Back Cover Blurb
- Donald Davidson presents a new edition of the 1984 volume which set out his enormously influential philosophy of language. Inquiries into Truth and Interpretation has been a central point of reference and a focus of controversy in the subject ever since, and its influence has extended into linguistic theory, philosophy of mind, and epistemology. This new edition features an additional essay, previously uncollected.
- The central question which these essays address is what it is for words to mean what they do. Davidson argues that a philosophically instructive theory of meaning should acknowledge the holistic nature of linguistic understanding, in that it should provide an interpretation of all utterances, actual and potential, of a speaker or group of speakers; and that it should not rely upon the concepts it attempts to explain, in that it should be verifiable independently of knowledge of the detailed propositional attitudes of the speaker.
- Among the topics covered in the essays are the relation between theories of truth and theories of meaning, translation, quotation, belief, radical interpretation, reference, metaphor, and communication.
Book Comment
- Clarendon Press; New Edition (27 Sept. 2001)
- Volume II of Davidson's Collected Works
- I also have a copy of the First Edition (1984)
"Davidson (Donald) - Inquiries into Truth & Interpretation: Preface/Introduction"
Source: Davidson - Inquiries into Truth & Interpretation
Introduction (Full Text1)
- What is it for words to mean what they do? In the essays collected here I explore the idea that we would have an answer to this question if we knew how to construct a theory satisfying two demands:
- It would provide an interpretation of all utterances, actual and potential, of a speaker or group of speakers; and
- It would be verifiable without knowledge of the detailed propositional attitudes of the speaker.
The first condition acknowledges the holistic nature of linguistic understanding. The second condition aims to prevent smuggling into the foundations of the theory concepts too closely allied to the concept of meaning. A theory that does not satisfy both conditions cannot be said to answer our opening question in a philosophically instructive way.
- The first five essays are mainly concerned with the question what sort of a theory would satisfy the first condition.
- Essay 7, "Davidson (Donald) - On Saying That", urges that a satisfactory theory must discover a finite basic vocabulary in the verbal phenomena to be interpreted if it is to prove useful to a creature with finite powers. If this is so, there is no escape from the need to treat the semantic features of the potential infinity of sentences as owed to the semantic features of the items in a finite vocabulary. It turns out that a number of familiar theories fail to meet this condition: Frege’s analysis of oblique contexts, Church's logic of sense and denotation, Tarski’s informal treatment of quotation are examples. Standard theories of adverbial modification might well be added to the list.
- Essay 2, "Davidson (Donald) - Truth and Meaning", argues that a theory of truth along the lines of Tarski’s truth definitions, but modified in various ways to apply to a natural language, would be enough for an interpreter to go on. Such theories have clear virtues. They make no use of meanings as entities; no objects are introduced to correspond to predicates or sentences; and from a finite set of axioms it is possible to prove, for each sentence of the language to be interpreted, a theorem that states truth conditions for that sentence. Further, the proof of such a theorem amounts to an analysis of how the truth or falsity of the sentence depends on how it is composed from elements drawn from the basic vocabulary. If such theories really do satisfy the two conditions listed in the first paragraph, we can take the word ’theory' in ’theory of meaning’ seriously.
- Many objections have been made to the claim that truth theories can do duly as theories of meaning. Some of the objections I have tried to meet or deflect in other essays in this book. But whether or not the claim can be made good, some of the arguments for it in ’Truth and Meaning’ are faulty. The reader will find that I shifted ground more than once as I tried to improve or clarify this central thesis. One thing that only gradually dawned on me was that while Tarski intended to analyse the concept of truth by appealing (in Convention T) to the concept of meaning (in the guise of sameness of meaning, or translation) I have the reverse in mind. I considered truth to be the central primitive concept, and hoped, by detailing truth’s structure, to get at meaning. These are remarks about theories of truth, of course, not remarks to be found in them.
- Something else that was slow coming to me was that since I was treating theories of truth as empirical theories, the axioms and theorems had to be viewed as laws. So a theorem like ‘“Schnee ist weiss” is true in the mouth of a German speaker if and only if snow is white' has to be taken not merely as true, but as capable of supporting counterfactual claims. Indeed, given that the evidence for this law, if it is one, depends ultimately on certain causal relations between speakers and the world, one can say that it is no accident that ’Schnee ist weiss’ is true if and only if snow is white; it is the whiteness of snow that makes ‘Schnee ist weiss’ true. How much of a concession this is to intensionality depends, I suppose, on one’s analysis of the concept of law. What seems clear is that whatever the concession comes to, it is one that must be made for any empirical science. These matters are discussed in Essay 12 ("Davidson (Donald) - Reply to Foster").
- Essay 3, "Davidson (Donald) - True to the Facts", asks whether a theory of truth in Tarski's style should be called a correspondence theory. Such theories do not, like most correspondence theories, explain truth by finding entities such as facts for true sentences to correspond to. And there are good reasons, which can be traced back to Frege, for rejecting facts as entities that could play this role. On the other hand, theories of truth of the kind considered here do require that a relation between entities and expressions be characterized (‘satisfaction’). It is not easy to see how a satisfactory route to truth can escape this step if the language the theory treats has the usual quantificational resources.
- "Davidson (Donald) - Semantics For Natural Languages", Essay 4, urges that truth theories could provide a formal semantics for natural languages to match the sort of formal syntax linguists from Chomsky on have favoured. When this essay was written, the deep structures of syntax were thought to be the vehicles for semantic interpretation. Essay 4 suggested that the deep structure of a sentence should correspond to the logical form a theory' of truth assigned to that sentence.
- Tarski’s Convention T2, which is defended in Essay 5 ("Davidson (Donald) - In Defence of Convention T"), is an informal, but powerful, instrument for testing theories of truth against one’s prior grasp of the concept. In the most direct application, the test merely calls on us to recognize the disquotational feature of truth predicates; sentences like ‘“Snow is white” is true in English if and only if snow is white' are trivially true. Since the totality of such sentences uniquely determines the extension of a truth predicate for English, a theory that entails all such sentences must be extensionally correct. Critics have often made the error of thinking that since the theorems that show a theory to be correct are trivial, the theory or the concept of truth it characterizes, must also be trivial.
- A theory of truth would serve to interpret a speaker only if the theory were up to accounting for all the linguistic resources of the speaker. But is a theory that satisfies Convention T adequate to a natural language? Here there are two questions.
- One is what devices to make or consider available in the language of the theory;
- The other is how to apply these devices to the language of the speaker.
My working assumption has been that nothing more than standard first- order quantification theory is available. Indeed, I was long convinced that many alternative approaches to semantics, employing, for example, modal logics, possible world semantics, or substitutional quantification, could not be accommodated in a theory that met the demands of Convention T. I now know this was hasty. Convention T does not settle as much as I thought, and more possibilities for interesting theorizing are open than I had realized. The well-known virtues of first-order quantification theory still provide plenty of motivation, however, to see how much we can do with it. In the next three essays, collected under the head of application, I attempt the semantic taming of three related but recalcitrant idioms: quotation, indirect discourse, and mood operators.
- Essay 6 ("Davidson (Donald) - Quotation") points out that no current theory of quotation is entirely satisfactory, and it proposes an explicitly demonstrative approach which makes quotation a special case of the demonstrative reference of words to other words in the verbal neighbourhood.
- Essay 7, "Davidson (Donald) - On Saying That", concentrates on one of the many kinds of sentence used to attribute attitudes; the paratactic3 solution suggested has obvious affinities with the treatment of quotation in Essay 6. In Essay 3 there are hints (which I think could be developed) on how the analysis could be extended to belief sentences. If the strategy were to be pursued, it might serve to give a semantics (though not a logic) for the modalities, for counterfactuals, and further sentences about ‘propositional attitudes’.
- Essay 8, "Davidson (Donald) - Moods and Performance", stresses the often neglected distinction between grammatical moods on the one hand and various sorts of illocutionary force on the other. Only the first is of concern to a theory of what words mean. Here a paratactic4 analysis of imperatives is suggested which is intended to accommodate our natural feeling that imperatives don’t have a truth value while remaining within the resources of a theory of truth.
- In the companion volume to this one, "Davidson (Donald) - Essays on Actions and Events", I show how a theory of truth can be applied to a number of further problem cases: sentences about actions and other events, adverbial modification, and singular causal statements.
- The third section of the present book is addressed to the question whether a theory of truth for a speaker can be verified without assuming too much of what it sets out to describe.
- In "Davidson (Donald) - Radical Interpretation", Essay 9, as in the rest of the essays, I follow Quine in supposing that even if we narrow attention to verbal behaviour that reveals when, and under what conditions, a speaker gives credence to a sentence, there is no direct way of sorting out the roles of belief and meaning in explaining that credence. Eliciting separate accounts of belief and meaning requires a theory that can bring to bear on the interpretation of each sentence and its accompanying attitudes the contribution of further data. Only by studying the pattern of assents to sentences can we decide what is meant and what believed.
- Depending on evidence which, without the aid of theory, makes no distinction between the contributions of belief and meaning to linguistic behaviour, requires a method for effecting the separation to a degree sufficient for communication. Devices to this end are described and defended in the present essays. But all of them, in one way or another, rely on the Principle of Charity.
- The phrase and the basic idea come from "Wilson (N.L.) - Substances without Substrata". Quine puts it this way: ‘... assertions startlingly false on the face of them are likely to turn on hidden differences of language’ ("Quine (W.V.) - Word & Object", p. 59). Quine applies the principle primarily to the interpretation of the logical constants.
- Because I find I cannot use Quine’s notion of stimulus meaning as a basis for interpreting certain sentences, I apply the Principle of Charity across the board. So applied, it counsels us quite generally to prefer theories of interpretation that minimize disagreement. So I tended to put the matter in the early essays, wanting to stress the inevitability of the appeal to charity. But minimizing disagreement, or maximizing agreement, is a confused ideal. The aim of interpretation is not agreement but understanding. My point has always been that understanding can be secured only by interpreting in a way that makes for the right sort of agreement. The ‘right sort’, however, is no easier to specify than to say what constitutes a good reason for holding a particular belief.
- The subtle pressures on the Principle of Charity begin to emerge in Essays 10 ("Davidson (Donald) - Belief and the Basis of Meaning") and 11 ("Davidson (Donald) - Thought and Talk"). Yet here too there are only hints; in work now in progress I attempt to develop the subject in more detail.
- Essay 10, "Davidson (Donald) - Belief and the Basis of Meaning", insists on the symmetry of belief and meaning in the exploration of verbal behaviour. In one important respect it goes further. It develops a striking parallel between Bayesian theories of decision and theories of meaning, and gives reasons why the two theories should be considered mutually dependent. The hints dropped here, which give promise of a unified theory of speech and action, have been taken up in subsequent publications.
- In the Appendix to Essay 10 ("Davidson (Donald) - Belief and the Basis of Meaning: Replies to Lewis and Quine") I express agreement with David Lewis on several points. The Principle of Charity should be modified as suggested in my comments above on Essay 9 ("Davidson (Donald) - Radical Interpretation"). Charity prompts the interpreter to maximize the intelligibility of the speaker, not sameness of belief. This entails, as Lewis says, that interpretation must take into account probable errors due to bad positioning, deficient sensory apparatus, and differences in background knowledge. He is right also in saying that in practice an interpretation should explain, and so be checked against, observed non-verbal behavior. I do not take this to prove that the evidential base on which I depend is not in theory adequate. I grant, however, that it may not be. Quine and I are, as one would expect, pretty much in accord. A difference emerges, however, on the extent of possible indeterminacy both in logic and in the interpretation of theoretical sentences; my view that quantificational structure can be detected at an early stage and my preference for a semantical theory over translation stiffen the constraints.
- The first two essays on radical interpretation stress the fact that understanding the words of a speaker requires knowing much about what he believes. Essay 11, "Davidson (Donald) - Thought and Talk", attends to the reciprocal dependence, and concludes, rather speculatively, that only a creature with a language can properly be said to have a full-fledged scheme of propositional attitudes.
- Essay 125, "Davidson (Donald) - Reply to Foster", as remarked above, recognizes that if a theory of truth is to suffice for interpretation, it must be more than true: its axioms and theorems must be natural laws. If an interpreter knew such a theory, he could use it to understand a speaker, but only if he knew that the theory's pronouncements were nomic.
- The next four essays may be described as philosophical fall-out from the approach to truth and interpretation recommended here.
- A theory of truth can be called a correspondence theory in the unassuming sense of Essay 3 ("Davidson (Donald) - True to the Facts"), but that sense does not encourage the thought that we understand what it would be like to compare sentences with what they are about, since the theory provides no entities with which to compare sentences. Along related lines, Essay 13, "Davidson (Donald) - On the Very Idea of a Conceptual Scheme", scouts the intelligibility of claims that different languages or conceptual schemes ‘divide up' or ‘cope with' reality in importantly different ways. Our general method of interpretation forestalls the possibility of discovering that others have radically different intellectual equipment. But more important, it is argued that if we reject the idea of an uninterpreted source of evidence no room is left for a dualism of scheme and content. Without such a dualism we cannot make sense of conceptual relativism. This does not mean that we must give up the idea of an objective world independent of our knowledge of it. The argument against conceptual relativism shows rather that language is not a screen or filter through which our knowledge of the world must pass. Giving up the dualism of scheme and content amounts to abandoning a theme central to empiricism in its main historical manifestations. But I do not think, as friends and critics have variously suggested, that my argument against empiricism makes me, or ought to make me, a pragmatist, a transcendental idealist, or an ‘internal' realist. All these positions are forms of relativism that I find as hard to understand as the empiricisms I attack.
- According to Essay 13, "Davidson (Donald) - On the Very Idea of a Conceptual Scheme", no sense can be made of the idea that the conceptual resources of different languages differ dramatically. The argument that makes for this conclusion makes equally for the conclusion that the general outlines of our view of the world are correct: we individually and communally may go plenty wrong, but only on condition that in most large respects we are right. It follows that when we study what our language — any language — requires in the way of overall ontology, we are not just making a tour of our own picture of things: what we take there to be is pretty much what there is. This is the theme of Essay 14, "Davidson (Donald) - The Method of Truth in Metaphysics".
- A theory of truth is tested by theorems that state the conditions under which sentences are true; these theorems say nothing about reference. Essay 15, "Davidson (Donald) - Reality Without Reference", accordingly contends that how a theory of truth maps non-sentential expressions on to objects is a matter of indifference as long as the conditions of truth are not affected. The question what objects a particular sentence is about, like the questions what object a term refers to, or what objects a predicate is true of, has no answer.
- In Essay 15, "Davidson (Donald) - Reality Without Reference", I am with Quine in holding reference to be inscrutable. Essay 16, "Davidson (Donald) - The Inscrutability of Reference", warns against taking inscrutability as a reason for trying somehow to relativize the reference and ontology of singular terms and predicates. For since nothing can reveal how a speaker’s words have been mapped on to objects, there is nothing to relativize to; and interpretation being unaffected, there is no need to relativize.
- No discussion of theories of meaning can fail to take account of the limits of application of such theories. The scope must be broad enough to provide an insight into how language can serve our endless purposes, but restricted enough to be amenable to serious systematization. Essay 8 ("Davidson (Donald) - Moods and Performance") took a necessary step by distinguishing between grammatical mood, which the meanest theory must account for, and the force of utterances, which is beyond the reach of comparable regimentation. Essay 17, "Davidson (Donald) - What Metaphors Mean", is mainly devoted to the thesis that we explain what words in metaphor do only by supposing they have the same meanings they do in non-figurative contexts. We lose our ability to account for metaphor, as well as rule out all hope of responsible theory, if we posit metaphorical meanings.
- Essay 18, "Davidson (Donald) - Communication and Convention", draws another boundary. It is always an open question how well the theory an interpreter brings to a linguistic encounter will cope. In practice an interpreter keeps the conversation going by adjusting his theory on the spot. The principles of such inventive accommodation are not themselves reducible to theory, involving as they do nothing less than all our skills at theory construction.
- The essays have been retouched in minor ways to reduce repetition, to eliminate unnecessary or confused passages, or to bring early more into line with latter thoughts. These temperings have been limited to the trivial. Where my errors or lapses have earned attention I have let things stand, or marked the change with a footnote.
- An early influence on my thinking was Michael Dummett, who lectured on Frege and philosophy of language several times at Stanford University while I was there in the fifties. Our discussions took a public form in 1974 when we gave a joint seminar on truth while I was a visiting fellow at All Souls College.
- W.V. Quine was my teacher at a crucial stage in my life. He not only started me thinking about language, but he was the first to give me the idea that there is such a thing as being right, or at least wrong, in philosophy, and that it matters which. Without the inspiration of his writing, his patient tutelage, his friendly wit and his generous encouragement, this book would not be worse than it is. It would not be.
In-Page Footnotes ("Davidson (Donald) - Inquiries into Truth & Interpretation: Preface/Introduction")
Footnote 1:
- I’ve retained the full text here, as it explains how the various essays interrelate with one another. But I’ve not included the bulk of the credits.
- However, I also intend – hopefully shortly – to copy the relevant text to the papers referenced, to act as Introductions thereto.
- I was incited to look at this book because it was cited – rather obliquely – in "Forsey (Jane) - Humans and Dumb Animals" - with reference to Davidson’s alleged claim that non-human animals cannot think, because they have no language. It looks like Essay 11 ("Davidson (Donald) - Thought and Talk") is the only obviously relevant item. She also cited "Davidson (Donald) - Essays on Actions and Events".
Footnote 2: Footnotes 3, 4: Footnote 5:
"Davidson (Donald) - Theories of Meaning and Learnable Languages"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 1
"Davidson (Donald) - Truth and Meaning"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 2
Abstract1
- This paper, like "Grice (H. Paul) - Meaning" (1957) launched a whole program in the philosophy of language, the program alluded to above of giving a Tarski type truth definition for a natural language as a theory of meaning for that language.
- Here Davidson argues for this approach over others, illustrates some of its virtues, and ends with a sketch of what needs to be done to bring this project to fruition.
- For twenty years many philosophers of language have been attempting to complete this program.
Philosophers Index Abstract
- Apparently the speakers of a language can effectively determine the meaning or meanings of an indefinitely large number of expressions. A theory of meaning explains how this is possible if it gives a recursive account of the meaning of every sentence on the basis of its structure.
- The problem discussed in this paper is: how does a theory of meaning do this? It is argued that by characterizing a predicate satisfying (roughly) Tarski's convention t (which tests the adequacy of a formal semantical definition of truth), a theory of meaning describes the required kind of structure. A theory meeting this condition would provide a clear and testable semantics for a natural language.
Paper Comment
In-Page Footnotes ("Davidson (Donald) - Truth and Meaning")
Footnote 1: Taken from "Harnish (Robert M.) - Basic Topics in the Philosophy of Language: Introduction".
"Davidson (Donald) - True to the Facts"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 3
Philosophers Index AbstractThis paper defends the correspondence theory of truth, that is, the claim that the truth of a statement can be explained in terms of a relation between language and the world. It is argued that tarski's concept of 'satisfaction', relativized to occasions of speech, is such a relation. The notion of corresponding to the facts, on the other hand, while harmless in itself, cannot be elaborated into a non-trivial account of truth.
"Davidson (Donald) - Semantics For Natural Languages"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 4
"Davidson (Donald) - In Defence of Convention T"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 5
"Davidson (Donald) - Quotation"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 6
"Davidson (Donald) - On Saying That"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 7
Philosophers Index AbstractAn analysis of indirect discourse is offered according to which the "that" of (e.g.) "Galileo said that" is viewed as a demonstrative referring to an immediately subsequent utterance. It is argued that this analysis, unlike several discussed in Quine's Word and Object, does not prevent the recursive characterization of a Tarski-type truth predicate.
Paper Comment
"Davidson (Donald) - Moods and Performance"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 8
"Davidson (Donald) - Radical Interpretation"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 9
Philosophers Index AbstractWhat knowledge would suffice to yield an interpretation of an arbitrary utterance of a language when such knowledge is based on evidence plausibly available to a nonspeaker of that language? it is argued that it is enough to know
… (1) a theory of truth for the language and
… (2) that the theory satisfies Tarski's 'Convention t' (modified to apply to natural language) and
… (3) that it gives an optimal fit (in a sense described) to data about sentences held true, under specified conditions, by native speakers.
"Davidson (Donald) - Belief and the Basis of Meaning"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 10
Philosophers Index AbstractA theory of "radical interpretation" gives the meanings of all sentences of a language, and can be verified by evidence available to someone who does not understand the language. Such evidence cannot include detailed information concerning the beliefs and intentions of speakers, and therefore the theory must simultaneously interpret the utterances of speakers and specify (some of) his beliefs. Analogies and connections with decision theory suggest the kind of theory that will serve for radical interpretation, and how permissible evidence can support it.
Paper Comment
Also in "Martinich (A.P.) - The Philosophy of Language"
"Davidson (Donald) - Belief and the Basis of Meaning: Replies to Lewis and Quine"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 10 (Appendix)
"Davidson (Donald) - Thought and Talk"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 11
Paper Comment
Also in "Rosenthal (David), Ed. - The Nature of Mind"
"Davidson (Donald) - Reply to Foster"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 12
"Davidson (Donald) - On the Very Idea of a Conceptual Scheme"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 13
Philosophers Index AbstractIt is argued that a dualism of scheme and content, of organizing system and something waiting to be organized, cannot be made intelligible and defensible. But if this dualism is abandoned, there is no basis for conceptual relativism, the idea that there may be profound contrasts between conceptual schemes.
"Davidson (Donald) - The Method of Truth in Metaphysics"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 14
Philosophers Index AbstractIn sharing a language we share a picture of the world that must, in its large features, be true. It follows that in making manifest the large features of our language, we make manifest the large features of reality. A translational theory of truth provides a method for revealing metaphysically significant structure in language. Examples are given to show how the method yields results for ontology.
"Davidson (Donald) - Reality Without Reference"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 15
Philosophers Index AbstractA dilemma concerning reference is posed: on the one hand it seems essential, if we are to give an account of truth, to first give an account of reference. On the other hand, reference is more remote than truth from the evidence in behavior on which a radical theory of language must depend, since words refer only in the context of sentences, and it is sentences which are needed to promote human purposes. The solution which is proposed is to treat reference as a theoretical construct whose sole function is to serve a theory of truth. Since more than one relation between words and objects will serve a theory of truth equally well, this amounts to giving up the concept of reference as basic.
"Davidson (Donald) - The Inscrutability of Reference"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 16
Philosophers Index AbstractMany different ways of assigning a reference to singular terms and predicates may account equally well for the linguistic behavior of a speaker. This leads quine to hold that the ontology of a speaker is fixed only relative to the arbitrary choice of a translation manual and of a background language. But the inscrutability of reference, so far from supporting such relativizations, shows that there is no way of settling matters of ontology.
"Davidson (Donald) - What Metaphors Mean"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 17
Paper Comment
Also in "Martinich (A.P.) - The Philosophy of Language"
"Davidson (Donald) - Communication and Convention"
Source: Davidson - Inquiries into Truth & Interpretation, Chapter 18
Philosophers Index Abstract
- Three ways in which convention has been said to be involved in language are considered:
- a general convention that in speaking one is trying to say what is true;
- a convention linking the meaning of each sentence with a standard use;
- a convention that individual words and constructions are to have a certain meaning.
- It is argued that there are no conventions of the first two kinds, and that the concept of expectation is all that is needed to explain the third.
- It remains a question whether convention helps explain how language works.
Text Colour Conventions (see disclaimer)- Blue: Text by me; © Theo Todman, 2025
- Mauve: Text by correspondent(s) or other author(s); © the author(s)