Replies to Michael Kremer

Since Michael so neatly summarized his objections in the form of three questions, all I need to do now is to answer these questions.

First, is existence really not essential by my definition? My answer is that it is neither universally essential, nor universally non-essential, and not by the definition Michael considers. The reason for this answer is that the definition Michael considers does not express the sense in which I claim (as I believe, together with Aquinas) that ‘exists’ is not an essential predicate of creatures, while it is an essential predicate of God.

Let me clarify. I introduced the notion of a substantial predicate by means of the following definition:

(SP) P is a substantial predicate if and only if SGT(‘exists’)(SGT(P)(u)(t))(t)=SGT(‘exists’)(u)(t)

Michael correctly interprets this definition as saying that a predicate P is substantial if and only if the existence of the significate of P in a thing u is the existence of u. But in this definition the predicate ‘exists’—since at this stage of my argument it is already treated as a distinguished logical predicate—does not fall within the range of the metavariable P, which ranges only over the "ordinary" non-logical predicates of the system. Therefore, Michael’s next step in his argument, substituting ‘exists’ for P, is not licensed by this definition. In fact, as can be seen from the subsequent course of the paper, the role of this definition is only to allow me to define the semantic values (the significata and supposita) of the term ‘essence’ in this system, as being the same as the significata of the substantial predicates thus defined. It is this introduction of the term ‘essence’, then, that further enables me to discuss, perhaps too briefly, the question of whether ‘exists’ is an essential predicate in the stronger sense that it would signify the essence of each and every thing. So the stronger sense of ‘essential’ I propose is not the sense of ‘substantial’ provided by this definition. The stronger sense of ‘essential’ I propose is that a predicate is essential to a thing in this strong sense, if and only if it signifies the essence of the thing in question. But then, given this stronger sense of ‘essential’, my Thomistic claim is that in this sense ‘exists’ is neither universally essential nor universally non-essential, but it is essential only to God, and non-essential to everything else. However, this is a metaphysical claim, the truth or falsity of which is not predetermined by the semantic definitions. The semantic definitions are needed only in order that we understand the claim properly, so as to be able to judge competently the soundness of the arguments for or against it. In any case, it seems that I have to add this clarification in some form to the final version of the paper, and I am grateful to Michael for pointing out this need by asking this question.

Michael’s second question is whether contemporary essentialism is necessarily so impoverished as I claim it to be, or is it only my representation of it which is impoverished. My answer is that the standard apparatus of possible worlds semantics is just as impoverished as I claim it to be, while, of course, this fact does not exclude the possibility of its "enrichment" in any number of ways. However, the further problem I have with the "enrichments" one can encounter in the contemporary literature is that although they are indeed enrichments in many respects, they do not remedy a specific kind of metaphysical poverty, namely, the lack of "the required expressive devices systematically connecting the semantics of concrete and abstract terms to the semantics and metaphysics of being", that is, the specific kind of poverty I characterized in these terms in the part of the sentence which Michael omitted from his quotation.

The particular problem which Michael mentions as solved by Gupta within the possible worlds account, namely, that existence is one of the "trivial" essential predicates in the standard possible worlds account, simply brings out the point that the standard possible worlds machinery does not automatically yield an adequate definition of the term ‘essential’ (as it does for some intuitive uses of ‘necessary’). I certainly did not regard or present this problem as the single, insurmountable difficulty which the possible worlds semantics account faces, which cannot possibly be solved with its own resources, and which therefore would be the best justification for my own project, which alone would be capable of solving this problem. In fact, in his own very apt summary, Michael does not have such a narrow view of my project. As he writes: "Gyula argues that the more traditional metaphysical framework deserves reconsideration, both because it can help us with problems arising from the contemporary approach, and because it possesses greater expressive power than the contemporary approach. He presents a fragmentary formal semantics for the traditional approach, and argues that this semantics enables us to see how the problems of contemporary essentialism can be avoided while at the same time other properly metaphysical issues, which are unapproachable from within the contemporary model due to its expressive weakness, become available for investigation."

In view of this characterization of the project of the paper, which I find to be correct, I don’t see why I could not acknowledge Gupta’s clever tricks within the possible worlds account, and at the same time claim that even with those tricks the possible worlds account is incapable of expressing such fundamental metaphysical claims as for example Aquinas’s thesis of the real distinction between essence and existence in the creatures, precisely because Gupta’s tricks and the framework in which they work are fundamentally different from the framework in which Aquinas’s claim is meaningful, while Aquinas’s framework solves all the problems for which Gupta’s tricks and their likes were devised, and some more.

[So, to put my position somewhat more sharply than in the paper, my first complaint against possible worlds "essentialism" is that although I can appreciate the particular solutions the possible worlds account can provide to the particular problems it generates, possible worlds "essentialism", even with all the clever fixes that I have seen, just keeps talking past the tradition whose name it quite inappropriately bears.

My second complaint is that, as a result, possible worlds "essentialism" is bound to keep reinventing the wheel, indeed, in several new, revolutionary, polygonal designs, and so no wonder that even those who deal and wheel with the next new design keep wondering just why it doesn’t roll as smoothly on every road as it should.

Accordingly, sticking with this metaphor, my paper says to these people the following: "try to put aside for a while the wonderful precision ruler with which you so aptly draw your clever polygons, and try this simple old device, called the compass, and see whether you can design something better with that".]

But enough of this, let us see whether this approach really has the metaphysical advantages I claim it has, despite the doubts Michael advanced against this claim summarized in his third question: do those metaphysical arguments work?

My answer is that they do, and Michael’s objections don’t show that they don’t, because the objections are guilty of ignoratio elenchi.

This is most evident in the case of his first objection. For, indeed, an argument based on an obviously invalid inference is an obvious howler, and the inference Michael refers to is obviously invalid. But my argument is not based on this kind of inference. Instead, the argument works in the following way. Let’s assume that there is a thing that has no substantial predicates. This means that all predicates of the thing besides ‘exists’ are such that the existence of the significata of these predicates is distinct from the existence of the thing, which is, of course, equivalent to the claim that the thing, if it exists, has existence, but no essence. But then it is clearly possible to construct a model in which such a thing is actual, and thus its existence is also actual. But since this existence is not identical with the existence of the significate of any other predicate in the thing, it is possible in the same model that the significata of all other predicates in this thing are not actual. So, contrary to Michael’s imputation, rather than trying to conclude from the possible falsity of any of the thing’s predicates to the possible falsity of all of its predicates (besides ‘exists’), I conclude to the possible falsity of all the thing’s predicates (besides ‘exists’) from the distinctness of the existence of the significata of all its predicates (besides ‘exists’) from its existence. But then, since this conclusion, namely, that the thing exists and it has no true predicates besides ‘exists’, is unacceptable, we have to reject the assumption from which it followed, namely, the assumption that there is a thing that has no substantial predicates. So, we have to concede that every thing has to have some substantial predicates, quod erat demonstrandum.

On the other hand, I think it is also worth pointing out that, although unwittingly, Michael’s objection provides a very nice confirmation of my claim concerning the metaphysical advantage of this approach over the possible worlds approach. For in the possible worlds approach it is indeed only the howler Michael points out that would be available for the purposes of a similar argument. To see this, consider the following. Informally, the gist of the argument is that the assumption that there are no essential predicates would lead to the unacceptable conclusion that there could exist something that would have no true predicates, besides ‘exists’. As we could see, in the traditional account this does indeed follow, because the assumption would mean that the act of existence of the thing would be distinct from the act of existence of any of its predicates (besides ‘exists’), and thus the thing could exist while the significata of any of its predicates would not exist.

In the possible worlds framework, however, the assumption would only mean that no predicate (besides ‘exists’) of anything is a rigid designator. So, this would only mean that any predicate of any thing is such that it is possible for the thing to exist and not to fall within the extension of that that predicate; or, in other words, for any predicate P of any thing u there is a possible world w such that u exists and u is not P in w. But this of course does not imply that there is a possible world w such that for any P of any u it holds that u exists and u is not P in w. So this would indeed be a simple modal howler:

(definition) (x)(P)(ESS(Px) Û N(Ex ® Px))

(assumption) (x)(P)(~ESS(Px)) [Û (x)(P)M(Ex & ~Px) by (definition)]

(howler) (x)(P)M(Ex & ~Px) Þ M(x)(P)(Ex & ~Px)

(metahowler) (u)(P)($ w)[u Î D(w) and u Ï R(P)(w)] Þ
($ w)(u)(P) [u Î D(w) and u Ï R(P)(w)] [where D is the domain-assignment and R is the denotation function]

But then, again, this would be a howler of the possible worlds account.

The charge of question begging in connection with the second argument is similarly based on a mistaken imputation. Michael says: "This argument assumes that for something to be is for it to be "what it is" in a sense answering to Aristotelian essence, so that for a thing to change without ceasing to be "what it is" in this sense is for it to continue to exist." I agree with Michael that if the argument assumed that for something to be is for it to be what it is, then it would be question begging. For this would be just another way of putting the claim that the existence of a thing is the same as the existence of the significate of its predicates that answer the question what it is, that is to say, this would be just another way of putting that things have essences, and, consequently, essential predicates.

However, the premise I actually used in the argument was the following: "... whenever a thing changes, but without ceasing to be what it is, it continues to exist." This premise by no means entails or is entailed by the claim Michael says I assumed in the argument, namely, that for a thing to be is for it to be what it is. All this premise asserts is that when a thing changes in respect of a predicate which does not signify the thing’s essence, then the thing may go on existing. But of course the truth of this claim neither requires nor excludes the possibility that the thing has some essence possibly signified by other predicates. So I may assume this claim without assuming that things have essences, and so it may legitimately be used by an argument intending to prove this conclusion. Therefore, to make this aspect of the argument clear, let me restate it in the following way.

1. For a living thing to live is for it to exist. [True by the meaning of the terms]

2. So, for a living thing to cease to live is for it to cease to exist [from 1, with obvious meaning postulates.]

3. Things have no essences, or equivalently, no predicate of a thing says what the thing is [Assumption to be refuted]

4. So, the predicate ‘living’ does not say what a living thing is [from 3]

5. So, when a thing changes from living to non-living, it does not change in respect of a predicate which says what the thing is [from 4, with obvious meaning postulates]

6. When a thing changes in respect of a predicate which does not say what the thing is the thing can stay in existence [This is common knowledge: we simply know that there are several such predicates, and we know that whenever things change in respect of those predicates, they do not have to cease to be. This premise is totally independent from the question whether things have other predicates which would say what the thing is and whether changing in respect of those predicates would imply the thing’s destruction.]

7. So, when a thing changes from living to non-living it can stay in existence [from 5 and 6]

8. So, for a living thing to cease to live is not for it to cease to exist [from 7, with obvious meaning postulates]

But 8 contradicts 2, therefore, granting all the other, self-evident premises, we have to reject the assumption. This completes the proof of the claim that things have essences, without ever assuming its truth by the assumption of any one of the premises.

Well, I would probably have to lay out also the other arguments in a similar manner in order to defend them against Michael’s sweeping charge of their question-begging character. But unfortunately I did not have sufficient space for that in the paper, nor do I have any time left for it here.