Meinongian Theory Research Articles

Paradoxical Aspects of the Russellian Conception of Existence

In this paper, the authors try to clarify the relations between Meinong’s and Russell's thoughts on the ontological ideas of existence. The Meinongian theory on non-existent objects does not in itself violate the principle of non-contradiction, since the problem that this hypothesis offers to the theory of definite descriptions is not so much a logical problem as an ontological problem. To demonstrate this we will establish what we believe are the two main theses basic to the theory of descriptions: the epistemological thesis and logical thesis.

Fictional Objects

Abstract The article, in the first part, aims at providing a positive answer to the ontological question about fictional objects starting from semantic reflections on the negative existential statements containing terms purporting to refer to such objects and the principle of compositionality. In the second part is proposed a Meinongian account of ficta, showing how it handles the problem of negative existentials without giving up neither the principle of compositionality nor our way of talking about fictional characters in the natural languages. It is also considered a particular kind of Meinongianism, namely Modal Meinongianism which seems to be the best Meinongian theory in order to explore the modal properties of ficta.

Modal Meinongianism, Russell’s Paradox, and the Language/Metalanguage Distinction

The subject of my article is the principle of characterization–the most controversial principle of Meinong’s Theory of Objects. The aim of this text is twofold. First of all, I would like to show that Russell’s well-known objection to Meinong’s Theory of Objects can be reformulated against a new modal interpretation of Meinongianism that is presented mostly by Graham Priest. Secondly, I would like to propose a strategy which gives uncontroversial restriction to the principle of characterization and which makes it possible to avoid Russell’s argument. The strategy is based on the distinction between object language and metalanguage, and it applies to modal Meinongianism as well as to other so-called Meinongian theories. Every so-called Meinongian theory has to face a well-known argument which was presented by Bertrand Russell in “On Denoting” (Russell, 1956).1 The aim of this critique was to point out that Meinong’s Theory of Objects is inconsistent, false, and worthless from a theoretical point of view. This is so–Russell argued–because one of the fundamental assumptions of this theory (the principle of characterization) leads to a consequence which is ridiculous from an ontological perspective. Although nowadays over a hundred years have passed since Russell first published his classic paper in 1905, and thereafter many philosophers have tried to reply to it, Russell’s criticism is still regarded as a serious challenge. Among the plentitude of theories inspired by Meinong’s views, Graham Priest’s theory (called “Noneism” or “modal Meinongianism”) is the newest one. Because of this one can expect it to deliver an interesting reply to Russell’s argument. Moreover, Priest assures us that, in fact, it does. In this 1 Thanks to Arkadiusz Chrudzimski, Graham Priest and Janine Reinert for their comments which helped to improve a previous version of the paper presented here. 64 Modal Meinongianism, Russell’s Paradox, and the Language/Metalanguage Distinction article I would like to put this claim to the test. Nonetheless, the conclusion of my paper is rather positive–Priest’s theory is immune to Russell’s critique and, in fact, older Meinongian theories are as well. This is so because the above-mentioned argument might be taken to be based on a methodological misunderstanding, and its reformulation in terms of Priest’s theory will help me to show why it is invalid. I believe that showing what is wrong with a Russell-style argument against Noneism will allow me to show that this argument was invalid in the first place, i.e. when it was used against Meinong’s original theory. In order to present my thesis I will start with a brief sketch of the main assumptions and claims of the original Theory of Objects and I will show how it was criticized by Bertrand Russell. Further, I will recall how various advocates of Meinong’s theory (especially Priest) have been trying to defend “non-existence” against Russell’s critique. At the end I will sketch a proposal for solving the puzzle of the principle of characterization. 1. Theory of Objects and the Principle of Characterization According to the main and well-known thesis of Meinong’s theory, “there are objects concerning which it is the case that there are no such objects” (Meinong, 1960). This controversial claim seems to be paradoxical, but in fact it expresses a simple idea according to which there are objects which we consider to be non-existent ones. Popular examples of such objects are:1) fictional objects such as Sherlock Holmes or Bilbo Baggins, 2) merely possible objects like a golden mountain, and 3) impossible ones like a round square. In some sense it is true to say that each of these objects does not exist. On the other hand, it seems that we truly ascribe some properties to them, e.g.: “Sherlock Holmes is a detective,” “The golden mountain is not green,” “The round square is round and square.” Because of that the so-called Meinongian claim is sometimes presented in a less paradoxical formulation, i.e. as a claim according to which there are nonexistent objects or that not every object exists. The Theory of Objects is based on three principles: 1) The Principle of Intentionality, according to which every intentional act has its object. This principle applies to all objects–no matter whether this object exists or does not. According to this principle it is argued that we can think, worship, or be afraid of, for example, George Bush as well as Sherlock Holmes or Bilbo Baggins. 2) The Principle of Independence states that the nature (Sosein) of an object is independent of its ontological status (Sein). In other words,

Non-Alethic Meinongian Logic DOI:10.5007/1808-1711.2010v14n1p99

The purpose of this work is to provide an answer to two fundamental questions: 1) Can a non-alethic logic be a Meinongian logic? And consequently 2) Can a non-alethic logic be an adequate logic for a Meinongian theory of objects? Using the results of da Costa (1989) and da Costa & Marconi (1986) and furthermore of da Costa (1986 and 1993) I propose a minimal non-alethic logic of the first order with identity and Hilbert’s "-symbol (da Costa et al. 1992) which can bring into Meinongian spirit the most relevant aspects of Meinongian logic underlying Meinong’s theory of objects. This is just a first approach of mine, to account for a complex thought, but so interesting, too as Meinong’s thought. Furthermore, giving a positive answer to 1) and 2) I indicate a plausible way which can avoid both difficult approaches and the attempt to refuse the theory of objects in order to do not compromise standard logic and some of its own laws. My approach shows that Meinong’s theory can be a valid ontology, because there is adequate and not banal logic underlying it.

Fictional objects, non-existence, and the principle of characterization

I advance an objection to Graham Priest’s account of fictional entities as nonexistent objects. According to Priest, fictional characters do not have, in our world, the properties they are represented as having; for example, the property of being a bank clerk is possessed by Joseph K. not in our world but in other worlds. Priest claims that, in this way, his theory can include an unrestricted principle of characterization for objects. Now, some representational properties attributed to fictional characters, a kind of fictional entities, involve a crucial reference to the world in which they are supposed to be instantiated. I argue that these representational properties are problematic for Priest’s theory and that he cannot accept an unrestricted version of the principle of characterization. Thus, while not refuting Priest’s theory, I show that it is no better off than other Meinongian theories.

Ways of reference to Meinongian objects. Ontological commitments of Meinongian theories

A. Meinong’s views are usually associated with an highly inflated ontology including various kinds of nonexistent objects, incomplete and impossible ones among others. Around the turn of the century B. Russell strongly criticised this ontology accusing it of inconsistency. And perhaps because of this criticism Meinong’s views have been forgotten for a long time. Only recently some philosophers have created theories of objects which are formalisations of Meinong’s ontology or which are directly inspired by the ontology 1 . However there still remains a need to discuss the problem of ways of reference to Meinongian objects regarding their specific structure of predication, and the interrelated problem of ontological commitment of those theories.Meinong’s ontological theory seems to be not so bizzare if we interpret it as a theory of possible intentional objects, i.e. possible objects of consciousness. And this interpretation perfectly agrees with phenomenological tradition from Brentano to Husserl and with the contemporary philosophy of mind. Needless to say reference is a special form of intentional relation so semantical investigations into these objects are of great relevance for general theory of intentionality.

The Denoting Reader

Reviews 77 THE DENOTING READER NICHOLAS GRIFFIN Russell Research Centre / McMaster U. Hamilton, ON, Canada L8s 4M2 NGRIFFIN@MCMASTER.CA Gary Ostertag, ed. Definite Descriptions: a Reader. Cambridge, Mass., and London: MIT P., 1998. Pp. xii, 411. US$32.95 (cloth); US$30.00 (pb). Russell's theory of definite descriptions has been widely and continuously debated in the half century since Strawson published "On Referring".I Given the number of important papers written on the topic in that timenot to mention the even larger number of unimportant ones-it is more than a little surprising that there has been, until Ostertag's book, no substantial collection of readings on the topic. Ostertag's Definite Descriptions fills this gap admirably-though, as he admits at the outset, many important papers and some important topics have been left out. Among the latter he mentions descriptive pronouns (p. xi); I shall mention some others below. The papers can be divided into two main groups: four selections in which Russell's original theory is expounded and a group of eight papers taking sides on various current controversies over the adequacy of Russell's theory. In between are three more miscellaneous papers: Strawson's "On Referring", which couldn't well be omitted and, in any case, set several of the current controversies going; an extract from Carnap's Meaning and Necessity giving a version of Frege's conventional denotation proposal; and Karel Lambert's "A I Mind (1950); reprinted here as Chap. 6. 78 Reviews Theory of Definite Descriptions", which represents free logic.2 Lambert's paper is a decent introduction to free logic, but it was written too early to capture the important link between Strawson's paper and free logic which emerged with the development of supervaluational semantics for free logic. Sadly, there is nothing on Meinongian theories of descriptions, which have had a substantial revival since the 19705. Nor is there anything on more purely linguistic work on definite descriptions. Rereading Strawson's paper after all theseyears, one is struck first by how beautifully written it is. One is also struck by the fact that although it pointed to all sorts of fruitful developments, it does hardly anything to bring them about. An amusing case in point is the "very special and odd sense of 'imply'" (p. 145) in which, Strawson wants to say, "The king of France is wise" implies that there is a king of France. Strawson repeatedly invokes this sense and toward the end of his paper even refers to the sense as "by now familiar" (p. 157). But repeatedly calling something "strange" does not make it familiar, and Strawson says not one thing about it-except that it does not mean either "entail" or "assert". It is, of course, the useful notion of presupposition , given formal articulation by van Fraassen,3 against every warning by Strawson who maintained that "ordinary language has no exact logic" (p. 159). Here, and elsewhere, Strawson's paper proved seminal in spite of his best efforts. The four expository selections include three by Russell: "On Denoting" (of course), the much more accessible chapter on descriptions from Introduction to Mathematical Philosophy,4 .and the relevant sections of Principia Mathematica (pp. 30-2 and 66-71 from the Introduction, and the Summary of *14). These are well-nigh essential, but the dotty notation of Principia does pose problems for using the text in the classroom. Perhaps because I had assumed that the students would get the hang of it relatively quickly, I found myself continually obliged to stop and translate into a notation that used more brackets and fewer dots. Fortunately, Ostertag himself gives a lucid summary in more modern notation in his Introduction.5 The fourth paper in 1 From his Philosophical Applications of Free Logic (Dordrecht: Kluwer, 1962; New York: Oxford U. P., 1991)-wrongly cited in the bibliography as "Philosophical Foundations of Free Logic". 3 Cf "Presuppositions, Supervaluations, and Free Logic", in K. Lambert, ed., The Logical ~y of Doing Things (New Haven, Conn.: Yale U. P., 1969); and Formal Semantics and Logic (London: Macmillan, 1971), pp. 153-63. 4 Reprinted complete with typos: "a propositional fUnction x' (towards the foot...

Existence assumptions in knowledge representation

If knowledge representation formalisms are to be suitable for semantic interpretation of natural language, they must be more adept with representations of existence and nonexistence than they presently are. Quantifiers must sometimes scope over nonexistent entities. I review the philosophical background, including Anselm and Kant, and exhibit some ontological problems that natural language sentences pose for knowledge representation. The paraphrase methods of Russell and Quine are unable to deal with many of the problems. Unfortunately, the shortcomings of the Russell-Quine ontology are reflected in most current knowledge representation formalisms in AI. Several alternatives are considered, including some intensional formalisms and the work of Hobbs, but all have problems. Free logics and possible worlds don't help either. But useful insights are found in the Meinongian theory of Parsons, in which a distinction between nuclear and extranuclear kinds of predicates is made and used to define a universe over which quantification scopes. If this is combined with a naive ontology, with about eight distinct kinds of existence, a better approach to the representation of nonexistence can be developed within Hobbs' basic formalism.

Meinongian Theories of Generality

Facing problem of making semantic sense of some complex expression in ordinary language, one can sometimes adopt easy way out claiming that expression is not to be taken at face value. Then one feels free to appeal to some purportedly deeper level of analysis, where no urge is felt, for one reason or another, of assigning any one semantic value to it. E.g., one can claim that what appears to be a complex but unitary expression is such only at surface level, whereas it is to be paraphrased away at level of logical form. Being entirely faithful to ordinary language, i.e., keeping appeal to surface/depth dichotomy to a minimum, is usually much more difficult. Meinong, it seems to me, tried to adopt latter strategy, while Russell gave an outstanding example of former in his theory of definite descriptions. In this, however, as in so many other ways, he was simply following Frege. There are sentences containing definite descriptions for which Russell's analysis obviously fails, namely those such as whale is a mammal, horse is a four-legged animal, etc., where no individual whale or horse is meant in particular. Russell, however, did not have to bother with them: in fact, Frege had already dealt with them and way he had done so is paramount example of strategy of paraphrasing away. As we all know, sentence horse is a four-legged animal is taken by Frege to mean All horses (or, all normal specimen of horses) are four-legged. The expression the horse, as a whole, has disappeared from logical form and so has need of making semantic sense of it and assigning it some entity or other. In fact, idea of having indefinite entities play this role was ridiculed by Frege (see, e.g., What is a Function?) just as much as he did in case of variable

To Be and Not To Be

Since the mid-1970s, there has been a revival of interest in the philosophy of Alexius Meinong and an attendant flurry of Meinong-inspired theories.' One of the pioneering efforts was Terence Parsons's 1974 article, A Prolegomenon to Meinongian Semantics (Parsons, 1974), which was followed by a series of articles in which he extended and elaborated his theory, culminating in his 1980 book, Nonexistent Objects (Parsons, 1980). The present essay is a critical and comparative study of Parsons's seminal and exciting work in this area, concentrating on the informal and formal versions of his theory as presented in his book.2 I begin with a discussion of the nature of intentional objects, their properties, and modes of predication as presented in Parsons's informal version of his theory. I argue that his view of objects does not adequately reflect our ordinary ways of speaking and thinking, and I defend Meinongian theories that recognize two modes of predication against Parsons's objections, which are based on his preference for two kinds of properties. I then consider Parsons's application of his theory to fictional objects, pointing out problems with his view that can be avoided by maintaining (contra Parsons) that no existing entities ever appear in works of fiction. I conclude with an outline of one of Parsons's formal versions of his theory, raising some questions and pointing out some difficulties and a curious consequence about modes of predication.

Meinongian Theories and a Russellian Paradox

This essay presents a re-examination of Alexius Meinong's article Uber Gegenstandstheorie (On the Theory of Objects) [26] and undertakes a clarification and revision of it which, I hope, is both faithful to Meinong and capable of overcoming the various objections to his theory that have appeared in the literature.1 I then turn to a discussion of a historically and technically interesting Russell-style paradox that arises in the modified theory. I also examine the alternative Meinonginspired theories of Hector-Neri Castafieda and Terence Parsons, using the modified theory as a sharper tool for investigating their worth than that provided by unaided intuitions or comprehensive, ad-hoc theory fragments. As with all theories, many of my claims are not susceptible of proof but, rather, gain their plausibility and value from their ability to deal with data and to provide solutions to various problems. The two main problems which, I believe, a properly constructed Meinongian theory ought to be capable of handling are, first, a linguistic problem of long-standing philosophical concern: that of providing a foundation for a semantics of natural languages, and, second, the problem of intentionality and the analysis of the structure of psychological discourse. Even Quine, ordinarily no friend of intentional language, attests to the importance of the latter problem, considering such discourse to be less clearly dispensable than other modalities ([36]: 336). For this problem, the theory must embody a characterization of the objects of thought (in the sense of that which is thought about). In order to account for the psychological phenomenon illustrated by puzzles concerning objects