the existence of entities, and hence are not whatish properties. In this essay, however, I shall not discuss this topic in detail, and only notify this point briefly wherever it is necessary.
11. Therefore, the whatness of an object is a more general notion than that of its essence, which usually refers only to the essential properties.
12. By the several modes of existence Rescher (1978, p. 64) means cases like the actual existence of particulars, the OF-correlative existence of ground-level [End Page 130] universals (actual or possible), the OF-correlative existence of higher-level universals, the OF-correlative existence of the dispositional properties of individuals, etc.
13. Having the same common element in spite of having unlike instances is the property of notions that admit graduation and degree. Consider, for example, the notion of force. Both weak forces and strong forces are the same in being force, and at the same time they differ with respect to their weakness and strength in forceness. According to a traditional belief in philosophy that goes back at least to Plato, existence or being is such a notion and admits of degree. That is, existences are different in degree and grade, not in kind, so that different existents are indeed the different levels of the same notion in proportion to their shares from that notion. This idea is explicitly acknowledged by many Muslim philosophers (see, e.g., Ibn Sīnā 1992, p. 41, and  2004, 2 : 37–38; Sohravardi 1977, 2 : 119–120; and Shirazi ( 1981, 1 : 427–46). Also, Descartes acknowledged such an idea explicitly: "There are various degrees of reality and being; a substance has more reality than an accident" (1984–1991, 2 : 117), and Kant used "this quality of reality to underwrite the claim for the existence of a continuum between reality and negation" (Caygill 1995, p. 345).
14. Considering the points mentioned above, perhaps it is better that, by using Aquinas' (1947, pp. 28 [P1, Q4, A3], 86 [P1, Q13, A5]) idea of analogical predication, we say that "existence" is an analogous term. If a univocal term, such as "tiger," is one that is applied to a number of similar things in an identical sense, and an equivocal term, such as "bank," is one that is applied to a number of things in entirely different meanings, an analogous term is one that applies to unlike things, partly for the same and partly for a different reason.
15. "… what a man is and that a man is are different." Goichon (1937, p. 132) has pointed to passages of similar content in Plato's Phaedo and Republic. For a criticism of this view, see Morewedge 1972, p. 426 n. 6.
17. According to Jean Jolivet (Wisnovsky 2000, p. 181), the origins of Avicenna's distinction between essence and existence lay not in ancient Greek philosophy but, indeed, in early Islamic theology (kalam) and the Muslim theologians' debates over how the terms "thing" and "existent" relate to each other. Moreover, some contemporary scholars, e.g., Moody (1965, p. 263), believe that Aristotle opposes this distinction in its Avicennian formulation.
20. It should be noted that the existence/whatness conceptual distinction, as Nasir al-Din al-Tūsi (d. 1274) asserts, should not be taken as meaning that "quiddity" in the intellect is separated from "existence," because "being in the intellect" is itself a kind of "existence," namely, "rational [mental] existence," just as "being in the external world" is "external existence." … [This] must be understood in the sense that the intellect is of such a nature that it can observe "quiddity" alone without considering its "existence." Not considering something is not the same as considering it to be non-existent. (from Tūsi's Sharh al-Ishārāt, trans. Izutsu [1971, p. 121])
24. "Urūḍ a to b" might mean that a is an accident (in contrast to the substance) of b, or it might mean that a does not belong to the nature of b and is added to it from "outside" itself. Traditionally, the first meaning has been used and discussed in the "Chapter of Maghulāt (Categories)" in Islamic (Arabic) Logic, whereas the second meaning belongs to the "Chapter of Eisagoge." It will be shown that in the issue at hand, Ibn Sīnā has meant the latter, not the former.
25. For an excellent summary of whether existence is an accident and the reason for the misunderstanding that followed Ibn Sīnā's assertion of the "accidentality" of existence, see Izutsu 1971, pp. 118–129.
28. According to Rahman (1981, p. 13), the true meaning of the famous metaphysical dictum "Existence is accidental to essence" is that "the contingent is never rid of its contingency" and never becomes self-necessary like God.
30. In fact, according to Ibn Sīnā (1405 H.Q., Al-Ilāhiyyāt, 1 : 29–37), "thing" and essence are, in a way, posterior to "existent" and existence. Also, in Al-Ishārāt wa al-Tanbihāt, he explains: "It is not possible that the attribute called 'existence' be caused in a thing by its essence, which is quite distinct from its existence or any other attribute. For the cause precedes the effect ontologically, but nothing is prior to existence." (Quoted from Goodman's Avicenna [1992, p. 78].)
31. According to Sadra's ( 1981, 1 : 38–44, 54–67) thesis of the fundamentality of existence, whatnesses neither exist in themselves nor do they have any [End Page 132] actuality, so that whatness without existence is indeed nothing. That is, to be X is to be a special grade of existence. Therefore, the lack of existence is the same as the lack of X. Existence, therefore, is the most fundamental reality of all entities. Whatnesses are the limitation of existence, as a point is the limitation of a line.
32. It is worth mentioning in passing that this idea does not imply that we can dispense with existence in favor of the notion of "something"—as D. C. Williams (1962, p. 763) claims. Nor does it imply that the whatness of an entity entails its existence such that all entities become necessary beings—as Vallicella (2002, pp. 71–72) argues.
33. Although many Thomists hold that there is a real existence/essence distinction, Aquinas himself, as a follower of Ibn Sīnā on this issue, can be interpreted as a denier of the real distinction. It should also be noticed that there are some differences between the views of, e.g., Aquinas and Suárez, and Descartes and Gassendi. These philosophers, however, all agree that there is not a real distinction between essence and existence.
34. Some advocates of the Frege-Russellian approach also believe that existence cannot be a property of individuals, although not for the reasons that are suggested here. Russell (1971, p. 252), for example, wrote: "there is a vast amount of philosophy that rests upon the notion that existence is … a property that you can attribute to things, and that the things that exist have the property of existence and the things that do not exist do not. That is rubbish."
35. This idea was proposed by Nakhnikian and Salmon (1957). However, the fact of the matter is that existence is ontologically too basic to be a property. For more discussion see Vallicella 2002, pp. 42–55.
37. As our discussion shows, these "axioms" might also be drawn indirectly from Kant's own theses about existence. In particular, "Axioms" I, II, IV, and VII are other ways of expressing Kant's thesis that existence is not a real predicate, and III and V are presupposed by Kant, since he is engaging in a discussion of what "existence" means in general.
38. It should be noted that Hume is discussing the relation between two ideas, i.e., the idea of existence and the idea of existent, not the relation between genuine existence and existents.
39. The contradictory positions that Hume takes concerning negative existential judgments—in one place he interprets them as judgments in which the idea of the object is conjoined with the idea of non-existence ( 1989, p. 15), and in another place he flatly rejects such an interpretation (ibid., p. 96 n.)—shows that he has no correct account of this issue. [End Page 133]
40. This is the case, provided that Hume here speaks about genuine existence and existent, and not their concepts.
41. Although Kant does not mention actuality when he describes his categories of modality (B106), he uses existence and actuality interchangeably here and also in his criticism against the ontological argument.
42. This indeed shows that Kant has another, what might be called a "cognitive," conception of syntheticity, according to which a relation with a cognitive faculty is added to the subject-concept. In this type of synthetic proposition the subject-concept is related to a cognitive faculty (understanding, power of judgment, or reason) of the epistemic/cognitive subject. It is clearly on the cognitive or subjective account of syntheticity that Kant calls not only existential but also all modal propositions synthetic.
43. It should be noted that for Kant "being" and "existence" are synonymous.
44. Even if, according to our current knowledge, having atomic number 79 is an essential property for being gold, we can imagine a counterfactual situation in which it is discovered that scientists were wrong, and the atomic number of gold is, say, 80.
45. Indeed, from Kant's point of view, this is a contradiction, for it is equal to saying that "There is x such that x does not exist, x is gold, x's atomic number is 79, and x is yellow"—or, in formal language, D&¬D&A&B&C.
46. It is worth noting that "real" derives from "res," which means thing, and is interchangeable with quid, or what. Also, reality appears under the rubric "Quality" in Kant's table of categories, while existence is booked under "Modality" (B106).
47. Hintikka (1981, p. 141) takes the very same view when he says, "Kant's injunction that existence is not a real predicate might perhaps be interpreted as saying merely that it must not be used in the definition of anything," and in the footnote he adds that Kant here returns to the view of Aristotle, "who similarly believed that 'being is not the essence of anything.'"
48. There is, of course, another option, that is, to interpret Kant's view on the basis of the post-Fregean ideas concerning existence. However, in a paper that I hope to have published shortly, I have tried to show that none of the post-Fregean interpretations of Kant have provided a satisfactory solution for Kant's problems so far.