Donald Davidson's Philosophy
  • Donald Davidson's Philosophy

Donald Davidson's Philosophy (Print)

Gerhard Preyer

Donald Davidson’s Philosophy
From Radical Interpretation to Radical Contextualism

332 pages, paperback
Second Edition 2011
ISBN 978-3-941743-11-3

Buch 29,80 Euro
E-Book (PDF) 19,80 Euro

29,80 €

Donald Davidson has contributed to many subjects in contemporary philosophy. He was one of the most influential Anglo-American analytical philosophers in the late 20th century. His leading topics are the theory of meaning, the philosophy of action and the theory of mind. His claim is to give a Unified Theory of Thought, Meaning, Action, and Evaluation as a new foundational account of language. In the history of his work emerged an overall view of mind and its relation to the world. This approach argues for a total revision of the Cartesian tradition and of traditional empiricism in epistemology.

This book systemizes his philosophy and refers to the critiques of his theory of meaning, action, and mind since the 1970s years. It is not fixed what Davidson’s place in the history of philosophy will be, but many patterns of philosophical thought in the last fifty years would not have been without him.

Teachers and students (graduate and advanced undergraduate) in the field of philosophy of language, mind, action, and moral philosophy.

Gerhard Preyer is professor of sociology at the Goethe-University Frankfurt am Main, Germany. Among his publications are Intention and Practical Thought and Interpretation, Language and the Social. Philosophical Articles. He edits ProtoSociology. An International Journal of Interdisciplinary Research (



Preface 9
Introduction 11

Part I
Radical Interpretation, Logical Form and Events 17

1. Donald Davidson's Philosophy: An Overview 19
2. Truth, Meaning and Radical Interpretation 67
2.1. From Radical Interpretation to Radical Externalism 67
2.2. From the Idiolect Theory to the Third Dogma of Empiricism 133
2.3. The Dismantling of a Myth 143
2.4. On Epistemic Restrictions of Understanding 156
3. The Logical Form of Action Sentences and Singular Causal Statements 168
3.1. Logical Form and Adverbial Modification 171
3.2. Causal Relations 184

Part II
Primary Reasons, Body Movements and Actions 193

4. The Logical-Connection Argument 195
5. Basic Acts 205
6. Primary Reasons and the Identity-Thesis 212
6.1. Pro Attitudes, Beliefs, Primitive Actions and their Causal Relations 214
6.2. From Anomalism of the Mental to the Unified theory 239
7. Flight from Body Movements 252
7.1. I. Thalberg's Theory of Action and the Accordion Effect 252
7.2. A. I. Goldman's Critique on the Identity-Thesis 257
7.3. Body Movement, Events as Actions and their Causation 267
7.4. On Hempel's Account 274

Part III
Radical Contextualism 279

8. The Elugardo-Problem 282
9. Analyzing Utterance Meaning 288
10. Indexicality 300

Bibliography 309
Detailed Table of Contents 320
Name Index  324
Subject Index  327


Since the seventies of the last century Donald Davidson, besides Willard van Orman Quine, has been the most famous American philosopher of analytic philosophy. Until the year 2004, more than nine volumes were published on his building a unified theory of thought, meaning and action (= unified theory) and its ontological implications (2004 (1980)).1 His unified theory has systematized parts of classical analytic philosophy. Davidson goes along with B. Russell's doctrine: "I propose to consider whether anything, and if so, what can be inferred from the structure of language as to the structure of world." (B. Russell 1950: 3) The unified theory gives classical analytical philosophy a new turn: "? it is plausible to hold that by studying the most general aspects of language we will be studying the most general aspects of reality" (Davison 1984 (1977): 201). Davidson does not go along with an analytical theory of meaning as put forward by R. Carnap according to Quine's critique of the two dogmas of empiricism. He considers this to be a consequence of breakdowns of classical analytic philosophy, the logical atomism (Russell, L. Wittgenstein Tractatus), and also of the semantics of logical empiricism of the Wiener Kreis, the so-called Sinnkriterium. The unified theory evolves from Quine's work; he has responded extensively to Davidson in his Pursuit of Truth (1992 rev. edn).

It is one of the main subjects of the theory of meaning and linguistic behavior to explain the ascription of propositional attitudes; evaluative attitudes like desires, intentions, hopes, moral convictions, views about duties, and obligations also count as propositional. Davidson calls all propositional attitudes thoughts because true beliefs play a significant role among them. They have semantic properties o r specifiable content. Understanding speech is bound by the ascription of propositional attitudes, that is, it is the way of making behavior intelligible. The problem is arising in the continuation of Quine's philosophy of language. He has shown that in principle there is an interconnection between belief and meaning, and our understanding of meaning is bound by sentence meaning. In this context, Davidson has developed a unified theory as a total theory of behavior, that is, a connected theory of desires and beliefs, and consequently of all propositional attitudes to explain both linguistic and non-linguistic behavior.

RI and its holistic truth-centered theory is initiated by radical translation (Quine) and leads us to a theoretical revised version of this fundamental problem in the theory of language. Speaking in terms of terminology, I call the finitely axiomatizable, semanticly compositional theory of truth: holistic truth-centered theory. This theory is not definitional or epistemic. At first I will give an outline of Davidson's philosophy with its systematic intent. This means that all the main themes of his philosophy of language, mind, rationality, and communication emerge from radical interpretation (RI). In the following steps I will analyze the hard core of his philosophy of language and action with respect to the relationship between interpretation (language)–rationality (charity)–triangulation born from the holistic truth-centered theory. This shows how the mental (attitudes), language, the social and the rest of the world are connected. It is my goal to make comprehensive that there is just no contingent connection between the concept of language given by RI and the distal theory of meaning (reference) and cognition (thought), that is, the epistemological turn in his philosophy. The epistemological turn in his philosophy as an anti-foundationalism is: the system of beliefs and other propositional attitudes is to relate to external circumstances not epistemologically but semanticly. From the triangulation point of view, the certitude of beliefs of our own propositional attitudes is not a foundation of knowledge. Consequently the foundation of interpretation, but not the foundation of knowledge, is that the interpreter has to treat first person authority of speakers.

It is to mention here that the key to link the theories of language, mental, action and epistemology is the coarse-grained concept of events. I will mention here that events are also significant for our theory of time. In contemporary philosophy the concept of event plays a dominating role in the philosophy of mind, language and action. Yet it seems to me that the concept of event in contemporary philosophy is too coarse and, at the same time, too fine-grained.

The step from RI to the theory of action and the rationalization of behavior is taken with Davidson's concept of reason and his ontology of events: actions are events and, as such, body movements that we redescribe with an intentional vocabulary. This shows that ontology, interpretation and the conceptual/linguistic dualism work together in his philosophy. Yet, in order to understand Davidson's modified concept of causal explanation of action it is useful to look back to the three background theories he claims to reject: Hart's ascriptivism, the logical-connection argument (A. I. Melden), and Danto's theory of basic acts. In this context I will also discuss the critiques of Davidson's identity thesis of I. Talberg and A. I. Goldman. I have reached the conclusion that the relationship of act-pairs A-A' like, for example, moving the hand and turning on the light, is not one of the acts themselves but between upshots (results). In this context I will introduce a radical theory of agency. An outline has been given in Preyer (2002b). Analyzing the route from RI to triangulation I expand the frame of reference of RI by taking steps to an analytical contextualism, and I take into account the epistemic capacity as the limit of so-called explanatory redescription, of linguistic behavior and actions.

In the analytical philosophy of langugage and semantics, since the nineties a schism has emerged between semantic minimalism, for example S. Somas (2002), E. Borg (2004) and H. Cappelen, E. Lepore (2005) on the one hand and radical pragmatism (contextualism), for example R. Carston (2002), F. Recanati (2004), S. Neale (2004) on the other hand. The reason for this aggravate disagreement is born from the analysis of sentence meaning (see Preyer 2006 b). What I called radical contextualism a long time ago is not a so-called original utterance centrism which argues that there is no significant basic set of content sensitive expressions and there is no autonomy of meaning. Therefore the technical term radical contextualism is misleading in the contemporary scene of the philosophy of language. I have not changed the terminiology in this book because for me it is a matter of intellectual honesty not to modify something theoretically only from an opportunistic point of view. Therefore it must be pointed out that I use the expression "contextualism" with a particular meaning, that is, we theorize the meaning of the utterance of a single sentence s by the sentences s* that just s implies. This is not to be constructed for a language in general, but for the uttered sentence. This way I will give RI a further turn. This turn is not truth-conditionally pragmatic or linguistically pragmatic in any of its versions. Furthermore I will also mention here the new researches on contextualism of K. Bach (2003). My old view was that linguistic communication and understanding only work if we assume any concept of linguistic meaning semanticly and not in essential pragmatically. This is not in any conflict with my use of the technical term in this book. We are in a continuous conversation about the problem of context sensitivity in semantics which gives the analysis of the relationship between semantic content a non-semantic content (G. Preyer and G. Peter eds. 2005a, 2006a). Every analysis of this relationship has to answer the question "On what level do we refer to the social frame of reference when understanding linguistic behavior (communication) basically?" That is not a trivial question because there may be some truth in "we are not born, speaking a language" (Davidson). The principle of autonomy of meaning is a hint to give us an answer.

We have to reckon that there is no total understanding and comprehensibility of linguistic behavior and consequently not of people either. Yet this is no defect of theories of interpretation and understanding in general. Both are a matter of degree, contexts and background theories. Just this is in harmony with the principle of autonomy of meaning. Therefore we are faced with the fact that the degree of understanding does not mean that we understand a natural speaker as a whole. We are black boxes in the case of successful interpretation also. I conclude this from Davidson's unified theory. Possibly this is a bit bulky to grasp. In Quine's philosophy it is valid: from the ontological point of view reference is scrutable, from the radical translation point of view (epistemological, behavioral) it is not. In a reversal of this: we are black boxes ontologically but not epistemologically, that is, if the ascription of attitudes works, the indeterminacy between the ascription of attitudes and the given interpretation of the speaker's words is limited (also in cases where discrepancies are). But from the point of view of radical interpretation we cannot discover the attitudes themselves, because from our theory of interpretation we conclude what attitudes are: attitudes cannot be located (principle of non-location). To be a black box ontologically takes also effect on rationalization behavior?the "climber-example"?because there is no way to explain how our attitudes cause an action that may also be caused in another way (Davidson 1980 (1973)): 79). That is my belief in contrast to many other colleagues. Yet it is to mention in this context that RI is not an inferential account.

It is in harmony with the autonomy of meaning that we link the mental, language and the social by compositionality (sub-sentential aspect). This is the new linguistic turn in philosophy partially followed from RI. Compositionality is built in on every stage of RI. This is the basic theory of the unification of both a theory of meaning and a theory of action. The language that we speak shows us sufficient structure to analyze the meaning of sentences and their parts; our thoughts and beliefs could not be completely false, so we have true world views at our disposal, and our actions and communications are not absolutely unsuccessful, therefore we have an understanding of people. The picture to link the mental, language and communication (the social) is built in compositionality because, if the theory of meaning is compositional, then (and only then) we ask how it is possible that the utterances of individual speakers are to redescribe by the specific compositional theory. Coping with the circle between belief and meaning by the evidence of holding a sentence true we find considerable insight into the nature of propositional attitudes, language and the mental. Therefore it follows from the truth-centered theory of RI as a semantic theory for a language that satisfaction conditions, compositionality, translation (RI) and intentional explanation work together. Firstly, a semantic theory assigns satisfaction conditions to semanticly valuable expressions and, in the same way, a theory of action assigns intentional contents to behavior. Such content is semanticly valuable as a quality of propositional attitudes. Secondly, the theory explains how the satisfaction conditions for complex linguistic expressions are determined by the satisfaction conditions of their parts. This explains us the nature of systematicity and productivity of natural languages. Thirdly, meaning is given by a successful translation, that is, in the procedure of RI. Here also pretheoretic intuitions come into play because speaking a language is also a skill. Fourthly, the unified theory applies the notion of content of attitudes to explain behavior intentionally. These issues of a semantic theory are accepted among the interpretive truth theorists. Philosophy of language has made progress in substance because we have realized that the traditional distinctions between "linguistic (conventional) meaning versus use", "truth-conditional versus non-truth-conditional meaning", and "context independence versus context dependence" do not work in principle. This is also something we can learn from the unified theory. I speak of understanding as knowing what something means. This implies skills also, and the knowledge of how to do something. An agent realizes his understanding in his first person authority immediately but an interpreter in a third person perspective has to theorize it. Yet this necessary, selective understanding of behavior is not contrary to the claim that making behavior intelligible is a matter of objective truth. And so our knowledge is grounded in truth. One central problem in philosophy of language must be mentioned, namely whether RI is possible and what consequence there is if there is no RI. The answer to the question "Is RI possible?" is dependent on how strong the epistemic restrictions of interpretation are assumed to be. And this leads to the question whether the concepts of meaning and propositional attitudes are theoretical in principle. Davidson's philosophy is significant in a transdisciplinary way. From the unified theory an orientation evolves to show us that the theories of language, communication, decision, epistemology and also ontology work hand in hand.

1 A part of Davidson's articles was published in: D. Davidson Vol. 1 1980, Vol. 2 1984, Vol. 3 2001, Vol. 4 2004, Vol. 5 2005. On Davidson's philosophy grounded in his theory of meaning and philosophy of language E. Lepore, Kirk Ludwig (2005); on Davidson's theory of language, action, and epistemology K. Ludwig ed. 2003, U. M. Zeglen ed. 1999, G. Preyer, F. Siebelt, and A. Ulfig eds. 1994, J. Brandl and W. Gombocz eds. 1989, E. Lepore, B. P. McLaughlin eds. 1986, E. Lepore ed. 1986, Vermazen, and M. B. Hintika eds eds. 1985, G. Evans, and J. H. McDowell eds. 1976; on Davidson's philosophy of mental J. Heil, and A. I. Mele eds. 1993; on Davidson's program M. Root, and J. Wallace (1982); on Tarski and Davidson D. Larson 1988; on an overview of Davidson's philosophy G. Preyer, F. Siebelt, A. Ulfig 1994, K. Ludwig 2003; on Davidson's theory of language E. Lepore 1982; on criticism B. Loar 1976; M. Dummett 1975, M. Platts 1979, J. J. Katz 1975; on logical form R. M. Martin 1978, W. K. Essler 1984; E. Lepore, K. Ludwig 2002 have given Davidson's semantics a further turn with an interpretive fulfillment theory, on further continuation and research G. Preyer, G. Peter eds. 2002a, on new research on the slingshot argument S. Neale 2001. On Neale G. Preyer, G. Peter (eds.). Forthcoming. Further volumes to follow may be expected. On a "Bibliography of Davidson's Publications", Ludwig 2003, 207-213, on a "Selected Commentary on Davidson," 214-15, on "Bibliographic References", 216-31; an "Interview with Donald Davidson by E. Lepore", 231-65. In: Davidson 2004.