Meaning and Publicity E-Book

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

Seitenzahl: 485

Veröffentlichungsjahr: 2018

Beschreibung

The papers collected in this volume all discuss the ways and extent to which the determinants of meaning must be public. In the philosophy of language there are currently two main traditions concerning the relationship between meaning and public phenomena. According to one tradition language is public in principle, so that there can be nothing to the meaning of linguistic expressions that cannot be accounted for in terms of the behaviour in context of linguistic subjects. According to the other tradition linguistic meaning is determined by the content of the mental representations that are expressed in overt speech acts. On such views, the properties of the mental are prior to language and linguistic meaning should be explained by appeal to mental concepts. There divergent traditions leave us with a question: Is linguistic meaning to be explained on the basis of a pre-linguistic biological or mental capacity which "goes public " in overt speech, or is it to be explained on the basis of pubic behaviour in context which "goes private "in thought, and which determines the contents of the mental?

Leseprobe

CONTENTS

Introduction

Meaning and Publicity: Two Traditions

Richard N. Manning

P

ART

I

H

ISTORICAL

B

ACKGROUND

Speaking Your Mind: Expression in Locke’s Theory of Language

Lewis Powell

Meaning, Communication, and the Mental

Patrick Rysiew

Intentionality and Publicity

Madeleine L. Arseneault

P

ART

II

M

EANING AND

I

NTERPRETATION

Quine, Publicity, and Pre-Established Harmony

Gary Kemp

Reflections on Davidsonian Semantic Publicity

Richard N. Manning

Meaning, Publicity and Knowledge

Marija Jankovic and Greg Ray

P

ART

III

C

ONTEMPORARY

C

RITICISMS AND

D

EVELOPMENTS

A Puzzle about Context and Communicative Acts

Daniel W. Harris

The Publicity of Meaning and the Perceptual Approach to Speech Comprehension

Berit Brogaard

Local Meaning, Public Offense

Robert Shanklin

O

N

C

ONTEMPORARY

L

INGUISTICS AND

S

OCIOLOGY

Analyses on Arbitrariness of Chinese Characters from the Perspective of Morphology

Feng Li

Formal Semantics of English Sentences with Tense and Aspect

Wenyan Zhang

The Axial Age and Modernity: From Max Weber to Karl Jaspers and Shmuel Eisenstadt

Vittorio Cotesta

Contributors

Impressum

On ProtoSociology

Ordering

Published Volumes

Bookpublications of the Project

INTRODUCTIONMEANING AND PUBLICITY: TWO TRADITIONS

Richard N. Manning

The papers collected in this volume all discuss the ways and extent to which the determinants of meaning must be public. In the philosophy of language there are currently two main traditions concerning the relationship between meaning and public phenomena. According to one tradition, exemplified by figures such as Ludwig Wittgenstein, W. V. Quine, and Donald Davidson, language is public in principle, so that there can be nothing to the meaning of linguistic expressions that cannot be accounted for in terms of the behaviour in context of linguistic subjects. On this view, the determinants of meaning are public. According to the other tradition, exemplified by figures such as Noam Chomsky, Jerry Fodor, and Ernest Lepore, linguistic meaning is determined by the content of the mental representations that are expressed in overt speech acts. On such views, the properties of the mental are prior to language and linguistic meaning should be explained by appeal to mental concepts. There divergent traditions leave us with a question: Is linguistic meaning to be explained on the basis of a pre-linguistic biological or mental capacity which “goes public” in overt speech, or is it to be explained on the basis of pubic behaviour in context which “goes private” in thought, and which determines the contents of the mental?

There is of course no univocal answer to the question what is meant by “public”, or by “meaning”. What is meant by “public” will vary with both the sorts of motivations one has for claiming that meaning must be public and with the conception of meaning under consideration. The conception of meaning under consideration might itself be underwritten by antecedent considerations concerning the publicity of its determinants. There is no general answer to the question whether antecedent conceptions of meaning drive publicity requirements, or antecedent considerations underwriting a publicity requirement motivate taking a certain view of meaning, and there may be no answer in the case of any particular thinker either.

Be that as it may, I will try in this introduction to bring focus to this variegated terrain by articulating some of the considerations, many overlapping, that provide positive motivation for a publicity requirement on meaning. The reason for focusing on the positive is itself a concession to the negative: Substantial grounds for doubting that the determinants of meaning are public are not hard to find. For one thing, though we will see that the matter is complex, people evidently have access to their own thoughts in a way that is not mediated through experience of their own behaviour or of other public evidence, and this mode of access is unavailable to others. The mental is private in this sense. If speech involves the expression of thought and thought is in this way private, then it seems that the meaning of speech acts should likewise be determined by what is private. For another, in principle arguments seem to show that, if public evidence is the sole determinant of meaning, then meaning is radically underdetermined, and perhaps even indeterminate. This is counterintuitive, given our sense that there is something we understand in grasping a person’s words, and in light of the fact that, once again, language is expressive of thought, and surely our thoughts have determinate contents. For yet another, it seems, as an empirical matter, that language acquisition takes place under the condition of a poverty of stimuli; moreover, linguistic studies reveal that languages have substantially more grammatical structure than could possibly be the product of exposure to purely public stimuli. For these latter reasons, among others, the idea that the determinants of meaning must be public is simply not a recognized principle among empirical and theoretical linguists, and is antithetical to the dominant traditions of psycholingusitics and the cognitive psychological study of mental contents and the contents of speech episodes as the culmination of cognitive processes. In this light, the burden of proof seems to be on those who insist the determinants of meaning must be public. In the remainder of this introduction, I will survey some broad considerations, historical and perennial, that push in favour of publicity.

Historically, one such consideration derives from the early modern transition from the way of being to the way of ideas. Once philosophers starting supposing it was more or less mandatory to conceive the direct objects of our experience as ideas in the individual’s mind, which at best represent objects external to it, the question what grounds we could have for supposing those ideas were in fact about objects external to the mind, let alone adequate to them, became acute. So too becomes the question whether and how the words that express those ideas can refer to any external objects or properties they might have. The problem is especially pressing for empiricists, who deny there are any innate ideas, and thus suppose the content of all ideas is derived from experience, generally (if not in all cases, cf. Hume) in the form of sensation. Yet it also plagues those who accept innate ideas, since even they (again typically, cf. Leibniz) think some ideas are not innate and do concern an external world. The standard move here is to claim that what sensory ideas are about is the external objects that cause them, and that the words that are associated with ideas derive their intentional content – meaning – from the content of those ideas. Leaving aside the question whether the causal inference involved here could ever be sound, the upshot is a view on which, if our words are to be meaningful at all, what determines their meaning must be something outside the mind, and in that sense, public rather than private. Similar arguments in more modern guise have the same tendency. The rejection of “magical” theories of reference on which reference and intentionality are intrinsic features of representations (linguistic or otherwise), in favor of causal or relational view of reference, points in the direction of meanings that are fixed by objects that are in principle intersubjectively accessible rather than merely private and subjective. If the meaning of an expression is, or is dependent upon, its referent, then extrinsic, hence at least potentially public, factors determine meaning.

Another consideration in favor of publicity derives from the idea that, if language is to be a vehicle of communication, it must be possible for one language user to convey her thoughts to another by means of language. On the reasonable assumptions that the linguistic expression used in some way expresses the speaker’s thought and that a linguistic expression expresses its meaning, it follows that, for linguistic communication to be possible, meanings must be shared, and, in that sense, public rather than private. This line of thinking, closely related to Frege’s anti-psychologism, fixes on the intersubjectivity of meaning. Meanings reified into abstract objects, knowable through intuition (rather than understood as empirically accessible), might satisfy any publicity demand underwritten by this concern alone. But materialists and nominalists who reject reified abstracta seem compelled to account for communication by taking the determinants of meaning to be intersubjectively accessible in a shared external environment.

Another motivation for adopting a publicity requirement stems from externalist arguments, such as Putnam’s, that press intuitions about meaning to support the view that meaning “ain’t in the head”. Twin Earth-style thought experiments and consideration of the division of linguistic labor suggest that the determinants of linguistic meaning are not narrowly psychological, and are thus public in the sense of non-psychological. However, such considerations are consistent with the idea that, though external and not private, the determinants of current meaning may be hidden (like molecular structure), and hence, though not private, in no clear way public either. They may also be known to only a few experts, and thus not fully public in a broader sense: Public facts that determine what a person’s utterance means may be, though not in principle inaccessible, hidden from her, perhaps inevitably. This has obvious consequences that run counter to intuitions about first-person authority, over both the contents of one’s thoughts and the meanings of one’s utterances.

Considerations of first-person authority can in fact both pull for publicity and push against it. If we view the mental as interior and private, and as that to which the individual subject has immediate access in a way that she does not have access to what is external, then it seems that knowledge of the contents of one’s own mind, and hence of the meanings of the words one uses to express those contents, cannot depend on what is essentially public. We have already seen the difficulties this poses for the possibility of communication and the skeptical and solipsistic worries it invites. Yet in the modern, largely materialist context, the idea of first-person authority also pushes against the idea that what determines meaning is publically inaccessible. For on the materialist view, the contrast between the private and the public is the contrast between the inner and the outer – between what is within the subject’s body and what is on or outside her skin. And it seems obvious that we know what is outside our skins better than we do what is inside. Knowledge of our thoughts and meanings seems immediate, whereas knowledge of what’s in our heads requires surgical or scanning techniques (which techniques publicize those inner goings-on). So if what determines meaning must be material, then meaning must depend upon material facts that are, because public, the more evident.

A different kind of consideration animates Wittgensteinian arguments that claim that the normativity of language depends on the publicity of meaning. For an expression to be used correctly, it must be possible in principle for it to be used incorrectly. But for familiar reasons concerning rule following, it seems that this possibility cannot be made intelligible if meanings are determined privately. For if meaning is determined privately there would seem to be nothing that could mark the difference between a speaker’s using a term in accord with its meaning and the speaker’s merely thinking she had done so. This seems to require a criterion for correct usage, and thus a determinant of meaning, that is public in the sense of being independent of the speaker’s own self understanding or pattern of use.

Logical behaviorist and verificationist accounts of meaning also favor publicity. The behavioral dispositions in terms of which mental phenomena are to be analyzed by logical behaviorism leave nothing but behavior situated in public context to determine meaning. Similarly, where meaning is means of verification, and verification conditions are taken to be predictions of observations, the determinants of meanings are as public as whatever can be observed. Logical behaviorism and positivist style verificationism are very close kin, each animated by a desire to avoid the supposed abuses of metaphysics by making sure that whatever meaning linguistic expressions might have is traceable back to the directly observable, hence public. Rylean behaviorism, rather more subtly, seeks to avoid a whole untenable picture of the relation of the mental to the physical that results from thinking of the mind as itself a mechanism like the body, albeit a ghostly, non-physical one. Naïve logical behaviorist and verificationist accounts of meaning have long been discredited. But more sophisticated descendants of these views persist. Here, the antimetaphysical roots of both behaviorism and verificationism should be born in mind. The anti-metaphysical, therapeutic aims of verificationist and behaviorist thinking about meaning and the mental survive in Wittgensteinian accounts and generally in theories of meaning as use. They likewise survive in other neo-pragmatist accounts of meaning, exemplified in different ways by thinkers like Quine, Davidson, and Brandom, each of whom shows a hostility to the idea that meaning facts can depend on matters that are inner and hidden. This hostility itself owes in significant part to the supposed link between this idea and perennial philosophical problems of skepticism, solipsism, idealism, as well as those associated with hypostasizing meaning entities.

Another approach to the question of meaning derives from a concern with language acquisition. If we think of language acquisition as language learning, and if we think of learning as a matter of either stimulus-response training or of the development of a theory of what expressions mean on the basis of evidence of the conditions under which they are used, language acquisition will depend on facts about stimulus and use conditions, which must be in some sense public. As noted above, considerations of the poverty of the stimulus, and of how fantastically quickly children acquire whole swaths of language, seem to discredit the idea that language acquisition is a matter of learning, conceived either as stimulus-response conditioning or theory building. However, if we turn our attention away from empirical facts about language learning and towards conceptions of what linguistic capacity consists in, the motivation for a publicity requirement may survive. Whether and to what extent the determinants of meaning must be public will then depend on one’s view of what that capacity consists in. Here we can contrast the capacity conceived, on the one hand, as an ability to distinguish grammatically well-formed expressions from ill formed ones, and on the other, as the ability to predict and coordinate actions on the basis of our own and others’ linguistic utterances in public context. A theory accounting for the first capacity may perhaps make no reference to public determinants of meaning. But it is harder to see how a theory of meaning designed to account for the second capacity could fail to tie meanings to public phenomena characterizing conditions of use.

Reflection on a pair of influential and competing methodological principles can provide a final, expansive perspective on the general topography of the terrain at stake here. Both arise from a rejection of the excesses of early twentieth century behaviorism and logical empiricism. Each of these, as we saw, pushes towards publicity by showing a hostility to the idea that meaning facts could depend on what either inevitably private or merely inner and thus contingently hidden. Realizing that this attitude towards meaning misunderstands and inappropriately fetters scientific practice, materialist and scientific minded philosophers sought to revitalize the idea that psychological and semantic notions could be usefully accounted for by means of the postulation (and eventual uncovering) of inner mechanisms, which, however in principle public they may be, are certainly not publicly accessible to ordinary language users in any ordinary sense.

But it is crucial just how we think of scientific postulation about the inner mechanisms involved in thought and language. One way is to think of inner states and processes that can be correlated with, and perhaps be identified as causing and being caused by, the subject’s environmental conditions. Fodor’s principle of methodological solipsism arises from the realization that a variety of distal stimuli can cause or correlate with the same inner states and processes, along with the conviction that the relevant subject of scientific study is the individual human subject. According to this principle, we should conceive the environmental correlates, causes and effects of the mechanisms as being bounded by the skin of the human subject. For the methodological solipsist, the only scientifically respectable kind of content is so-called narrow content, and the determinants of narrow content are not public, if that means external and evident. Moreover, the pressure to think of the inner correlates as having representational contents is the apparent compositional structure of the efferent and afferent impulses with which they correlate, which seems to call for treating the inner structures as themselves having a syntax. Methodological solipsism thus adopts Quine’s early conception of stimuli, but rejects his narrowly behaviorist methodology as well as his insistence that language is a social art. A scientific approach to meaning, on the methodological solipsist’s view, concerns what goes on in the subject’s body, just as a does a scientific approach to neuro-anatomy. What cannot be found through such study has no scientific status.

A different methodological stricture, itself adopted in the course of making way for an explanation of the mental in scientific style, is found in Sellars’s methodological behaviorism. The idea here is that, while science may legitimately posit inner entities to serve as thoughts which are expressed in overt linguistic expression, the evidence on the basis of which such inner episodes are to be posited and upon which they and their intentional features are to be modeled is overt linguistic expressions understood as already semantically significant. Here the traditional picture of meaningful language as the outer expression of inner intentional content is reversed. Now the intentionality of thoughts as inner episodes is derivative on that of overt language. But if thought gets its intentional content from overt language, then the determinants of the meanings of the overt linguistic expressions themselves cannot be inner. How exactly overt linguistic episodes come to have originary meaning is of course a complex and vexed matter. But for the methodological behaviorist, the determinants of such meaning are out in the shared world.

The motivations for adopting methodological behaviorism, and thus the motivation for adopting a corresponding public view of meaning, are large scale philosophical ones, rather than piecemeal and empirical. Sellars himself was motivated, among other things, by a desire to avoid falling into the Myth of the Given (episodes of awareness of kinds that are prior to and explanatory of conceptual and linguistic capacities), while at the same time to respect the idea of the inner, both scientifically and in traditional philosophical of privileged access. As we have seen, other motivations for publicity are derived from similarly large philosophical concerns. Attention to these big picture issues can, I hope, help to frame the more particular and detailed perspectives presented in the papers contained in this volume. Much exploration and detailed mapping of the interrelations between ways of thinking about meaning and ways of thinking about the public remains to be done. The essays printed here contribute to the investigation of this important philosophical space.

Part I

Historical Background

SPEAKING YOUR MIND: EXPRESSION IN LOCKE’S THEORY OF LANGUAGE

Lewis Powell

Abstract

There is a tension between John Locke’s awareness of the fundamental importance of a shared public language and the manner in which his theorizing appears limited to offering a psychologistic account of the idiolects of individual speakers. I argue that a correct understanding of Locke’s central notion of signification can resolve this tension. I start by examining a long standing objection to Locke’s view, according to which his theory of meaning systematically gets the subject matter of our discourse wrong, by making our ideas the meanings of our words. By examining Locke’s definition of “truth”, I show that Lockean signification is an expression relation, rather than a descriptive or referential relation. Consequently, the sense in which our words signify our ideas is roughly that our utterances advertise our otherwise undisclosed mental lives to each other. While this resolves one aspect of the public/private tension, I close with a brief discussion of the remaining tension, and the role for normative constraints on signification to play in generating a genuinely shared public language.

Introduction

John Locke opens book three of his Essay Concerning Human Understanding by highlighting the social role of language. It is because we are social creatures, Locke tells us, that we require language. At the same time, the actual account of the workings of language that he offers is hyper-individualized and based in the psychologies of particular speakers. Locke explains the importance to humans of having a shared language, but, in essence, offers a theory on which each speaker has their own idiolect. So, there is a prima facie tension between Locke’s view of language’s fundamentally social purpose and his account of its fundamentally individualistic mechanisms. In this paper, I show how Locke resolves this tension between the social purpose of language and its individualistic mechanisms.

One of the most common concerns about Locke’s theory, from his own day forward, was the objection that his individualistic, psychologistic account of the meaning of language winds up getting the subject matter of our discourse wrong. Locke has long been accused of incorrectly maintaining that when we speak, we are always talking about our own ideas and mental states, rather than ever talking about objects and qualities in the external world. If correct, this objection is a major concern for Locke’s philosophy of language, as it is untenable to offer an account of language on which all of our assertions are making claims about the contents of our own minds. Fortunately for Locke, this objection turns on a misunderstanding of his views.

In the first section of this paper, I present this “wrong subject matter” objection, found in the writing of Locke’s contemporary John Sergeant, which was also offered later by John Stuart Mill, and which continues to be raised as an objection to Locke’s account. In section two, I show how a careful reading of Locke’s claims about truth show that the objection is misplaced. Locke’s theory of language does not render our own minds the constant and essential subject matter of our discourse. Instead, Locke’s theory presages an expressivist approach to thinking about language. In the third section of the paper, I turn to a remaining set of worries about how Locke’s individualistic psychological account of meaning leaves him with a theory of individual idiolects, and no theory of a common public language. While some thinkers could rest content without accounting for a shared language, Locke’s concern for the social nature of communication seem to require him to address this worry. In the conclusion of the paper, I articulate Locke’s normative approach to resolving the idiolect problem without positing a genuine public language.

Section 1: The Wrong Subject Matter Objection

John Locke is one of the few figures of the early modern period to offer an explicit, systematic treatment of the workings of language. The entirety of book three of the Essay concerns his account of language, and the very opening of his discussion helps to establish the outlook and orientation of his theorizing:

§1. god having designed Man for a sociable Creature, made him not only with an inclination, and under a necessity to have fellowship with those of his own kind; but furnished him also with Language, which was to be the great Instrument, and common Tye of Society. Man therefore had by Nature his Organs so fashioned as to be fit to frame articulate Sounds, which we call Words. But this was not enough yet to produce Language; for Parrots, and several other Birds, will be taught to make articulate Sounds distinct enough, which yet, by no means, are capable of Language.

§2. Besides articulate Sounds therefore, it was farther necessary, that he should be able to use these Sounds, as Signs of internal Conceptions; and to make them stand as marks for the Ideas within his own mind, whereby they might be made known to others, and the Thoughts of Men’s Mind be conveyed from one to another. (Essay, 3.1.1–2)

In these two sections Locke establishes some of the most integral and core commitments of his account of language, including both the view that the purpose of language is to allow humans to enter into proper social relationships with one another, as well as the view that the mechanism by which language functions is the use of articulate sounds as outward manifestations of one’s thoughts.

According to Locke, the contents and events of one person’s mind are obscured from others. But humans are social creatures, and so, cannot flourish in isolation from each other. The primary purpose of language, then, is to act as antidote to our natural state of social isolation, by permitting us to actually share the goings on of our inner lives with each other. Locke then proposes that the way language serves to do this is for an individual’s utterances to serve as signs of their thoughts, permitting them to create outward manifestations of their inner lives.

Stated this way, Locke’s basic picture has a lot of prima facie appeal. However, because of Locke’s focus on the social role of language in allowing us to share our mental lives with each other, his primary semantic relationship is one that obtains between words or utterances and an individual speaker’s thoughts and ideas. This feature has led to some fairly longstanding criticisms of Locke’s views.

To outline the basics of Locke’s view: a term like “Elizabeth” will signify your idea ELIZABETH.1 And a term like “human” will signify your idea human. In general, substantive terms will signify ideas. A word like “is” (or “is not”) on the other hand, will signify an act of the mind, like affirming (or denying).2 So the sentence “Elizabeth is human” signifies affirming human of ELIZABETH. A slightly less cumbersome way to phrase this would be to say that the sentence “Elizabeth is human” signifies the judgment/belief that Elizabeth is human. In Locke’s own day, John Sergeant offered what I term the “wrong subject matter” objection against his idea-based theory of meaning, with a similar objection offered later by Mill, and still offered casually today.3 Here is Sergeant’s presentation of the objection:

16. It may be perhaps replied, that the Ideas are only meant by the Words; because when we speak, we intend [Note:Proof 10th. Because when the thing it self is intended to be made known, the Thing it self is the first meaning, or what is first meant by the words. ] to signify our Thoughts. I answer, that, however it may be pretended that what is meant immediately by the words, is our Thoughts, when our own Thoughts or Judgments about any matter, are the things desir’d to be known; yet, when the Things are the Objects enquired after, as, when a Master teaches a Scholler Natural Philosophy, or any other Truth, the Intention of the Speakers does primarily aym and mean to signify the Things or Truths themselves; and not our Thoughts concerning them; and, therefore, the Things themselves are in the Intention and Mind, or are the Meanings of the Speakers, or Discoursers. And this passes generally in all other occasions, except only when the Knowledge of our Interiour Thoughts is ultimately aymed at. Thus, when a Gentleman bids his Servant fetch him a Pint of Wine; he does not mean to bid him fetch the Idea of Wine in his own head, but the Wine it self which is in the Cellar; and the same holds in all our Commerce and Conversation about things without us. (Sergeant 1697, Preliminary Second [emphasis added])

Sergeant’s objection, despite it’s now-archaic presentation, is clear. When I say “get me the wine” I am asking for someone to retrieve actual wine, not to produce the idea of wine in my mind. Generally when I use the word “wine”, unless I am specifically trying to talk about my interior mental life, I am talking about wine itself, not my idea of wine. Here is a slightly more formal presentation of this objection:

Wrong Subject Matter (WSM) Objection:

If “Elizabeth” signifies

ELIZABETH

, then the sentence “Elizabeth is human” is about

ELIZABETH

, not Elizabeth herself.

The sentence “Elizabeth is human” is about Elizabeth herself, not

ELIZ

ABETH

.

So, it is not the case that “Elizabeth” signifies

ELIZABETH

.

Let us grant premise (2), as it seems extremely plausible. So, the strength of this objection to Locke’s view hinges on the strength of premise (1). The WSM objection maintains that “Elizabeth” signifying the idea ELIZABETH means that utterances involving the term are thereby about the idea. And this point relies on the assumption that signification and aboutness will go hand in hand: if the things signified by a sentence are the things the sentence is about, it would be impossible for the Lockean view to escape this objection. So behind premise (1) is something like the principle that signification implies aboutness, and for individual words, that would be something like the principle that they refer to the things they signify.

Locke doesn’t give us an explicit account of aboutness, making it harder to identify directly whether this is a principle Locke would subscribe to. However, there is still good reason to find this assumption questionable, and it has been questioned by a number of Locke scholars. Norman Kretzmann (1968) has argued that ideas on Locke’s theory of meaning function more like Fregean senses than referents (and thus, would not necessarily go hand in hand with “aboutness”). EJ Ashworth (1981), Paul Guyer (1994) and Walter Ott (2003) have also argued that that the sense of signification which Locke is using, inherited from ancient, medieval, or other early modern sources, does not track reference (and thus would not track aboutness). One notable exception is Michael Losonsky (2007), who grants that, for Locke, names refer to ideas, but attempts to defend the plausibility of such a view.

My own interpretation differs in some important ways from those of Kretzmann, Ashworth, Guyer, and Ott but, importantly, aligns on this issue: that a term signifies something, for Locke, does not mean the term refers to that thing, or that utterances involving the term are automatically about the thing signified.

So, the plan is to avoid the objection by denying premise (1). However, in order for this to be a successful reply to the objection, it is important to provide a sustainable account of how Locke avoids commitment to (1). And here, it is helpful to outline a positive account of Lockean signification.

Section 2: Locke’s Expressivist Account of Meaning And Assertion

In this section, I will show how Locke articulates what I term a sincerityconditional semantic theory. Many contemporary semantic theories we are familiar with are primarily in the business of providing truth conditions for sentences. The meaning assigned to a sentence on such accounts may be the truth conditions themselves, or a proposition which maps directly on to some truth conditions, or the like. If one is accustomed to such a perspective, it is natural to look at what Locke assigns as the meanings of sentences, and attribute to him a confusion about the subject matter of our discourse. Against that background perspective, if someone proposes that the meaning of the sentence “Elizabeth is human” is the belief that Elizabeth is human, rather than the state of affairs of Elizabeth being human, or a third-realm proposition that Elizabeth is human, or a function from worlds where Elizabeth is human to the value “true”, it is natural to think that Locke’s account of the meaning of the sentence will get the result that the sentence is true provided the speaker has that belief, rather than it being true provided that Elizabeth is a human being. But this way of interpreting his view deeply misunderstands the relationship Locke sees between language and mind.

For Locke, the role of language is to help us advertise our otherwise undisclosed mental states to others. So when someone utters the sentence “Elizabeth is a human” the thing that Locke’s account treats as having primary importance is the relationship between the speaker and the psychological state they are attempting to put on display. The WSM objection effectively assumes that Locke’s account does so by making the sentence describe the speaker’s mental state. But, describing and displaying are different. Their difference is similar to the difference between:

1: I am feeling pain.

2: Ouch!

These two sentences are arguably related to the same mental state (a first personal pain sensation), but they differ in that the former straightforwardly describes oneself as having that mental state, while the other one seems to be an utterance we can make sense of because, typically, the person who (sincerely) says it is in that mental state. But they are not describing themselves as being in that mental state by saying “ouch”, they are merely allowing it to have an outward manifestation.4

So now, compare these two sentences, for Locke:

3: I believe that Elizabeth is a human.

4: Elizabeth is a human.

The defense of Locke here is that (3) and (4), like (1) and (2), can be related to the same mental state (a belief that Elizabeth is a human). But, whereas (3) describes that mental state, (4) displays, manifests, or expresses it. In other words, when someone utters (4), they have done so in order to advertise being in that mental state. They are not talking about that mental state, they are, instead, giving voice to that mental state.5

Locke’s fundamental semantic interest isn’t in the descriptive facet of language, it is in language’s power to advertise or display what is going on in our head. Locke’s account of the meaning of a sentence doesn’t map a sentence to its truth conditions, rather it maps the sentence to the mental state that one is advertising through uttering that sentence, which I’ll term the sentence’s sincerity conditions.

If we think about the meanings of the sentences as their sincerity conditions, we can see why having the meaning of a given sentence be a judgment does not automatically involve getting the subject matter wrong. Instead, Locke is taking a stance on a different relationship than the one that Sergeant seems to be concerned with.6

Showing that the utterance isn’t about my own judgment is one thing, but it is a further step to show how Locke’s theory can secure the result that the utterance is in fact about Elizabeth. As mentioned, Locke doesn’t offer an account of aboutness. He does, however, offer us a reasonably robust discussion of truth. And for the relevant issue here, that serves our purposes well. We can see whether Locke delivers on premise (2) by seeing what his account of truth would say about the truth conditions for a sentence like “Elizabeth is human”.

Here is Locke’s general definition of truth:

§2 Truth then seems to me, in the proper import of the Word, to signify nothing but the joining or separating of Signs, as the Things signified by them, do agree or disagree one with another. The joining or separating of signs here meant is what by another name, we call Proposition. So that Truth properly belongs only to Propositions: whereof there are two sorts, viz. Mental and Verbal; as there are two sorts of Signs commonly made use of, viz. Ideas and Words. (Essay, 4.5.2)

Here we get a sort of correspondence account of truth. A joining (or separating) of signs – a proposition – is true just in case the the things signified by the signs agree (or disagree) with one another. So, propositions join (or separate) signs, and the things signified by those signs stand in relations of agreement (and disagreement). If the signs are joined, and the things signified agree, the proposition is true. And, if the signs are separated, and the things signified disagree, then the proposition is true. Otherwise, the proposition will not be true. Importantly, Locke distinguishes between two types of propositions here, mental and verbal. As Locke clarifies in 4.5.5, mental propositions are composed of ideas, and (many) ideas signify objects in the world, while verbal propositions, on the other hand, are composed of words and words signify ideas. So far all that has been said, there is still room for the Sergeant/Mill worry: my relevant beliefs may be directly about the world, but my utterances might not turn out to be about the world (they may even still turn out to be about my beliefs). However, looking to Locke’s further elaboration on these two types of truth, we can see that this is not the case:

§6 [ … ] When Ideas are so put together, or separated in the Mind, as they, or the Things they stand for do agree, or not, that is, as I may call it, mental Truth. But Truth of Words is something more, and that is the affirming or denying of Words one of another, as the Ideas they stand for agree or disagree: And this again is twofold. Either purely Verbal, and trifling, which I shall speak of, or Real and instructive; which is the Object of that real Knowledge, which we have spoken of already. (Essay, 4.5.6)

Locke’s treatment of mental truth here is very brief. A judgment is mentally true just in case the ideas joined (separated) in it, or the things they stand for, agree (disagree).7 As to the truth of words, Locke tells us that it is “the affirming or denying of Words one of another as the Ideas they stand for agree or disagree”. Here we see that the account for an utterance like “Elizabeth is human” will affirm “human” of “Elizabeth” which will then be true provided that our idea HUMAN agrees with our idea ELIZABETH.

What is crucial to appreciate here about Locke’s account is that there are two relevant pairs of concepts being deployed: joining/separating, and agreement/disagreement. For Locke, a verbal proposition is true when the verbal signs are joined (separated) in parallel with the agreement (disagreement) of the ideas signified, not when words are joined (separated) in parallel with the joining (separating) of the ideas signified. So, as I argued above, an utterance of “Elizabeth is human” isn’t made true by the speaker having the belief that Elizabeth is human. Instead, it is made true by some concordance between the speaker’s idea ELIZABETH and their idea human.8 In fact, a few section later (4.5.11), Locke defines a concept of “moral truth” (not in the sense of the truths of morality, but rather a morally relevant definition of truth) as “speaking Things according to the perswasion of our own Minds, though the Proposition we speak agree not to the reality of Things”. This is a definition of what we’d more likely call honesty, and Locke specifically distinguishes it from his account of “the truth of words”. One way to put my proposal is that the core of Locke’s account of language tells you what it takes to utter a sentence honestly, rather than what it takes for an utterance of it to be true. The truth conditions will be a secondary feature, derived from this primary assignment of sincerity conditions.9

To summarize: “Elizabeth is human” is a verbal proposition, composed of “Elizabeth” signifying the idea ELIZABETH, “human” signifying the idea human, and “is” signifying the act of affirmation or the joining of those two ideas. The whole thing signifies a judgment joining human and ELIZABETH. But for “Elizabeth is human” to be true, requires that HUMAN and ELIZABETH agree with each other, not that they have been joined by the speaker. And for HUMAN and ELIZABETH to agree with each other requires that the things they stand for agree with each other (namely Elizabeth and the astra idea HUMAN). Setting aside whether Locke winds up having the right truth conditions for the belief that Elizabeth is human, the general structure of his account of verbal truth seems correct, here: the truth conditions for the sentence “Elizabeth is human” and the judgement that Elizabeth is human wind up being the same, systematically. And this seems right. The truth conditions of your utterances should match the beliefs you are displaying by making such utterances.10 And this, it is clear, his account captures.

However, there is one remaining related concern, which will be the focus of the next section. So far, we’ve been ignoring the fact that my idea ELIZABETH and your idea ELIZABETH might be different in their particulars. Locke talks explicitly about people having different qualities built in to their idea of gold or of swans, and the same thing will go for ideas of individual people, or really, any of our ideas. At first glance, this means we would all be speaking subtly different languages, and talking past each other constantly. If it is part of my idea of Elizabeth that she was 30 feet tall, then my utterance of “Elizabeth is 30 feet tall” would turn out to be true, rather than false. And that seems like the wrong result.

Section 3: The Problem of Idiolects

Locke’s theory, as it has been so far laid out, looks like it fundamentally offers an account of individual idiolects.11 In fact, Locke even implicitly recognizes this when describing our ideas of swans as an example of complex ideas of substances (emphasis added):

Thus the Idea which an English-man signifies by the Name Swan is white Colour, long Neck, red Beak, black Legs, and whole Feet, and all these of a certain size, with a power of swimming in the Water, and making a certain kind of Noise, and, perhaps, to a Man, who has long observed those kind of Birds, some other Properties, which all terminate in sensible simple Ideas, all united in one common subject. (Essay, 2.24.14)

We can use this case to lay out the concern for Locke’s view. Consider four individuals, First, Locke’s standard Englishman (the layperson), and his swan observer (the expert). Let’s also consider someone with an underdeveloped views of swans (the novice) and someone with mistaken views of swans (the bungler). So, the idea that the first signifies by the term “swan” is composed of the qualities enumerated by Locke in the passage above. The idea that the second signifies by the term is composed of all of those, plus some extra qualities common to swans (e.g. that they are largely herbivorous). The idea the third signifies by the term is composed of a partial subset of the qualities Locke enumerated (omitting the redness of the beak). And the idea the the fourth signifies by the term includes all those enumerated by Locke, and also some qualities not actually possessed by swans (e.g. that they can breathe underwater).

Let’s examine what the view of truth outlined in the last section seems to commit Locke to for utterances of some swan related sentences by these four:

S1. Swans are white.

S2. Swans have red beaks.

S3. Swans are mostly herbivorous.

S4. Swans are able to breathe underwater.

It seems that regardless of which of the four utters (S1), it will turn out true. All of them signify an idea by “swan” that agrees with the idea WHITE. The layperson, the expert, and even the bungler would also clearly speak truly when they utter (S2), though we don’t yet know what Locke would say about the novice’s utterance here, because while the sort of agreement between ideas that is involved when the one idea is partially composing the other is clearly sufficient, we haven’t established what else could be sufficient for the ideas to agree. Similarly, the expert’s uttering (S3) is clearly true on the account we have so far, but it is as yet unclear how to assess whether anyone else speaks truly in uttering it. And lastly, it seems like Locke is committed to saying the bungler speaks truly when uttering (S4), and though we need to know more to decisively assess this, it is unlikely that any of the other three would speak truly by uttering (S4).

However, even merely allowing that the bungler can utter (S4) truly is sufficient to raise this worry. Locke’s view is obviously going to be a non-starter if the swan expert speaks truly in saying “Swans are able to breathe underwater”. But that would mean that when the bungler utters that sentence, and the expert says “Swans are not able to breathe underwater”, the two are talking past each other, because they are effectively speaking different languages. The bungler is saying a true thing about a fictional type of bird, while the expert is denying a false claim about an actual species.

An ideal treatment would have room for error in addition to ignorance. We want a view on which all four people are speaking the same language, and so, a view on which the truth-value of the four sentences above doesn’t depend on which of the four is speaking. And there is a special challenge for Locke in offering such a view, because of Locke’s specific focus on language’s role for displaying the contents of our minds to others. So, prima facie, the utterances of the bungler should be signifying the idea he possesses.

Interestingly, Locke has the resources to develop a view on which he could offer a reply that maintains that the bungler both signifies his bungled idea and that he utters something false when he speaks, though it is not clear that he ultimately offers such a reply. Towards the end of the second book of the Essay Locke outlines several pairs of contrasting attributes which can be (in some sense) applied to ideas.12 Significant for our purposes is the contrast he draws between real and fantastical ideas, and the discussion of true and false ideas.

With respect to the first, the bungler’s idea is fantastical, rather than real, because there is no external archetype for his idea of a long-necked, red-beaked ( … ) white bird that breathes water. Locke would say that, at best, the Bungler is believing (and thus) saying something on a par with claims about centaurs and gorgons. But this is not good enough for our purposes, as it does nothing to render anything about the claim false.

At the same time, it is helpful in explaining how Locke can get the intuitively correct results for the novice’s utterance of (S2) and both the novice and layperson’s utterances of (S3). Because the other three all do have ideas with an external archetype, Locke can appeal to the actual co-existence of red-beakedness and herbivorousness with the other qualities that are in their ideas of swans to explain why their swan ideas in fact agree with the ideas of herbivorousness and red-beakedness.13

However, this real/fantastical distinction will only help with ideas of substances (swans, humans). If the dispute concerns ideas of modes, which don’t require external archetypes, there would be no distinction to be drawn, as the direction of fit is reversed. So, Locke’s distinction between true and false ideas is much more helpful for a general response to the worry:

§1 THOUGH Truth and Falshood belong, in Propriety of Speech, only to Propositions; yet Ideas are oftentimes termed true or false (as what Words are there, that are not used with great Latitude, and with some deviation from their strict and proper Significations?) Though, I think, that when Ideas themselves are termed true or false, there is still some secret or tacit Proposition, which is the Foundation of that Denomination: as we shall see, if we examine the particular Occasions, wherein they come to be called true or false. (Essay, 2.32.1)

Ideas are never, strictly speaking, true or false. But we do sometimes label them true or false, and Locke here claims that this is done only when there is an accompanying proposition or judgment which is itself genuinely false. Locke goes on to outline the two primary ways in which ideas are rendered false in this derivative sense:

§4 When-ever the Mind refers any of its Ideas to any thing extraneous to them, they are then capable to be called true or false. Because the Mind in such a reference, makes a tacit Supposition of their Conformity to that Thing: which Supposition, as it happens to be true or false; so the Ideas themselves come to be denominated. The most usual Cases wherein this happens, are these following:

§5 First, When the Mind supposes any Idea it has, conformable to that in other Men’s Minds called by the same common Name; v.g. when the Mind intends, or judges its Ideas of Justice, Temperance, Religion, to be the same, with what other Men give those Names to. Secondly, When the Mind supposes any Idea it has in it self, to be conformable to some real Existence. Thus the two Ideas, of a Man, and a Centaur, supposed to be the Ideas of real Substances, are the one true, and the other false; the one having a Conformity to what has really existed; the other not. Thirdly, When the Mind refers any of its Ideas to that real Constitution, and Essence of any thing, whereon all its Properties depend: and thus the greatest part, if not all our Ideas of Substances, are false. (Essay, 2.32.4–5)

Here, Locke tells us that the main judgments that accompany ideas permitting them to become (derivatively) true or false are (a) when one assumes that the idea they signify by a name is conformable to the idea that other people signify by that name, (b) when one assumes that the idea in question is conformable to some really really existing thing, and (c) when one assumes that the idea captures the things real essence. This last one is largely irrelevant for our purposes here, but the first two both seem applicable to the case of the bungler: the bungler is tacitly assuming that their idea swan corresponds to other people’s idea swan, and that their idea SWAN corresponds to certain actual long-necked white birds out in the world.

Locke’s take on the bungler’s utterance then is that the bungler signifies their judgment that breathe underwater agrees with their ideaSWAN, and that that judgment is, in isolation, true, but their idea SWAN is doubly false, neither corresponding to other people’s use of “swan” nor to reality. Locke doesn’t tell us that these tacit judgments are in fact conveyed, signified or displayed by the bungler’s utterance of the sentence. That view would roughly amount to the idea that most utterances express a conjunction of beliefs, including a belief that the words you just used signify corresponding ideas to the ones other people used as well as a belief that your ideas correspond to things out in the world. And so the utterance by the bungler would be overall false because it amounts to a conjunction of one true (irrelevant, fantastical belief) and two false beliefs about their idea SWAN. Instead, Locke seems to take the stand that the bungler’s utterance was true in his idiolect, but that there are some important/relevant false beliefs involved in the bungler’s overall situation.

So, as it has been presented here, Locke’s view accepts the counterintuitive consequence that the bungler has said something true in this case, but attempts to mitigate that consequence by pointing out the related falsities in the beliefs of the speaker. But if the bungler simply didn’t have views about what other people’s ideas of swans were like (or vandalism, if we want to use an example of a mode), they’d escape the falsity Locke is able to identify here. But if we are each speaking our own individual languages, how can Locke offer any criticism of a speaker like Lewis Carroll’s Humpty Dumpty,14 who intentionally uses words in a manner indifferent to their commonly accepted meanings?

Conclusion: A Lockean Public Language?

William Lycan (2008) provides a nice statement of just this concern as an objection to ideational theories of meaning:

Meaning is a public, intersubjective, social phenomenon. An English word has the meaning it does for the entire community of English speakers, even if some members of that community happen not to understand that word. Butideas, images, and feelings in the mind are not intersubjective in that way; they are subjective, held only in the minds of individual persons, and they differ from person to person depending on one’s total mental state and background.Therefore, meanings are not ideas in the mind. (Lycan 2008, p. 68)

Here, I think the best way to understand Locke’s avenue for reply involves recognizing that the social purpose of language produces normative force for speakers.15 Simply put: if the primary goal of speaking is to disclose the contents of your mind to others, you have good reason to try to coordinate the signification of your terms with them. As Locke puts it in his discussion of the abuses of language:

§29 [ … ] He that applies the Words of any Language to Ideas, different from those, to which the common use of that Country applies them, however his own Understanding may be fill’d with Truth and Light, will not by such Words be able to convey much of it to others, without defining his Terms. For however, the Sounds are such as are familiarly known, and easily enter the Ears of those who are accustomed to them; yet standing for other Ideas than those they usually are annexed to, and are wont to excite in the Mind of the Hearers, they cannot make known the Thoughts of him who thus uses them. (Essay, 3.10.29)

I think this fairly plausible point from Locke is why he is not concerned with generating a full account of a public language of the sort Lycan is describes. Having a public language would allow Locke to say that Humpty Dumpty is speaking falsely, but Locke’s view is rather that Humpty Dumpty’s use of language is impractical. Locke doesn’t need his utterances to be false; the important complaint is that Humpty Dumpty’s meaning is opaque to his audience. And if the point of his uttering something is to give us a glimpse of what’s going on in his head, then he’s not going to succeed – unless he explains what he means by the words (or, more effectively) conforms his signification to that of other speakers. So, for Locke, speakers are normatively encouraged to seek as much overlap as possible in the constellation of their individual idiolects.16 One could even say that when there is a certain critical mass of overlap, that becomes the meaning of the term in some public language, but the background normative system will do the actual work of explaining what Humpty Dumpty and the bungler are doing wrong.17 So, for Locke, the individualist mechanics of language are prerequisites for achieving its social purpose, but they can function in isolation. The social purpose further provides norms on how to communicate effectively, and those norms are what do the work of encouraging us to speak language the same way.

References

Ashworth, E. J. (1981). “Do words signify ideas or things?” The scholastic sources of Locke’s theory of language. Journal of the History of Philosophy 19 (3):299–326.

Carroll, Lewis, John Tenniel, George Dalziel, Edward Dalziel, and Richard Clay. 1872. Through the looking-glass, and what Alice found there. Macmillan and Co.

Guyer, Paul (1994). 5 Locke’s philosophy of language. In V. C. Chappell (ed.), The Cambridge Companion to Locke. Cambridge University Press. pp. 115.

Mill, John Stuart (2006); A system of logic ratiocinative and inductive. Books I-III. In The Collected Works of John Stuart Mill. Liberty Fund.

Kretzmann, Norman (1968). The main thesis of Locke’s semantic theory. Philosophical Review 77 (2):175–196.

Locke, John (1979). The Clarendon Edition of the Works of John Locke: An Essay Concerning Human Understanding. Clarendon Press.

Losonsky, Michael (2007). Language, meaning, and mind in Locke’s Essay. In Lex Newman (ed.), The Cambridge Companion to Locke’s “Essay Concerning Human Understanding”. Cambridge University Press.

Lycan, William G. (2008). Philosophy of Language: A Contemporary Introduction. Routledge.

Schroeder, Mark (2008). Being For: Evaluating the Semantic Program of Expressivism. Oxford University Press.

Sergeant, John (1697). Solid Philosophy Asserted Against the Fancies of the Ideists. Garland.

Marušić, Jennifer Smalligan (2014). Propositions and Judgments in Locke and Arnauld: A Monstrous and Unholy Union? Journal of the History of Philosophy 52 (2):255–280.

Stainton, Robert J. (1996). Philosophical Perspectives on Language: A Concise Anthology. Broadview Press.

1 I here adapt a convention, common to contemporary philosophy of mind, of designating ideas or concepts with all caps.

2 These positions are outlined straightforwardly in Essay 3.1 (for substantive terms) and 3.7.1 (for the copula).

3 Mill (2006), 1.2.1

4 This distinction is familiar in much contemporary literature on expressivism, See, for example, Schroeder (2008).

5 The difference between this approach and Ashworth’s “making known” is subtle, but significant. Ashworth’s interpretation includes in the signification of a word/sentence a great deal more than just the idea/thought being displayed.

6 It is worth noting that Sergeant’s objection dealt with an imperative sentence, and everything I am saying here is focused on declaratives. The point I am making can be generalized fairly easily. As an example, if imperative sentences are used to display desires (rather than beliefs) we can see how the reply to Sergeant would go: my utterance “(you) bring me the wine” signifies a desire of mine built out of my idea of the addressee, my idea of the wine, my idea of the act of bringing, and my idea of myself. That’s not to say the desire is about my ideas, but that it is built out of them.

7 The disjunction here is to cover the case of ideas without archetypes (Pegasus, Centaurs, etc.). In such cases, the agreements will terminate with the ideas themselves, rather than depending on some further agreement between the objects, since there are no objects.

8 My treatment of Lockean judgments here is cursory and oversimplified in some ways. For a better, more thorough treatment of Locke’s account of judgment, see Smalligan-Marusic (2014).

9 By advertising your belief that Elizabeth is human, you have publicly put yourself on the hook for possessing that belief, and can be held to account or challenged by someone with a contradictory belief. On this account, assertion is the result of publicizing one’s beliefs.

10 Other approaches can obviously capture this systematic alignment in the other direction: because the sentence has certain truth-conditions, believing what you say means that you would only utter it if you possessed the belief that has those truth conditions.

11 A somewhat similar worry to the one developed here is presented in Stainton (1996), p. 101.

12