Language and Value E-Book

Yi Jiang

• Herausgeber: Books on Demand
• Kategorie: Geisteswissenschaft
• Serie: ProtoSociology
• Sprache: Englisch

Seitenzahl: 495

Veröffentlichungsjahr: 2015

Beschreibung

This volume represents the first substantial collaborative work from Chinese and Western scholars on philosophy of language. We believe that recent developments in the Anglo-American philosophy and Chinese philosophy of language not only suggest a tableau of shared problems, but also a number of similar methods and goals. It is the goal of this volume to juxtapose these recent philosophical developments to illustrate the similarities between the traditions, but also to spur common dialogue and hopefully efforts to jointly engage shared philosophical problems. The issues engaged range from the role of cultural and biological factors in linguistic competence and language use, to figurative speech in different languages, the comparative study of work in semantics and philosophy, and the role of language in establishing the legal and ethical norms and values. Content and abstracts: www.protosociology.de

Leseprobe

ProtoSociology

An International Journal of Interdisciplinary Research

Volume 31, 2014

Language and Value

Edited by Yi Jiang and Ernie Lepore

CONTENTS

Introduction

Ernest Lepore and Yi Jiang

S

EMANTICS AND

O

NTOLOGY

The Relation of Language to Value

Jiang Yi

Refutation of the Semantic Argument against Descriptivism

Chen Bo

Semantics for Nominalists

Samuel Cumming

Semantic Minimalism and Presupposition

Adam Sennet

Compositionality and Understanding

Fei YuGuo

Values Reduced to Facts: Naturalism without Fallacy

Zhu Zhifang

W

ORD

M

EANING,

M

ETAPHER, AND

T

RUTH

Philosophical Investigations into Figurative Speech Metaphor and Irony

Ernie Lepore and Matthew Stone

Norms of Word Meaning Litigation

Peter Ludlow

The Inconsistency of the Identity Thesis

Christopher Hom and Robert May

Describing I-junction

Paul M. Pietroski

Predicates of Taste and Relativism about Truth

Barry C. Smith

Mood, Force and Truth

William B. Starr

A Semiotic Understanding of Thick Terms

Aihua Wang

F

EATURES OF

C

HINA’S

A

NALYTICAL

P

HILOSOPHY

An Echo of the Classical Analytic Philosophy of Language from China: the Post-analytic Philosophy of Language

Guanlian Qian

The Chinese Language and the Value of Truth-seeking: Universality of Metaphysical Thought and Pre-Qin Mingjia’s Philosophy of Language

Limin Liu

Mthat and Metaphor of Love in Classical Chinese Poetry

Ying Zhang

On ProtoSociology

Digital Volumes available

INTRODUCTION

Ernest Lepore and Yi Jiang

It is our great pleasure to introduce this volume of essays, which represent (in our modest opinion) the beginning of an era of Chinese and English language philosophical cooperation. The Chinese philosophical tradition and the Western philosophical tradition (starting with the Ancient Greeks) have often been considered orthogonal to each other—two ships passing in the night without much to say to each other. Whatever the merits of this thesis, we believe that recent developments in the Anglo-American philosophy and Chinese philosophy of language not only suggest a tableau of shared problems, but also a number of similar methods and goals. It is the goal of this volume to juxtapose these recent philosophical developments to illustrate the similarities between the traditions, but also to spur common dialogue and hopefully efforts to jointly engage shared philosophical problems.

The essays in this volume (by contributors from the United States, Great Britain, Australia and China) address issues about language and value from philosophical perspectives that overlap in surprising ways. The issues engaged range from the role of cultural and biological factors in linguistic competence and language use, to figurative speech in different languages, the comparative study of work in semantics and philosophy, and the role of language in establishing the legal and ethical norms and values.

This volume represents the first substantial collaborative work from Chinese and Western scholars on philosophy of language. Although there has previously been some collaboration between Chinese and English language academics, there has never been a contemporary philosophical collection of this design, where each essay appears in both Chinese and English, and the two versions are presented side by side. It is our hope that by this endeavor we will be able to encourage greater collaborations among scholars both in China and in the English-speaking world. Our mission is to provide a bridge between traditions, leading to mutual understanding, philosophical collaboration, and ultimately a level of philosophical progress that could not be achieved were we working separately. We believe there is much to be gained by our working together in such an enterprise.

To this end, in addition to assembling the essays in this volume, we paired each contributor from China with an English speaking philosopher; the paired philosophers read, edited and commented on each others work. The philosophers were thus able to rewrite their papers with the help of advice from a fresh perspective. In some cases this led to the introduction of arguments and moves that had not previously been considered; in other cases it led to a better handle on how the ideas should be presented—it opened the ideas to a broader philosophical audience. We believe that this has made for a compelling, globally accessible, and yes revolutionary new body of philosophical work.

The Chinese philosophers who have contributed represent a broad spectrum of the current academic world in China.

CHEN Bo is a professor of Philosophy at Peking University. He is one of most distinguished scholars in logic and analytic philosophy in China. He has published several essays in top journals in A&HCI list. His interests are philosophy of logic, Quine’s philosophy, and philosophy of language.

Professor JIANG Yi of the School of Philosophy and Sociology, at Beijing Normal University, is one of most distinguished scholars in analytic philosophy and philosophy of language in China. He has published several books in Chinese on Wittgenstein and analytic philosophy. His interests are analytic philosophy, philosophy of language and Wittgenstein.

Professor LIU Limin, of School of Foreign Languages, is at Sichuan University. His interests are philosophy of language and linguistics.

Professor QIAN Guanlian, of Guangdong University of Foreign Studies, has for some time now been a important and distinguished scholars in pragmatics and the philosophy of language. He has published several books in Chinese on pragmatics.

YUGUO Fei, is a member of the Philosophy of Department at Yunnan University in Kunming. He received his PhD from Wuhan University, and his chief interests include logic and philosophy of language.

WANG Aihua is an Associate Professor of School of Foreign Language, at the University of Electronic Science and Technology of China in Chengdu. Her interests are also in pragmatics and philosophy of language.

ZHANG Ying, of Philosophy Department, at Huazhong University of Science and Technology, received her PhD from Sun Yat-sen University. She was a visiting professor at Rutgers for two years, and her chiefs interests are the philosophy of language, logic and pragmatics.

Lastly, Professor ZHU Zhifang, of Philosophy Department, at Wuhan University. He is one of more distinguished scholars in analytic philosophy and logic in China. He has published several books and essays in logic and philosophy of language. His interests are logic, philosophy of language and semiotics. From the English speaking community our volume includes previously unpublished papers by a number of distinguished theorists of language.

Adam SENNET, an Associate Professor, Dept. of Philosophy, UC Davis, specializes in the philosophy of language and has published papers in Mind and Language, Philosophical Studies, and the Journal of Philosophical Logic.

Samuel CUMMING, an Assistant Professor of Philosophy, UCLA, has interests in the philosophy of language, philosophy of mind, semantics and pragmatics.

Peter LUDLOW is Professor of Philosophy at Northwestern University. He has published on topics ranging from the philosophy of linguistics to the metaphysics of time. His current interests include the dynamics of communication (including the dynamics of the lexicon) and the optimization of group knowledge in adversarial environments.

Christopher HOM, Assistant Professor of Philosophy, Texas Tech University. He earned his PhD from the University of California, Irvine (LPS). His research interests are in philosophy of language, metaethics, and philosophy of race. He has published articles on racial slurs and normative language generally.

Paul PIETROSKI, Professor of Linguistics and Professor of Philosophy, University of Maryland. His research interests are focused on how linguistic meaning is related to human psychology.

Robert MAY is Distinguished Professor of Philosophy and Linguistics at the University of California, Davis. He is the author of Logical Form: Its Structure and Derivation,and with Robert Fiengo of Anaphora and Identity and De Lingua Belief. He is well-known for his work in the syntax and semantics of natural language, especially on natural language quantification, and has written extensively on Frege, along with other topics in philosophy of language and philosophy of logic.

William STARR, Assistant Professor at the Sage School of Philosophy, Cornell University. His research is on communication and cognition, drawing on ideas across philosophy, linguistics, logic, artificial intelligence and psychology. His published work explores these themes through various linguistic phenomena including conditionals, questions, imperatives, modality and speech acts.

Matthew STONE completed his Ph.D. in the Computer and Information Science Department at the University of Pennsylvania in 1998. Since then he has had an appointment in the Computer Science Department and Center for Cognitive Science at Rutgers, the State University of New Jersey. Stone has had visiting positions at the University of Edinburgh and the Universität Potsdam. He works on problems of meaning in human-human and human-computer conversation.

Barry C SMITH is a Professor of Philosophy and Director of the Institute of Philosophy in the School of Advanced Study, University of London, where he co-directs the Centre for the Study of the Senses. He has written mostly on the philosophy of mind and language, on the topics of self-knowledge and our knowledge of language. He co-edited The Oxford Handbook of Philosophy of Language (2006) with Ernest Lepore. Following his 2007 collection, Questions of Taste—the philosophy of wine (Oxford University Press), he began working with psychologists, neurologists and neuroscientists on flavour perception and is now the co-organiser of an international research project on the Nature of Taste. He has been a Visiting Professor at the University of California at Berkeley and at the Ecole Normale Supèriere, and was the writer and presenter of the BBC World Service radio series, The Mysteries of the Brain.

Ernest LEPORE is the Director of the Center for Cognitive Science at Rutgers University and a Professor of philosophy. His chief interests are philosophy of language and philosophy of mind.

I. Semantics and Ontology

THE RELATIONOF LANGUAGE TO VALUE

Jiang Yi

Abstract

How does language relate to value? Why do we concern with the relation up to now? I will analyze the background of increasing interests in the relation of language to value in contemporary philosophy of language, provided with ideas that language has meaning with intention which determines the way of acts in relation with values in societies, and that, when we consider the value in language, we are searching for consequences of our speech acts for final goals of language.

In 1836 Alexander Bryan Johnson (1786–1867), so-called a philosophical banker, published his unrecognized book, A Treaties on Language, which was the continuity of his first writing, Philosophy of Human Knowledge in 1828. In the book he committed the meaning of words as reminder of human knowledge which seems to be seen to have anticipated the thrust of logical positivism, at least in arguing that misunderstandings of how language operates bedevil philosophical questions, and theories of modern linguistics. After a century of his death a conference on the life and works of Johnson was held in Utica in1967. The proceedings of the conference was published in 1970, entitled Language and Value, in which he was interpreted as a generalist as a banker, businessman, essayist, satirist, and philosopher. The title of the proceedings hints his binary character of banker and philosopher. This might be the first time to relate language to value, though not in professional philosophy.

In 2002 Diana Mary Kilpert published her Language and Value: The Place of Evaluation in Linguistic Theory, in which she tries to address the evaluation in language studies. But it is just in linguistic sense that we can evaluate languages in our social activities. Much writings on languages of evaluation from different perspectives, such as sociological and political, appeared recently. Most of them are concerned with applications of the theory of evaluation in language rather than the theory itself. Philosophers of language would, in contrast, consider the relation of language to value, concentrating on value elements of language in use, not evaluation in language. Thus we should clarify firstly some distinctions among those concepts which are confused with in our discourse of the relation of language to value.

The Pragmatic are concerned mostly with terms which are full in evaluation and appraisal. Thus they discuss implications of those terms in use, not the meaning of such a term in its self. So in this sense there is difference between language of evaluation and evaluation of language, which makes our discussion clearer in clarification of the relation of language to value. The difference is that, while talking in the language of evaluation, we are not evaluating the language but the implications of language in which we express our intention to evaluate. Further on, even if we acknowledge the difference of the two we have to know that we could not understand the terms with speaker’s intentions of evaluation without any knowledge of language itself. So we have to know what makes us to use the language of evaluation to appraise what we want to do so.

Philosophers of language do not want to talk about the value or evaluation of language but the correlation of language to value which is central to discussion of implication of languages in use. The term, value, here refers to varieties of implicatures of language when we use them to demonstrate and fulfill our goals for expressions. The question why language is concerned with value is involved in most flourished discussions on Oxford Ordinary Philosophy since 1950s. Stanley Cavell’s well-known paper, Must we mean what we say, in 1957 explained the reason why we should not assert what we know when we talk about what we might think we know. He said:

The nature of the Oxford philosopher’s question, and the nature of his conception of philosophy, can be brought out if we turn the question upon itself, and thus remind ourselves of when it is we need to remind ourselves of what we should say when. Our question then becomes: When should we ask ourselves when we should (and should not) say “The x is F” in order to find out what an F(x) is? The answer suggested is: When you have to.1

That you have to means that there is something tacit you would know already when you say in the ordinary language. And this is a sense in which we have to deal with significance of words in the language.

Charles Travis in his paper, Pragmatics, provided us with a pragmatic view which is opposite to a semantic one. The pragmatic view is that “it is intrinsically part of what expressions of English mean that any English sentence may, on one speaking of it or another, have any of indefinitely many different truth conditions, and that any English expression may, meaning what it does, make any of many different contributions to truth conditions of wholes in which it figures as a part.”2 It is involved in discussion of implicature of words in utterances when we use them to mean something else that some problems on intensionality and propositional attitudes would be part of being central to the philosophy of language.

Scott Soames in his recent book, What is Meaning, challenges the traditional theory of propositions from Frege and Russell by providing a new view which is explained as being cognitive relative. In this view propositions are take to be cognitive-event types, provided that one’s acquaintance with and knowledge of propositions is acquaintance with and knowledge of events of one’s cognitive life. He says:

Propositions, as I understand them, can play the roles for which they are needed in semantics, pragmatics, and other areas of philosophy. However, they are not the source of that which is representational in mind and language. Sentences, utterances, and mental states are not representational because of the relations they bear to inherently representational propositions. Rather, propositions are representational because of the relations they bear to inherently representational mental states and cognitive acts of agents.3

As we know, mental states and cognitive acts of agents are intention-directed which relate to the value relations of speakers and hearers. In this sense Soames locates meaning in thought, perception, and the cognitive acts of agents.

Ernie Lepore and Barry Loewer in their new book, Meaning, Mind and Matter, defended and expanded three views in the philosophy of language for several decades. They provided a much strong argument for the irreducibility of the mental to the physical. According to it the “global supervenience” does not require the existence of strict laws connecting physical with mental properties or that mental properties are identical to physical properties.4 In contrast,

The interesting point for us is that what will happen, or if we allow probabilities over micro histories, the probabilities of what will happen are given by adding one or the other decision to the state that is most similar to the actual state that contains the brain state corresponding to the decision. So the reason we are interested in evaluating counterfactuals along Lewisian lines is that conditionals so evaluated contain information about the likely results of our decisions and this information is enormously important to our getting what we want.5

It seems that the two philosophers would provide us with more interesting explanations of the gap in the mental and the physical. If we understand well the motives of the authors in their book, there will be some intentions in their explanations to clarify the extra significance of the mental a priori to the physical. And if we understand well the significance, we can locate it to our society, though it is not immediate in society.

All above demonstrate that philosophers of language began to pay much attention to the relevance of language to value in which the intention such as propositional attitudes and other properties of utterances in our language is expressed vaguely or explicitly. But the questions remain: how does language relate to value? And, why we concern with such a relation up to now?

First of all, we should make clear on the issue by discussing the relevance of language to the problem of value. As we mentioned the term value here is involved in the significance of language we use in society. The problem of value in the philosophy of language has been considered as a part of pragmatics when we use language to express our intentions, and evaluate consequences of speakers’ utterances on hearers and actions. There is, however, a substantial part in the problem of value in the philosophy of language. According to the previous consideration the problem and language might be distinct each other. So, we are expressing our intentions in language when we talk about the problem in the philosophy of language. But it is, I would say, on a wrong way. We are not expressing our intentions in language when we talk about the problem. We are communicating each other by language in which our intentions are expressed and understood. In this sense the value of language is shown or illuminated implicitly by utterances in language. In other words, speaking strictly, language is marked by value.

From this start point we can argue that language has meaning with intention which determines the way of acts in relation with values in societies, and that, when we consider the value in language, we are searching for consequences of our speech acts in advance for final goals of language.

The meaning of words has been considered as consisting of intentions the speaker intends to express in those words. From Frege on the problem of meaning of words is a puzzle, for not any single theory of meaning could settle down the puzzle for a century. The main reason I think is that we almost forgot the nature of language as the mark of human value. We are content to think language as an entity which bears varieties of function in different theories. However, language is not an entity which we can attribute properties to. Language should be thought of as a process which develops smoothly our intentions determining the way of acts we do in language. It is indispensable that the way of acts in language is shown in our society in which the function of language can be fulfilled with intentions speakers express in language.

References

S. Cavell, 1976, Must We Mean What We Say, Cambridge: Cambridge University Press.

B. Hale & C. Wright, ed.,1997, A Companion to the Philosophy of Language, Oxford and New York: Blackwell.

E. Lepore & B. Loewer, 2011, Meaning, Mind ,& Matter, Oxford and New York: Oxford University Press

S. Soames, 2010,What is Meaning, Princeton: Princeton University Press.

1 S. Cavell, Must We Mean What We Say, Cambridge University Press, 1976, p.21.

2 C. Travis, Pragmatics, in B. Hale & C. Wright (ed.), A Companion to the Philosophy of Language, Blackwell, 1997, p.87.

3 S. Soames, What is Meaning, Princeton University Press, 2010, p.7.

4 E. Lepore & B. Loewer, Meaning, Mind ,& Matter, Oxford University Press, 2011, p.7.

5Ibid, pp.231–232.

REFUTATIONOF THE SEMANTIC ARGUMENTAGAINST DESCRIPTIVISM

Chen Bo

Abstract

There are two problematic assumptions in Kripke’s semantic argument against descriptivism. Assumption 1 is that the referential relation between a name and its bearer is only a metaphysical relation between language and the world; it has nothing to do with our public linguistic practice. Assumption 2 is that if name N has its meaning and the meaning is given by one description or a cluster of descriptions, the description(s) should supply the necessary and sufficient condition for determining what N designates; it is possible for us to find out such a condition for fixing the referent of N. Emphasizing the sociality, conventionality and historicity of language and meaning, this paper criticizes Assumption 1 and Assumption 2, and concludes that Kripke’s semantic argument fails.

1. Opening

To refute descriptivism, Kripke reformulates its cluster version refined by Wittgenstein and Searle. For him, cluster-descriptivism consists of six theses, in which theses (1), (3) and (4) are the targets of his semantic argument:

(1) To every name or designating expression ‘X’, there corresponds a cluster of properties, namely the family of those properties φ such that [the speaker] A believes ‘φX’.

(3) If most, or a weighted most, of the j’s are satisfied by one unique object γ, then γ is the referent of ‘X’.

(4) If the vote yields no unique object, ‘X’ does not refer.6

In my view, Kripke’s semantic argument can be summarized as follows.

If descriptivism is correct, that is, name N is exactly synonymous with one description or a cluster of descriptions, then, the meaning7 of N should be the necessary and sufficient condition for determining what N designates. In other words, if an object satisfies the corresponding description(s), it is the semantic reference of N (i.e., the sufficiency of meaning of N for fixing the referent of N); if an object does not satisfy the description(s), it is not the semantic reference of N (i.e., the necessity of meaning of N for fixing the referent of N). However, for a great number of names it is not the case that the corresponding description(s) constitutes the necessary and sufficient condition for identifying their references. So descriptivism gets the semantic facts wrong.8

This argument can be reformulated more simply as follows, in which ‘P1’ for premise 1, ‘C’ for the conclusion, ‘N’ stand for a name, and so forth.

P1If descriptivism is correct, then, the meaning of name N, which is givenby one description or a cluster of descriptions, should be the necessaryand sufficient condition for determining what N designates.P2In fact, the corresponding description(s) cannot supply the necessaryand sufficient condition for determining the referent of N.CDescriptivism is wrong.

For this argument, I accept P2, but I reject P1; so I do not accept conclusion C. I think that there are two problematic assumptions in the argument:

Assumption 1 (A1 for short): The referential relation between a name (or a description) and its bearer is strictly ‘objective’ or ‘metaphysical’; in particular, it is not sensitive to the facts about our linguistic community; in other words, it has nothing to do with our linguistic community.

I will argue that A1 is wrong, because the referential relation between a name (or a description) and its bearer is actually a social relation, which concerns three elements, i.e. the name (or a description), the object, and our linguistic community. What a name (or a description) designates depends on what our linguistic community uses the name (or a description) to designate.

Assumption 2 (A2 for short): If name N has its meaning and the meaning is given by one description or a cluster of descriptions, the corresponding description(s) should supply the necessary and sufficient condition for determining what N designates. Also, it is possible for us to find out such a condition for fixing the referent of N.

I will argue that A2 is wrong for three reasons: (a) A2 is a misunderstanding

[Bedeutung] of a linguistic expression; in its narrow sense, ‘meaning’ only signifies to the sense of an expression, which could be understood and grasped by human minds. This paper only uses the word ‘meaning’ in its narrow sense.

or distortion of traditional descriptivism. (b) We cannot require that a proper name is exactly synonymous with some description(s), and cannot ask the necessary and sufficient condition for fixing what the name designates, because there is no such condition at all. (c) When determining the referent of a name by means of the meaning of the name, we should consider not only the factual satisfaction relation between an object and some description(s), but also speakers’ intention, Network and Background, all of which together determine what a name designates.

I will conclude that Kripke’s semantic argument fails.

2. Exposing the Assumptions of the Semantic Argument

2.1. Assumption 1

Kripke tries to disprove thesis (3) of cluster-descriptivism by offering some counterexamples, for there can be situations in which the family φ of descriptions corresponding to a name N is actually satisfied by a unique object y, but y is still not the referent of N.

Fictional Cases. Let us imagine a counterfactual situation. Gödel had a friend called ‘Schmidt’, who had actually proved the incompleteness of arithmetic. But Gödel somehow got hold of the manuscript and published it in his own name. Then Gödel achieved fame as ‘the man who discovered the incompleteness of arithmetic’. However, in fact, the semantic referent of that description is the man Schmidt. If ‘Gödel’ is synonymous with the description ‘the man who discovered the incompleteness of arithmetic’, does ‘Gödel’ change its referent into the man Schmidt? Kripke replies ‘No’, ‘Gödel’ still designates the person called ‘Gödel’ whereas ‘the man who discovered the incompleteness of arithmetic’ refers to the man Schmidt, because Schmidt is actually the person satisfying that description, and we make a mistake when using the description to refer to Gödel.

Non-fictional Cases. It has been commonly believed that Peano is the man who discovered certain axioms which characterize the sequence of natural numbers. But actually it is Dedekind who discovered these axioms earlier; thus the description ‘the man who discovered certain axioms which characterize the sequence of natural numbers’ denotes Dedekind. Many people mistakenly regard Einstein as both the discoverer of the theory of relativity and the inventor of the atomic bomb. But actually it was not a single person but a group of people who invented the atomic bomb. Similarly, many people regard Columbus as the first man to realize that the earth was round and the first man who discovered America. However, there might have been someone else who is the semantic referent of these descriptions, whereas ‘Columbus’ still refers to the person called Columbus.

From these cases, Kripke argues that one description or even a cluster of descriptions is not the sufficient condition for identifying what a name designates. It is possible that what satisfies the description(s) is not the referent of the name but that of another name.

I find an assumption hidden in the above argument by Kripke, namely, that the question of ‘how does some description(s) refer to an object?’ just concerns the relation between the description(s) and its satisfier, between a language and the world, and has nothing to do with us as the users of the description(s) and the language. Straightforwardly, the referential relation between some description(s) and its satisfier is only the matter of fact, not relevant with our intentions, conventions and customs to use the description(s); in other words, the semantic referent of some description(s) is exactly the object which in fact satisfies the description(s), rather than the object of which our linguistic community thinks the description(s) are true. For example, if the man Schmidt actually satisfies the description ‘the man who discovered the incompleteness of arithmetic’, then the description refers to Schmidt; If Dedekind in fact discovered certain axioms which characterize the sequence of natural numbers earlier than Peano did, the description ‘the man who discovered certain axioms which characterize the sequence of natural numbers’ refers to Dedekind rather than Peano; If someone else, not Columbus, is really the first man to realize that the earth was round and the first man who discovered America, then the guy is the semantic referent of the description(s). Though we use the descriptions mentioned above to separately designate the men Gödel, Peano, and Columbus, our uses are totally wrong.

I can list some other evidences for that Kripke holds A1 in his theory of rigid designation.

(1) Kripke separates two questions apart: one is ‘how does a name designate an object?’, which seems to be an objective relation between a name and its bearer, has nothing to do with us as the users of the name; the other is ‘how do we determine what a name designates?’, which is a social-historic relation between a name, its bearer and the users of the name. His response to the first question is the theory of ‘rigid designation’, while his response to the second is ‘by a causal-historic chain’.

Kripke asserts that a name designates an object rigidly and directly, without the help of its meaning as an intermediary. For example, proper name ‘Aristotle’ always designates the man Aristotle. However, we can imagine that Aristotle had an entirely different career, that is, he could have done nothing that is attributed to him in the actual world. Even under such counterfactual circumstances, we are still talking about the man Aristotle rather than someone else. Kripke maintains that a proper name is a rigid designator which refers to the same object in all possible worlds in which the object exists; Even if an object did not exist in some possible world, the name could still refer to it, if to anything. As far as their mechanism of reference is concerned, natural kind terms are similar to proper names. For Kripke, the key point is that the referential relation between a name and an object is almost a metaphysical relation between the two, and for determining the referent of the name we do not need any knowledge of the object to which we use the name to refer. After a name has been given to an object in a baptism, the name will be used to refer to the originally named object by all people who hear, speak and write the name, even if those people have no knowledge of the object.

(2) Kripke has expressed a strong doubt to the doctrine of ‘division of linguistic labor’.

Putnam expounds that the users of natural kind terms do not always know how to identify the referent of ‘gold’ and how to distinguish between the bearers of ‘an elm’ and the bearers of ‘a beech’; They have to rely on some experts in their linguistic community who are qualified to address on these issues. On the basis of this common phenomenon, Putnam proposes his hypothesis of universality of the division of linguistic labor:

Every linguistic community exemplifies the sort of division of linguistic labor just described, that is, possesses at least some terms whose associated “criteria” are known only to a subset of the speakers who acquire the terms, and whose use by the other speakers depends upon a structured cooperation between them and the speakers in the relevant subsets.9

However, Kripke does not agree with Putnam at this point, because he thinks that ‘what does a name designates?’ is a semantic question to which there is a definite answer; Experts could not help us on this issue, since they have no special semantic power or authority to determine the referent of a name.

Kripke talks about the terms such as ‘gold’, ‘member of the French Cabinet, Minister of State, in the twentieth century’, ‘elm’ and ‘beech’. First, these terms have determinate extensions. It is determinate whether or not something is in their extensions. The extensions of such terms have nothing to do with the time of speaking these terms, e.g. the term ‘gold’ spoken in the time of Ancient Greece has the exactly same extension as that spoken in this century. Also, the extensions of such terms have nothing to do with the speakers, e.g. the term ‘gold’ spoken in the ordinary people’s mouths has the exactly same extension as that spoken in experts’ mouths. As for the second description listed above, Kripke says:

… the term just means what it does. It may be difficult or hard to determine whether something is in the extension; this is a special problem of what we are going to know. Sometimes we may not know what terms are in the extension, what objects are in the extension or not, for a very long time. But the experts provide no help as far as actually determining the extension of the term. They only help us find out after a while which things actually fall into the extension of the term.10

Second, the experts might not be qualified, e.g. he might be only a fancied expert such as alchemist or astrologer. Even if he is eligible, he might make a mistake. For instance, his belief about the name’s referent might be false; even worse, there might be no experts who can determine the extension of some special names. Kripke emphasizes:

… in the case of natural kind terms, experts have no special linguistic authority. As Hilary Putnam himself says in another passage “there are just people who know a lot about gold”, they do not have any kind of authority analogous to the Académie française, a special authority over the extension of the term.11 … Now, actually, I think the term “division of linguistic labor” contains a strong suggestio falsi. I don’t know that it is false or wrong because, as meant by Putnam, it may be right. Almost all the connections that I can gather from it, and especially the ones that have been taken over by others such as Dummett, seem to me to be, first and most important, I suppose, false and second, and perhaps therefore, incompatible with the quite correct things Putnam has said elsewhere, even in the same papers that emphasized this concept.12

However, Kripke recognizes the role which experts play in two kinds of cases: (i) Some terms come from experts, since they are created by the experts and then spread into the community. Under such circumstances, experts indeed have some kind of authority. However, this is not because experts have special semantic power but because they are the producers of the terms in some baptisms. (ii) Experts have some roles to the reference shift of names. They can be guardians against contamination of samples by spurious items, which, if we do not watch out, may take over the role of central items and change the referents of the corresponding names. That is to say, a natural kind term, which originally designates item A, changes into another term denoting item B, when the samples have been contaminated. The more experts there are around, the less likely this is to happen.

Kripke also extends his arguments and conclusions to the cases of proper names such as ‘Peano’. He believes that, contrary to what people usually suppose, his own theory of names is incompatible with Putnam’s theory about division of linguistic labor. Furthermore, division of linguistic labor is even incompatible with some quite correct things Putnam says elsewhere. He emphasizes:

… normally we think of the relevant semantic feature as preserved. That is the essence of the historical theory. A speaker at any given time over time, and even if he has forgotten most of the descriptions he associates with the name of the being, or he may be an amnesiac, still counts normally as preserving the same reference that he had before.13

In sum, Kripke implicitly holds that the referential relation between a name (or a description) and its bearer is strictly the ‘objective’ or ‘metaphysical’ relation between language and the world; in particular, it is not sensitive to facts about the linguistic community, that is, it has nothing to do with our public linguistic practice. This is Assumption 1 (short for A1) of the semantic argument.

2.2. Assumption 2

To refute thesis (4) that ‘[i]f the vote yields no unique object, “X” does not refer’, Kripke made two comments as follows.

(1) It’s possible that the vote yields no unique object; since a description is not sufficient to fix the referent of a name, and there might be more than one object satisfying the description.

Kripke gives three cases as examples. About what the name ‘Cicero’ designates many people know only that he was ‘a famous orator of ancient Rome’, about ‘Feynman’ only that he was ‘a physicist’. It’s obvious that such description(s) is not sufficient for fixing a unique referent of the corresponding name. Kripke further argues that, even though we know nothing about the referent of N, N still refers to a particular object, e.g., in the case of ‘Nancy’:

… A mathematician’s wife overhears her husband muttering the name ‘Nancy’. She wonders, whether Nancy, the thing to which her husband referred, is a woman or a Lie group. Why isn’t her use of ‘Nancy’ a case of naming? If it isn’t, the reason is not indefiniteness of her reference.14

(2) It’s possible that the vote yields no object, that is, that there is no object satisfying all or most of the corresponding descriptions.

Kripke mentions that Biblical scholars generally held that Jonah did exist, but most of descriptions that the Bible attributes to him are false.15 Even so, ‘Jonah’ still refers to the person called ‘Jonah’, though he did not accomplish anything the Bible attributes to him. Likewise, it could be imagined that Moses did not accomplish anything described by the Bible, but from this it cannot be concluded that Moses did not exist or the name ‘Moses’ has no referent. Moreover, it could also be imagined that Aristotle did not accomplish anything that we usually attribute to him, but ‘Aristotle’ still refers to the man Aristotle.

From these cases, Kripke argues that one description or even a cluster of descriptions is not the necessary condition for fixing what a name designates. It is possible that what does not satisfy the description(s) is still the referent of the name.

In sum, Kripke’s semantic argument against descriptivism runs like this: if descriptivism is correct, and one description or a cluster of descriptions constitutes the meaning of name N, then the corresponding description(s) should provide the necessary and sufficient condition for fixing the referent of N. However, the description(s) cannot play such a role, because what does satisfy the description(s) may be not the referent of N but that of another name, and what does not satisfy the description(s) may be still the referent of N. Therefore, descriptivism is wrong.

From the semantic argument, I find another assumption A2, which says that, (a) If name N has its meaning and the meaning is given by one description or a cluster of descriptions, the description(s) should provide the necessary and sufficient condition for fixing what N designates; (b) It is possible for us to find out such a condition for determining the referent of N.

3. Refuting the Assumptions of the Semantic Argument

3.1. Refutation of Assumption 1

According to A1, language seems to be an automatic system relating itself to the external world, that is to say, names designate external objects by themselves, and propositions describe external states of affairs by themselves; all these things are independent of our linguistic community. I think that this is a totally wrong way of characterizing how languages work.

Oxford English Dictionary says that language is ‘the whole body of words and of methods of combination of words used by a nation, people, or race’. Here, I’d like to emphasize three characteristics of language as follows.

(1) Language is social. That is to say, language is closely connected with human beings and human society; it emerges and develops with the emergency and development of human society; children’s acquisition of a language is the process of how they are humanized and socialized.

Dewey asserts rightly:

Language is specifically a mode of interaction of at least two beings, a speaker and a hearer; it presupposes an organized group to which these creatures belong, and from whom they have acquired their habits of speech. It is therefore a relationship, not a particularity… The meaning of signs moreover always includes something common as between persons and an object. When we attribute meaning to the speaker as his intent, we take for granted another person who is to share in the execution of the intent, and also something, independent of the persons concerned, through which the intent is to be realized. Persons and thing must alike serve as means in a common, shared consequence. This community of partaking is meaning.16

Dummett also claims that,

…language is a social phenomenon, in no way private to the individual, and its use is publicly observable.17

He thinks that a language is constituted by the conventional practices and agreed standards of usage, so in using words individual language-users must hold themselves responsible to the standards of use of the language to which those words belong.

According to this standpoint, language is not an automatic system relating itself to the external world; It is human beings, who use a language, that build the bridge between the language and the world, and that create the referring or predicating relation between a name (or a proposition) and an object (or state of affairs). So, the referential relation between a name and its bearer depends on both our understanding of the name and the true situation of the object in the world. Likewise, the truth value of a predication also depends on two elements, i.e. the existent states of things in the world and our ways of speaking about the things. It is not the case that semantics takes no account of speakers; on the contrary, it just does not consider individual speakers, but must consider our linguistic community. Any talk about the meaning and reference of a linguistic expression is relative to our community.18 Therefore, it is an illusion to regard language as an autonomous and self-sufficient system, and it goes astray to investigate the relation between a language and the world without considering human beings.

(2) Language is conventional. That is to say, the fact that a language has become what it looks like now has no a priori necessary logic, but is the result of unconscious choices and conventions by a linguistic community.

Of course, such kinds of conventions are not established by way of negotiation or in the form of agreement or contract, but in a gradual process: when new names and expressions appear, some of them are unpopular and not accepted by our linguistic community, and eventually abandoned, whereas some others are popular and commonly used by our linguistic community. These commonly chosen names and expressions are unwritten ‘conventions’. Later, these accepted names and expressions are refined and revised, and enter into dictionaries or encyclopedias. The entries in dictionaries or encyclopedias become written ‘conventions’. However, even these written conventions can have some exceptions, and can be violated and changed.

More specifically, the referential relation between a name and its bearer must be traced back to the initial baptism of the object. In the causal chain of communication, the descriptive information about what a name designates is transferred from one person to another and from one generation to another; only informative descriptions of an object acknowledged by our linguistic community constitutes the meaning of the corresponding name. So, the meaning of a name reflects our consensus about the object to which the name refers. I will introduce some symbols to characterize the meaning of name N, which is the collection of descriptions of N’s bearer: let lowercase letters, a, b, c, d, e, f, g, h, i, j, k, … separately stand for a description of N’s bearer. Some descriptions are not accepted as true by our linguistic community, so they will not enter into the meaning of N; only those description of N’s bearer agreed by our linguistic community enter into the collection {a, b, c, d, e, f, …}. Since the collection illustrates the consensus of our linguistic community about the referent of N, and generally acknowledged by the community, so an operator for consensus can be put in the front of the collection as a superscript {a, b, c, d, e, f, … }. This kind of collection of descriptions determines what N designates. Of course, we could have some other collections of descriptions of N’s bearer by counterfactual imagination, for example, {–a,–b,–c,–d,–e, f, g, h, j, k, …}, {–a, b,–c,–d, e,–f, u, v, w, x, … }, in which ‘–a’ shows that a is absent, and so forth. The latter collections of descriptions of the referent of N do not constitute the meaning or even partial meaning of N, because they have not been acknowledged by our linguistic community. So we can’t use them to determine the object to which N refers, at least we can’t use them to identify the object to which we usually use N to refer.

Evans expresses some similar views:

… consideration of the phenomenon of a name’s getting a denotation, or changing it, suggests that there being a community of speakers using the name with such and such as the intended referent is likely to be a crucial constituent in these processes. With names as with other expressions in the language, what they signify depends on what we use them to signify. … There is something absurd in supposing that the intended referent of some perfectly ordinary use of a name by a speaker could be some item utterly isolated (causally) from the user’s community and culture simply in virtue of the fact that it fits better than anything else the cluster of descriptions he associates with the name.19 (Italics added)

(3) Language is animated. Because the world in front of us is changing, our cognition of the world is also changing. Our linguistic community adjusts language and its meaning to the needs of our practice and cognition. As a result, language and meaning are always in the process of change and growth.

Susan Haack has argued persuasively that meaning grows, which means not only words getting new meaning, but also losing old ones, and new words being invented to express new concepts and discriminations, some old expressions die or are abandoned, and even a whole language may become ‘dead’. Such changes may not be perceived in a short period, but in the long run they are evident and obvious. For instance, we can tell them by contrasting ancient English to modern English. She claims generally,

A natural language is an organic, living thing. Over the long haul a language may, like Latin, give birth to several different, new languages, and eventually fall into desuetude and die. And all natural languages slowly—and sometimes not so slowly—shift, change, and adapt: borrowing words from other languages and from the specialized jargon of scientists, soldiers, sailors, lawyers, bureaucrats, etc.; turning once-live metaphors to new purposes or domesticating them as comfortable clichés; sporting new idioms, buzzwords, slang, and catchphrases. I think the growth of meaning is much more significant than the recent philosophical mainstream acknowledges; but so far from being, as the radicals suppose, invariably a hindrance to rationality, it can contribute to the cognitive flexibility that rationality demands. 20

Based on this conception of language, I have developed a new theory of names—‘socio-historical causal descriptivism’ (abbreviated as SHCD)21. Obviously, it is beyond the space-limit of this paper to show all the details of SHCD. However, I will follow SHCD to challenge A1 hidden in the semantic argument of Kripke.

As is stated above, Kripke supposes that the question of ‘how does a name designate an object?’ only concerns the objective (or metaphysical) relation between language and the world, which has nothing to do with us as the users of the name. For him, a name, as a rigid designator, is a constant function which fixes its reference in all possible worlds without considering our intention, convention and custom to use the name. But for me, this assumption is absolutely wrong. I think that it is necessary for an agent to know what a word means before the word becomes a name; otherwise, he cannot distinguish a name from a pure noise. For instance, if I make a sound ‘soyola’, then, for the audience, is it a name? If it is, what does it designate? Even I myself may not know what it denotes because it is likely that I come up with a noise just for fun. Yet it is possible that someone happens to be called ‘Soyola’, but I don’t know. In another case, I speak out the sound ‘but’ or write the corresponding spelling, and then it is heard or seen by someone else. Is ‘but’ a name? Not necessarily. It may be the pronunciation of an English word, or the nickname of my best friend, or the name of my pet. Listeners will certainly ask me: ‘What is but? What do you mean by “but”?’ I would explain that ‘I am pronouncing the English word “but”’, or that ‘But is the nickname of my best friend’, or that ‘But is my dog’s name’, or, while pointing to an object, I say that ‘this is But’. Only after hearing my explanation can listeners determine whether ‘But’ is a name or not. Also, consider Kripke’s example of ‘Nancy’: I’d like to ask, how does the mathematician’s wife know that ‘nancy’ is a name rather than a noise made unconsciously by her husband, since he also pronouncing something like ‘haha’? Why ‘nancy’ is a name but ‘haha’ is not? What differences are there between ‘nancy’ and ‘haha’? I think that if an agent is completely ignorant of an object and cannot identify it ostensively, he or she could not regard any word as its name.22

There are two ways to explain the referential relation between a name and its bearer, that is, ostension and description. Ostension is to name an object through pointing to it. However, since a large number of objects are out of our horizon, the number of objects named by ostension is very small. Therefore, many words become names by way of description rather than ostension. When establishing the referential relation between a name and an object, we require at least some informative descriptions, consisting of a copula and a sortal, for example, ‘a is a newly discovered planet’, ‘b is a person’, ‘c is a dog’, ‘d is a painting’. Without this minimum informative description, ‘a’, ‘b’, ‘c’ and ‘d’ cannot become names for us. Kripke thinks that in order to guarantee that a name designates the same object in a causal chain, present speakers should be in accord with previous speakers in the respect of referential intention. I can ask some further questions: if we cannot identify the corresponding object by ostension, how could the accordance be guaranteed only by hearing some sound? What is transmitted on earth in the causal chain of a name? I think that these are serious problems to be treated by Kripke.

Moreover, the so-called ‘counterexamples’ of descriptivism given by Kripke in his semantic argument could be explained away.

Gödel/Schmidt. I can reply to Kripke as follows. Your fabricated story is not acknowledged by our linguistic community; your fancy about Gödel is not in the causal chain of the name ‘Gödel’. Therefore, we can still believe that the reference of ‘the man who discovered the incompleteness of arithmetic’ is the man Gödel rather than the man Schmidt. You have made a mistake. In addition, if your imagined situation is agreed by our linguistic community because of good evidence, perhaps we will cut off the connection between the name ‘Gödel’ and the description ‘the man who discovered the incompleteness of arithmetic’, and establish a new connection between the description and the name ‘Schmidt’. Perhaps, we will establish another relation between the name ‘Gödel’ and the new description ‘the notorious man who stole Schmidt’s proof of the incompleteness of arithmetic’.

Peano/Dedekind. My reply is similar to the above case. What is of great significance is not what Peano, Einstein and Columbus have actually done, but what is acknowledged by our linguistic community. Only those descriptions of the persons agreed by our linguistic community can be regarded as the part of the ‘official’ history of the persons and constitute the meanings or partial meanings of the relevant names. In contrast, those descriptions of the persons rejected by our linguistic community will be forgotten, or just become the topics of chat, gossip, or casual conversation at leisure time. We never consider those descriptions seriously.

Johna and Aristotle. As for the names of historical figures, such as ‘Johna’, ‘Aristotle’, ‘Cicero’ and ‘Confucius’, I think, what we really care about is various descriptions of their bearers in the classics or historical documents, since we cannot be acquainted with the figures and all our information about them come from the literature. What is of great significance for us is not who ‘Aristotle’ and ‘Confucius’ actually denote in history but who satisfy those descriptions of them. In some sense, what we really care about are the objects ‘constructed’ by those descriptions. As for the man who did not accomplish anything the Bible attributes to Moses, Kripke can give any name, even ‘Moses’, to him, but this ‘Moses’ is certainly not the Moses in the Bible. We only care about the Moses rather than ‘Moses’. As for the man who did not accomplish anything attributed to Aristotle in the literature, Kripke can give any name, even ‘Aristotle’, to him, but this ‘Aristotle’ is certainly not the Aristotle recorded in historical documents. We really care about the man Aristotle ‘living’ in those documents and in our cultural tradition. If later we find out new evidence which shows that our prior descriptions of a historical figure are totally wrong or inadequate, then the original descriptions should be revised or complemented, but these revisions or complements should also be agreed by our linguistic community and transferred down the causal chains of these names. Otherwise, these descriptions cannot constitute the meanings or partial meanings of the relevant names, or cannot be used to determine what the names designate. To a large degree, the meaning and reference of a name do not depend on what a particular person thinks about but on what our linguistic community acknowledges. The activities of giving the meaning and determining the referent of a name are social and historical.

3.2. Refutation of Assumption 2

I think that A2 is wrong because of the following reasons:

(1) Kripke misinterprets the most important principle of traditional descriptivism—‘sense determines reference’—as meaning that the sense of a name gives the necessary and sufficient condition for determining its reference, as if it assumed that a proper name can be exhaustively analyzed by some description(s). Searle clarifies the matter rightly:

… the issue is most emphatically not about whether proper names must be exhaustively analyzed in completely general terms. I do not know of any descriptivist theorist who has ever maintained that view, though Frege sometimes talks as if he might be sympathetic to it. In any case it has never been my view, nor, I believe, has it ever been the view of Strawson or Russell.23

In Searle’s view, what descriptivism actually claims is that when explaining how speakers determine the referent of a name, we need to show how an object satisfies the descriptive intentional contents in the speakers’ brains, which include speakers’ intention, descriptive features, Networks, and Backgrounds, related to the corresponding name.

Putnam also asserts:

I think, even Frege does not hold that the sense of a name is the necessary and sufficient condition for determining the referent of the name. His views of the sense of proper names can be summarized as follows.

First, the sense of a proper name is the associated mode of presentation of the object to which it refers. Only if a name has a sense could it refer to an object; which object a name denotes depends on whether or not the relevant object has the properties associated with the sense of the name. All this shows that the sense of a name affords the criterion or approach for identifying its reference. Conversely, the reference of a name does not determine its sense; identity of sense cannot be deduced from identity of reference, since different senses can determine the same referent. For example, the same triangles can be described as ‘equilateral triangles’ or as ‘equiangular triangles’.

Second, the sense of a proper name can be given by descriptions that represent the characteristics of the referent; moreover, the sense of the name can be given by different descriptions: i.e., there exist different explanations of the sense of a name by different people. Frege emphasizes:

So long as the Bedeutung remains the same, such variations of sense may be tolerated, although they are to be avoided in the theoretical structure of a demonstrative science and ought to not to occur in a perfect language.25

Third, because of the imperfection of natural language, a proper name may correspond to more than one sense (ambiguity), and there exist proper names which has sense but no referent, such as ‘Odysseus’, ‘the most distant celestial body from the earth’, ‘the least rapidly convergent series’. As Frege suggests, we can please ourselves so long as the same name has the same sense in the same context; when a name has no reference, we can stipulate artificially that it refers to 0 or the empty set.

According to Frege’s views, we know that: (a) The sense of a name is a sufficient condition for fixing its reference. That is to say, supposing a name has a sense, we could find its corresponding referent; if it has no such referent, let its referent be 0 or the empty set. In this way we can ensure that every name has a referent determined by its sense. (b) A single sense or partial sense of a name is not a necessary condition for fixing its referent. For Frege allows a name to have more than one sense, so long as the senses could determine their referents separately. This means that any single sense or partial sense of a name is not necessary for fixing its referent; even if the name lacks one of its senses, its reference could be determined by its other sense, so we can still say that sense determines reference. Therefore, when Kripke takes sense as the necessary and sufficient condition for fixing reference, he departs from or misunderstand Frege’s original position.