Inquiries Into Truth and Interpretation

by Donald Davidson

The eighteen essays in this collection address the question of what it is for words to mean what they do. Davidson covers such topics as the relation between theories of truth and theories of meaning, translation, quotation, belief, radical interpretation, reference, metaphor, and communication.

Genres: Philosophy, Language, Linguistics, Nonfiction, Politics, American

Paperback

First published April 26, 1984

About the author

Donald Davidson was one of the most important philosophers of the latter half of the twentieth century. His ideas, presented in a series of essays from the 1960's onwards, have been influential across a range of areas from semantic theory through to epistemology and ethics. Davidson's work exhibits a breadth of approach, as well as a unitary and systematic character, which is unusual within twentieth century analytic philosophy. Thus, although he acknowledged an important debt to W. V. O. Quine, Davidson's thought amalgamates influences (though these are not always explicit) from a variety of sources, including Quine, C. I. Lewis, Frank Ramsey, Immanuel Kant and the later Wittgenstein. And while often developed separately, Davidson's ideas nevertheless combine in such a way as to provide a single integrated approach to the problems of knowledge, action, language and mind. The breadth and unity of his thought, in combination with the sometimes-terse character of his prose, means that Davidson is not an easy writer to approach. Yet however demanding his work might sometimes appear, this in no way detracts from either the significance of that work or the influence it has exercised and will undoubtedly continue to exercise. Indeed, in the hands of Richard Rorty and others, and through the widespread translation of his writings, Davidson's ideas have reached an audience that extends far beyond the confines of English-speaking analytic philosophy. Of late twentieth century American philosophers, perhaps only Quine has had a similar reception and influence.

Donald Herbert Davidson was born on March 6th, 1917, in Springfield, Massachusetts, USA. He died suddenly, as a consequence of cardiac arrest following knee surgery, on Aug. 30, 2003, in Berkeley, California. Remaining both physically and philosophically active up until his death, Davidson left behind behind a number of important and unfinished projects including a major book on the nature of predication. The latter volume was published posthumously (see Davidson, 2005b), together with two additional volumes of collected essays (Davidson 2004, 2005a), under the guidance of Marcia Cavell.

Davidson completed his undergraduate study at Harvard, graduating in 1939. His early interests were in literature and classics and, as an undergraduate, Davidson was strongly influenced by A. N. Whitehead. After starting graduate work in classical philosophy (completing a Master's degree in 1941), Davidson's studies were interrupted by service with the US Navy in the Mediterranean from 1942-45. He continued work in classical philosophy after the war, graduating from Harvard in 1949 with a dissertation on Plato's ‘Philebus’ (1990b). By this time, however, the direction of Davidson's thinking had already, under Quine's influence, changed quite dramatically (the two having first met at Harvard in 1939-40) and he had begun to move away from the largely literary and historical concerns that had preoccupied him as an undergraduate towards a more strongly analytical approach.

While his first position was at Queen's College in New York, Davidson spent much of the early part of his career (1951-1967) at Stanford University. He subsequently held positions at Princeton (1967-1970), Rockefeller (1970-1976), and the University of Chicago (1976-1981). From 1981 until his death he worked at the University of California, Berkeley. Davidson was the recipient of a number of award and fellowships and was a visitor at many universities around the world. Davidson was twice married, with his second marriage, in 1984, being to Marcia Cavell.

-http://plato.stanford.edu/entries/dav...

Reviews

[yyy · Rating 4 out of 5]

Overall, I liked this collection very much. Not as much as Essays on A&E, but it's still very good. (Indeed, per convention 4 stars means "I loved it".)

props:
- Paratactic analysis: who would have thought solving problems related to oratio obliqua, quotation, and moods/performatives would have such a simple solution?

slops:
- The essays are repetitive, sometimes (not always) to a fault; perhaps part of the reason why Davidson's repertory has such coherence -- despite the fact that he never wrote a book -- is that he has two or three core notions that do most of his work. This is less a criticism of the essays than a caveat to the prospective reader: prepare for massive repetition.

- Davidson cities Tarski's "The Concept of Truth in Formalized Languages" in Every. Single. Essay. I exaggerate, but still... it's in at least 15 of 'em. I don't know if this is exactly a "slop", but the temptation to play find-the-Tarski-citation is too strong.

[Liedzeit Liedzeit · Rating 3 out of 5]

When I was studying philosophy, one of the frustrating things was that I could not understand Tarski. What he was saying seemed trivial to me. And everybody told me it was not trivial at all. What did I miss?

The same thing happened when I read this book. And no wonder, Davidson relies heavily on Tarski. Of the 18 essays in this book, originally written between 1965 and 1982, all but two refer directly to Tarski.

So ‘snow is white’ is true if and only if snow is white. How often can you repeat this (or read this) without going crazy?

But whereas Tarski tried to present a definition of ‘truth’ Davidson aims for a theory of truth. A theory of truth, he says, is a set of axioms from which for every sentence of a language the conditions under which this sentence is true follows. (In: Semantics for Natural Languages)

And in sixteen of the essay he tries to explain this again and again. (At least he is consistent. Have just one idea and stick to it.) But I still did not get it.

In his Reply to Foster he admits that he has not managed to find a satisfying explanation.

Fortunately, I remembered that I have an excellent book (by Bernard Harrison) on the philosophy of language on my shelf that has a chapter on Davidson. And after reading that, I now think that not only do I understand Davidson, but I even think that he is right (whereas Harrison thinks he is not).

So what is it all about? The snow sentence is called a T sentence. Another example is: ‘There are million stars out tonight’ is true iff (if and only if) there are a million stars out tonight.

The T sentence in fact offers a way of dispensing with the concept of truth. Every statement of the form ‘s is true’ is replaced by a statement ‘p’ which is identical or a translation of s.

Truth is defined extensionally by the totality of T sentences. Still, what does this mean?

It means that we can take the notion of truth as fundamental and use it for a theory of meaning. And this is best demonstrated by his idea of radical interpretation.

How does a T sentence like ‘Es regnet’ is true iff it is raining helps in understanding German?

Davidson suggests that we think of a sentence as the smallest unit of meaning. He rejects the idea that we need to construct a sentence out of words, and he rejects the idea that we need to know the beliefs of a speaker with whom we are communicating. Instead, he applies a 'principle of charity', which means that the beliefs of the individual do not matter, nor do his intentions, as long as in the long run (given enough data) utterances of 'Es regnet' will actually occur when it is raining.

This is how Harrison sums it up:

The primary process of radical interpretaion is that of distributing truth conditions to sentences in such a way as to maximise agreement and truth across the wides possible range of utterances and contexts of utterance. Radical interpretation is thus essentially holistic in nature.

Now it is easy to criticise this approach. Putnam does so in his Meaning of Meaning where he says that Davidson’s (and Quine’s) claim that we need to go from whole sentences to individual words is the opposite of the procedure upon which the study of natural languages actually are based.

But this is not a good argument. It certainly seems that we first learn individual words and then somehow manage to put them together into sentences (using the Chomsky transformational grammar rules that we are supposedly born with). But is this really so?

It seems to me that the success of modern AI proves that Davidson was quite right. Large language models seem to acquire knowledge of a language exactly as Davidson says, by analyzing huge amounts of data.

In the essay What Metaphors Mean, Davidson argues that metaphors mean only what they literally mean, that they have no second metaphorical meaning. He says that sentences with metaphors are false. This seems to me to contradict his own theory. A word used metaphorically is in principle no different from a homonym. And its meaning becomes clear if enough utterances with this word are analyzed. To guess the new meaning by seeing similarities to the existing meaning is just a shortcut we humans use.

In the last essay Communication and Convention Davidson denies that language is based on conventions. On the contrary, language is a precondition for having conventions. This sounds right to me.

This is an excellent book, but really poorly written. His ideas are not trivial but it would have been nice if he had been able to express them a little better.

7/10

[Micah Newman · Rating 4 out of 5]

These essays exposit some of Davidson's most influential (or at least commented-on) and well-known views. Starting with “Truth and Meaning,” Davidson takes Tarski’s theory of truth and puts it to a different use than originally intended to see if a semantic "theory of truth" (by which I think he generally means knowing what sentences in a language are true) can "do duty" as a theory of meaning. Gradually over the course of these essays, he works out the consequences to areas like translation, interpretation, and belief. Because the essays represent variations on just a very few central themes, and the essays were originally written to be presented on different occasions and so were each intended to stand alone, taken together there is a certain amount of repetition, especially in the early going. And in some of these essays, Davidson's writing is sometimes oblique and elliptical. For many instances of “Of course,” “It is obvious that,” and “Clearly,” I simply have to guess at what Davidson might be gesturing at. He seems to assume a certain background in formal linguistics or mathematical logic, or both, it’s hard to tell. In addition, the Tarski-heavy material I've always found a bit thin, a bit dry and unexciting. That plus the repetition in the first section of papers means it may not get your juices flowing right away. But it gets better.

The cardinal question in philosophy of language is what it is to speak, to understand, and to be understood. For Davidson, this is all approached through the lens of interpretation right from the start: as witness his central claim that only the interpreter of another could have meaningful speech and thoughts of his own. Interpretation seems to be the single central theme pervading all of Davidson’s work. The first major step in the advancement of this theme is made in "Radical Interpretation," where he takes Quine's notion of "radical translation" as a point of departure and describes how one might in principle test his theory of meaning-by-truth by generalizing it to a theory of the necessary and sufficient conditions of understanding a language from scratch. This sets the groundwork for further ideas.

The next major idea comes with the circumscribed notion of literal semantic meaning he had already been employing, in advancing the doctrine of the "autonomy of meaning" —that an utterance with a given literal meaning may, depending on the circumstances, be put to any (nonlinguistic) use whatsoever. He notes that people rarely talk with the mere of aim of saying true things, a startling claim whose significance and truth becomes apparent when examined. An application Davidson puts it to here is in his unique view that the meaning of a metaphor is just its literal meaning. This is another claim that seems obviously false at first, unless you bear firmly in mind the restricted and literal sense he in which he means "meaning"—in the case of metaphor, what one might think of as its "meaning" in the broader sense far outstrips its literal meaning, and in fact, the semantic meaning of anything that could be put into a literal expression, which is why what metaphors express really can't be paraphrased. A common metaphor dies (ceases to be a metaphor) when it comes to extend the literal meaning of a word through repeated use and understanding. His arguments are bracing and his command of examples in literature is deft and put to very effective use. Although this view is seemingly only cited by others for the purpose of a quick dismissal (as is often the case for Davidson's views), I think Davidson is absolutely right (about most things, in fact).

The "autonomy of meaning" comes up repeatedly in later writings (in volumes 4 and 5). He develops the idea further here by pointing out that it is an essential characteristic of language and language users that utterances are always made with "ulterior motives" that go well beyond the literal meaning of what is said – and in fact anything that did not have this characteristic would not count as a language. Overall, his arguments that are sprung from the "autonomy of meaning" kernel amount to a comprehensive theory of speech acts, although he does not refer to them often per se.

Another big, even epoch-shattering, idea is in "On the Very Idea of a Conceptual Scheme," in which he attacks what he calls the "third dogma of empiricism"—scheme-content dualism, according to which content can be organized by any of a number of incommensurable schemes (cf. Kuhn), which assumes there is uninterpreted "content" to begin with. Viewing this as a "third dogma" flows seamlessly from Quine's first "two dogmas," but it is seemingly even more radical. His dismantling of the scheme-content dichotomy is so thorough and penetrating that one finds that a dogma in common to empiricists *and* rationalists, going back at least to Descartes, has been simply swept away. One is left with *unmediated* access to the world through language by the triangulation of meaning with other language users (this forms almost the entire basis for the essays in volume 3) and no intelligible basis for a substantive relativism that would leave different groups of language users conceptually isolated and in principle unable to communicate meaningfully. In a nutshell, if we found that we simply could not make sense of an uninterpreted language or of the behavior of its users, we could not conclude that “their rationality” was different from ours; rather, we simply would not be in a position to attribute thoughts, language, or rationality to its users at all.

It's worth emphasizing that, although he always restricted his focus to a small cluster of philosophical issues, Davidson's outlook on them is *utterly unique*. That, combined with his dogged programmaticity and systematicity over his entire philosophical career, means that you can't even effectively argue against any of his positions without also arguing against pretty much all of them. That alone, at least in my view, puts Davidson on the same list with Kant, Hegel, Frege, Wittgenstein, Kripke, and Putnam.

[Rafael Nardini · Rating 5 out of 5]

Donald Herbert Davidson, filósofo norte-americano, catedrático de Filosofia da Linguagem em Stanford e Princeton, desenvolveu em sua produção intelectual o conceito do "princípio da acomodação racional", também denominado de "princípio da caridade interpretativa", que prescreve ao intérprete, num debate, maximizar o acordo e minimizar o desacordo, procurando entender o ponto de vista de seu interlocutor da forma mais forte e persuasiva possível, para que impere uma verdade partilhada entre os dois, viabilizando o entendimento mútuo.

Entendendo comunidades

Criar grupos em que todos pensam igual, no qual não existem diferenças e as afinidades se sustentam por uma linguagem e uma percepção da vida única não é estar em comunidade, mas sim em uma máfia, ou de forma delicada, em uma turma. Como a turma do candidato A e a turma do candidato B.
Na turma todos pensam igual e "o objeto central sou eu". Todos são, em realidade, espelhos bem-vindos das minhas crenças e certezas.
Mas quando pensamos em comunidades existem diferenças, existe diversidade, existem conflitos porque existe o EU e existe o OUTRO. A comunidade é bem mais interessante do que a turma, mas claro que fazer parte da turma é mais fácil e muito mais aconchegante.

[Juan Fuentes]

Demasiado para un gañán como yo. Lo que he entendido (la cuarta parte o así) muy estimulante.

[Leonardo]

"Verdad y Significado" e "Interpretación Radical" para Filosofía del Lenguaje. Unidad V.