Triangulation: From an Epistemological Point of View 9783110326833, 9783110326529

Citation preview

Maria Cristina Amoretti, Gerhard Preyer (Eds.) Triangulation From an Epistemological Point of View

Philosophische Analyse Philosophical Analysis Herausgegeben von / Edited by Herbert Hochberg • Rafael Hüntelmann • Christian Kanzian Richard Schantz • Erwin Tegtmeier Band 40 / Volume 40

Maria Cristina Amoretti, Gerhard Preyer (Eds.)

Triangulation From an Epistemological Point of View

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.

North and South America by Transaction Books Rutgers University Piscataway, NJ 08854-8042 [email protected] United Kingdom, Ireland, Iceland, Turkey, Malta, Portugal by Gazelle Books Services Limited White Cross Mills Hightown LANCASTER, LA1 4XS [email protected]

Livraison pour la France et la Belgique: Librairie Philosophique J.Vrin 6, place de la Sorbonne; F-75005 PARIS Tel. +33 (0)1 43 54 03 47; Fax +33 (0)1 43 54 48 18 www.vrin.fr

2011 ontos verlag P.O. Box 15 41, D-63133 Heusenstamm www.ontosverlag.com ISBN 978-3-86838-119-1 2011 No part of this book may be reproduced, stored in retrieval systems or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work Printed on acid-free paper FSC-certified (Forest Stewardship Council) This hardcover binding meets the International Library standard Printed in Germany by CPI buch bücher.de

CONTENTS

Introduction: Mind, Knowledge, and Communication in Triangular Externalism M. Cristina Amoretti and Gerhard Preyer 9 Part I: The Epistemological Turn

29

1. Triangulation and Philosophical Skepticism Claudine Verheggen

31

2. Triangulation between Externalism and Internalism M. Cristina Amoretti

47

3. Triangulation Triangulated Kirk Ludwig

69

4. Triangulation and Objectivity: Squaring the Circle? Adina L. Roskies

97

Part II: Communication and Environment

103

5. Knowing Me, Knowing You: Triangulation and Its Discontents Fredrik Stjernberg 105 6. Triangulation and the Beasts Dorit Bar-On and Matthew Priselac

121

7. Interpreting Assertions Sanford C. Goldberg

153

8

Part III: Philosophical Geography

177

8. The Short Happy Life of the Swampman: Interpretation and Social Externalism in Davidson Mario De Caro 179 9. The Externalism of Triangulation Gerhard Preyer

197

10. Triangulation in Action: A Rationalizing Proposal Ingvald Fergestad and Bjørn Ramberg

221

11. Triangulation and Philosophy: A Davidsonian Landscape Jeff Malpas

257

Contributors

281

Name Index

283

Subject Index

285

INTRODUCTION: MIND, KNOWLEDGE, AND COMMUNICATION IN TRIANGULAR EXTERNALISM M. Cristina AMORETTI (University of Genoa, Italy) Gerhard PREYER (Goethe-University Frankfurt am Main, Germany)

Well, that is just what we see when one of our triangular or other acquaintances comes toward us in Flatland. As there is neither sun with us, nor any light of such a kind as to make shadows, we have none of the helps to the sight that you have in Spaceland. If our friend comes closer to us we see his line becomes larger; if he leaves us it becomes smaller: but still he looks like a straight line; be he a Triangle, Square, Pentagon, Hexagon, Circle, what you will—a straight Line he looks and nothing else. (Abbott 1884: 16-17).

1. On the Naturalistic Turn and the Externalism/Internalism Debate about the Mind Since the mid-1960s, naturalism (physicalism, materialism) has had a great effect on American philosophy of mind, and more generally, on the analytic tradition. This trend is not self-evident and therefore requires further explanation. The successes of different sciences, such as biochemistry and neurophysiology, the behaviorism dominating psychology, the orientation of logical empiricism to natural sciences, and its critique of Scheinprobleme initiated this turn toward naturalism, which was underpinned by the overwhelming influence of mathematical logic in early twentieth-century philosophy. Since the mid-1950s, Wilfrid Sellars’ article “Empiricism and the Philosophy of Mind” (Sellars 1956), in which it is argued that mental states are neural states, has had particular significance. However, the work of Willard von Orman Quine, John C. Smart, and Ullin T. Place has also played a pivotal role for the naturalistic turn in the philosophy of mind.

10

In the 1960s, materialism was widely accepted. The critiques of the popular type-type-identity theory (as the one developed and defended by Smart) by philosophers such as Hilary Putnam (multi-realization of mental states) or Donald Davidson (anomalous monism), however, caused most materialists to abandon the type-type-identity theory in favor of some kind of token-token-identity theory. Among its other consequences, this move toward token-token-identity theory initiated the debate on mental causality.1 The reaction to type-type-identity theory was not only ontological, but also ideological (Burge 1993: 360). For instance, the answer to Putman’s above-mentioned critique gave birth to functionalism (analytical as well as scientific functionalism).2 It is worth stressing that functionalism and behaviorism both agreed that mental explanations are insufficient and should be replaced by non-mental explanations. Most functionalists assumed that their own account harmonizes with materialism. The central intuition of functionalism dictates that mental states (and events) are states within a causal (or functional) network in the mental architecture of a living organism. Mental states (and events) are to be specified exclusively as sequences (or places) of inputs and outputs, which are to be described in a nonmentalistic vocabulary. The theoretical model at the base of functionalism is the analogy of the mind as a computer program (Block 1995): the mind is considered the software that runs on the hardware of the brain (and, at least in theory, this mental software may be implemented by other kinds of hardware). Since the mid-1970s we have been confronted with the difficulties raised by two accounts of the philosophy of mind: internalism and externalism. For internalists, mental content is independent from the external environment, that is to say, mental content is narrow in principle. More precisely, they believe that the content of mental states is determined solely by “internal” features, that is, by something that is under the skin, inside 1

On mental causation and supervenience, see, for instance, Burge 1993, 2006; Heil J. and Mele 1993; Kim 1993, 1998; Rogler and Preyer 2001 2 On an analytic version of functionalism, see Armstrong 1968; Lewis 1972; Shoemaker 1984; on a scientific version, see Putnam 1967; Fodor 1981 had argued for a non-reductive version.

11

the body or the brain. Therefore, properties are regarded as intrinsic. Such a view is compatible with the basic tenets of functionalism and is defended by philosophers such as, for example, Jerry Fodor, Colin McGinn, Brian Loar, and John R. Searle, although Searle opposes functionalism. Moreover, it is important to mention the critiques of functionalism raised by Thomas Nagel, Ned Block, John R. Searle, and Frank Jackson, which have had a great influence in the philosophy of mind (Block 1978; Jackson 1982; Nagel 1974; Searle 1980). Since the 1990s, Gilbert Harman’s concept of representation (according to which representational states are nothing more than representational features) and his critique on the invert spectrum have also acquired a particular relevance (Harman 1990). Looking back to the philosophical tradition, we may classify René Descartes, John Locke, David Hume, Edmund Husserl, and Gottlob Frege, independent of their particular philosophies, as internalists. From the externalist point of view, however, mental content is to be individuated by the outside environment; more precisely, externalism argues that the content of mental states depends on or is individuated by external objects and events. Therefore, such content does not depend solely on internal states (or properties). That is, at least some properties of the content of thoughts are relational; they are external to the skin, to the body or to the brain. Externalists such as Hilary Putnam, John McDowell, Tyler Burge, Gareth Evans, and Donald Davidson argue that all content is wide and non-relational and agree with Putnam’s doctrine: ‘meanings’ just ain’t in the head (Putnam 1975a: 227, see also 223-227). Externalism is not a homogeneous position; we may distinguish, for example, between diachronic or synchronic externalism, social (linguistic) or physical (causal) externalism, and reductive or non-reductive versions of it (LePore and Ludwig 2005: 335-340). Burge’s version is a kind of social (linguistic, non-individualistic) externalism. Putnam, Davidson, and Dretske are physical (causal) externalists, even if in very different ways. For example, Dretske presents a diachronic version of externalism, whereas Davidson’s triangular externalism is more ambiguous: it may be seen either as synchronic (if we focus our attention on the social framework in which mental states are currently evaluated) or diachronic (if we underline the role of causal history for the determination of content). One may also distinguish between token physicalism, that is, every token of an

12

event is a physical event, and type externalism, that is, some types of mental states are to be individuated externally when we explain actions. More precisely, Cynthia Macdonald (Macdonald 1989, 1990, 1992) makes the distinction between 1. internalism, 2. externalism, 3. type externalism connected with token internalism, 4. type internalism connected with token externalism. According to Macdonald, 2. is a version of strong externalism, as externalism is strong if it rejects both type and token internalism, whereas 3. and 4. are versions of weak externalism, as externalism is weak if it rejects either type internalism alone or token internalism alone. On the motivation of externalism and internalism, Colin McGinn makes a similar distinction between strong and weak externalism (McGinn 1982a, 1982b, 1983, 1989). The former argues that some mental states are to be individuated by the properties of the environment, while the second argues that certain mental states are dependent on features of the body of the subject. Regardless of whether or not this is the case, the features occur in the environment of the thinker (speaker, agent). Burge’s definition of internalism (individualism) is a critique of strong externalism, and it is in harmony with weak externalism. From our point of view, it is obvious that strong externalism implies weak externalism (see also Edwards 1994). It is worth stressing here that internalism as a mind-body relationship should not be confused with ontological Cartesianism, nor should externalism be identified with the denial of ontological Cartesianism. The Cartesian view of the mind is that mental states are ontologically independent of the physical ones (ontological dualism). The turn from Cartesianism in the philosophy of mind is marked by the affirmation that there is no nonspatial mind substance (ontological monism). If Cartesianists hold that the mind is a substance ontologically independent from the physical, both internalists and externalists deny this ontological dualism, maintaining that the mind ontologically depends on the physical or coincides with it. However, according to the internalists, the physical must be identified solely with the body or the brain, whereas the externalists point out that the

13

physical may also comprise some features of the outside natural/social environment. Many philosophers agree that there are at least two problems with the individuation of content: first, the epistemic one, whether the thinker (speaker, agent) knows what she thinks, and second, the first person authority of the thinker (speaker, agent). The first problem is caused by the acquisition of knowledge from outside. This acquisition cannot be assumed a priori. The second problem is that first person authority means that the thinker (speaker, agent) has special authority, knowing the content of her own intentional states, even if content is to be individuated externally (see, for instance, Brown 2004; Ludlow and Martin 1998; Wright et al. 1998). More precisely, the thinker (speaker, agent) knows the content of her own intentional states immediately (directly, a priori), without the need of any evidence or inference from utterances and actions. Thus, her non-evidential knowledge is authoritative, in contrast to knowledge acquired by evidence (Davidson 1984, 1987, 1988, 1991). This feature of Cartesianism—that is, the Cartesian view of first person thinking, namely, that the thinker (speaker, agent) has a special epistemic access to the content of her own mental states, intentional or not—is not disputed among many philosophers including most externalists. More precisely, Cartesian first-person thinking means that the subject knows her own occurring first-order mental states directly, without any conscious inference; this contrasts with the knowledge of mental states from the third-person point of view, which is inferred from utterances and actions. Although the Cartesian intuition about the nature of thought is compatible with internalism, there is a potential conflict with externalism. For Cartesians, the nature of thought is considered from the first-person point of view: the thinker (speaker, agent) knows the content of her own mental states and is authoritative about the knowledge of the content of these states. However, first person authority seems possible only if thought content is narrow content (as internalists hold). Externalism, in contrast, argues that content is, in principle, to be individuated by features of the external environment outside the thinker (speaker, agent) and, thus, that there is no narrow content. Even if the problem of non-evidential knowledge may seem particularly relevant for externalism, it is not one of externalism itself.

14

According to Davidson (1989), for example, the problem is not due to externalism itself but to the assuming of inner entities (abstract entities, Frege-thoughts, or similar) and to the thesis that content is determined by relations with such entities. If we argue in a Fregean manner, then we must ask, “What is the individuation of such entities?” The Frege’s turn is that thought is individuation dependent. Therefore, the individuation condition that is specified by external relations cannot be known from either the third- or first-point of view. British empiricists (such as Locke, Berkeley, and Hume) have the same problem that Cartesians have. They have not shown how inner entities stand in relation to the outer entities that cause them. The solution, according to Davidson, is doing away with these inner entities. In general, to answer the problem raised by first-person authority, externalists argue, grosso modo, in the following way: if we consider any first-order thought (mental content) such as “p”, it is fixed externally and is therefore a feature of the external world; however, the individuation conditions that fix the first-order thought are the same external conditions that also fix the second-order thought, “I am thinking that p”. That is to say, the second-order thought contains or inherits the first-order thought. If this reasoning is true, there is no room for error, and thus first-person authority is maintained. Moreover, another problem immediately arises. On the one hand, the ascription of intentional states to others, as well as the existence of other minds, can only be established by contingent correlations between mental states and behavior from the third person point of view, that is, ascriptions to others require evidence. On the other hand, first person authority is immediate, that is, we need no evidence for self-ascriptions. The difference between an evidential and a non-evidential basis for the ascription of mental states makes it difficult to answer the following question: “Why do the mental predicates mean the same in ascriptions to ourselves and to others?—I, you, he know(s)”.3

3

We is a special case. It also makes sense to ascribe knowledge, attitudes, and conscious states to a group.

15

2. On Davidson’s triangular externalism Davidson introduced the notion of triangulation in “Rational animals” (1982): If I were bolted to the earth, I would have no way of determining the distance from me of many objects. I would only know they were on some line drawn from me towards them. I might interact successfully with objects, but I could have no way of giving content to the question where they were. Not being bolted down, I am free to triangulate. Our sense of objectivity is the consequence of another sort of triangulation, one that requires two creatures. Each interacts with an object, but what gives each the concept of the way things are objectively is the base line formed between the creatures by language (Davidson 1982: 105).

This brief passage conceals all of the basic elements of the theory of triangulation: according to Davidson, in order to have any thought (and language), there must be (at least) two creatures, a speaker and an interpreter, linguistically interacting on the background of a public shared world. Although the theory of triangulation has become a central notion in Davidson’s whole philosophy, it is particularly crucial to characterize his own kind of externalism about mental content, that is, triangular externalism. According to Davidson, in fact, our thought (and language) emerges from the basic interaction involving a speaker, an interpreter, and the outside shared world. It is important to understand why that is so. The briefest answer is that the process of triangulation is, in principle, necessary to account for two elements that are both basic to propositional thought (and language): on the one hand, empirical content and, on the other, the concept of objectivity (see Davidson 1992, 1999a). More precisely, triangulation would explain how our beliefs about the external world, that is, our empirical beliefs, acquire their empirical (objective) content, and how the concept of objectivity may eventually emerge. It is quite easy to understand that empirical content is necessary to thought (at least if thought must be about something in the outside world), whereas the necessity of the concept of objectivity for the possession of thought is far more controversial.

16

According to Davidson, in the most basic situations, what determines (at least in part) the empirical content of a belief about the external world is its “typical” cause, but in order to identify such a cause, a single creature is not sufficient because she is not able to single out the relevant cause among all the possible causes (which include, for instance, both proximal and distal causes). With the introduction of (at least) a second creature— the interpreter—the situation radically changes: because the relevant cause must be common to the speaker and the interpreter, it must be situated in the shared public world, and thus, it must be a distal cause. The relevant cause may also be narrowed due to the presence of a second creature: the relevant cause is what typically causes relevantly similar responses by both the speaker and the interpreter, and thus, it may be found where the two lines connecting, respectively, speaker and world, and interpreter and world do intersect. It is important to note that, according to Davidson, the relevant cause may be successfully identified only in a linguistic framework (that is to say, only language can solve once and for all the underdetermination of the relevant cause), and thus, thought and language are deeply intertwined from the very beginning (Davidson 1999a: 130). Moreover, content is objective in the sense that (in most cases) it is true or false independent of the existence of a thought about it or of a thinker. Moving to the concept of objectivity, Davidson believes that a thinker—in order to be a real thinker—must be aware of the objectivity of thought, namely, that what she believes may be either true or false, and thus, she must posses the concept of objectivity. This concept is clearly linked to that of error: in order to understand the former, a creature should be able to understand that she can be mistaken. On this point, Davidson follows Ludwig Wittgenstein’s idea that one would not have the concept of error if one would not be (or at least have been) in interaction with other creatures similar to us. Because the process of triangulation, involving two creatures with the backdrop of a public shared world, is the most basic example of social interaction, it is easy to understand why this process is necessary to the concept of objectivity. Again, it is worth noting that language is also needed for the emergence of the concept of objectivity: only linguistic creatures can have it. The interdependence of thought and language may be seen as a consequence of Davidson’s strong anti-reductionism (intentional mental states

17

and concepts cannot be reduced to physical ones), but it makes it difficult, if not impossible, to explain the emergence of thought and language in scientifically acceptable terms (Amoretti 2009). Moreover, triangular externalism is revealed to be quite different from other kinds of externalism about mental content. It is a form of physical (or causal) externalism, but the presence of an interpreter is necessary to identify the relevant typical cause determining mental content. As Davidson would say, there is a social element (the interpreter) that enters into the causal process of content determination. The need for the social element is quite interesting because it reveals some pivotal features of Davidson’s triangular externalism. First, the theory of triangulation does not simply require that the speaker’s utterances may be possibly interpretable by a potential interpreter but rather implies the stronger claim that the speaker’s utterances are really interpreted by an actual interpreter. Without an interpreter concretely triangulating with the subject, there can be no content and thus no propositional thought and no language. Thus, thought and language are public in a very strong sense: the very possibility of thought and language emerges together with a community. Second, externalism, holism, and rationality are deeply connected: if the relevant typical cause is determined through the process of triangulation, then it is contaminated by holistic and rational constraints from the very beginning. Content is thus determined by both causal and holistic factors (that is to say, external elements are necessary but not sufficient for determining content). As a corollary, it is worth noting that externalism concerns (directly or indirectly, through holistic relations) the whole thought, not only some special thoughts, such as those concerning natural kinds. Triangular externalism has attracted numerous critics from the very beginning. To begin with, Davidson’s attempt at determining empirical content through the process of triangulation has been objected to in at least three different ways. First, the presence of the interpreter may cast doubt on the objectivity of content (Pagin 2001). Second, a shared cause may be considered as ambiguous as an unshared one (Føllesdal 1999; LePore and Ludwig 2005; Verheggen 1997). Third, we may question the implicit assumption that the speaker and interpreter perceive the same similarities (Fennell 2000). Other doubts have been raised about the idea that triangu-

18

lation is necessary for the concept of objectivity. For example, it may be argued that a single creature comparing past and present experiences is sufficient for the emergence of the concept of objectivity (Briscoe 2007; Child 2001; Engel 2001; Heil 1992; Montminy 2003). Moreover, the possession of the concept of objectivity (especially if it is regarded as involving the concept of truth) seems too strong of a requirement for having thought. Finally, a more general objection: given that we need to introduce language (and thus thought itself) to understand why triangulation is necessary for thought, why should we consider triangulation necessary at all? (Sinclair 2005). In addition to the above problems, it has also been argued that certain basic characteristics of Davidson’s externalism are themselves highly contentious, making triangular externalism an intrinsically instable theory. On the one hand, some critics think there is an irremediable tension between the interpreter’s point of view and the role of causal history for the determination of content. On the other hand, they see a conflict between the holistic character of content and its externalist individuation (De Caro 1998; Hahn 2003; Stjernberg 2002). To conclude, it is worth stressing that in order to evaluate triangular externalism from an epistemological point of view, it would be important to answer the following question: how does externalism truly affect Davidson’s epistemology? We may want to divide this issue into three subproblems: what are the consequences of triangular externalism for (1) our knowledge of our own mental states, (2) our knowledge of others’ mental states, and (3) our knowledge of the external world? For what concerns question (1), Davidson faces a problem common to all varieties of externalism, namely, that of reconciling the external nature of content with first person authority. As we have seen, a general reply is that what determines the content of our first-order thoughts, whatever it is, also determines the content of our second-order thoughts. But, of course, it is not sufficient to explain the asymmetry between first-person and thirdperson knowledge (Davidson 1984, 1987, 1988, 1989). On question (2), Davidson should explain how it is possible to know others’ mental states: because content is determined not only causally, but also holistically, and obviously it cannot be the case that two different subjects share exactly the same holistic constraints about any given content, it

19

cannot be the case that they share the same content. In this situation, as Michael Dummett notably argued, it is difficult to understand how communication would still be possible (Davidson 1986, 1994; Dummett 1986). Regarding question (3), it is worth mentioning that Davidson has explicitly stated that his own kind of externalism regarding mental content has the incredible virtue to block radical skeptics (or, at least, to tell the skeptic “to get lost”) and, thus, to guarantee our knowledge of the external world (Davidson 1983, 1990, 1999b, 1999c, 1999d). All of these issues regarding the process of triangulation and triangular externalism are variously examined and discussed by the contributors to this volume. 3. Triangulation in debate The question of objective knowledge and philosophical skepticism about the external world is scrutinized by both Claudine Verheggen in “Triangulation and Philosophical Skepticism” and M. Cristina Amoretti in “Triangulation between Externalism and Internalism”. Verheggen argues that those who analyzed Davidson’s anti-skeptical argument have often paid insufficient attention both to the claim, made explicit in the triangulation argument, that the conditions on belief-attribution are also conditions of belief-possession and to the lessons to be drawn from the rejection of the scheme/content distinction. According to Verheggen, taking into consideration these two most important elements of Davidson’s philosophy allows us to make the anti-skeptical claim that “belief is in its nature veridical” and, hence, to reach a strong anti-skeptical conclusion. Amoretti considers Davidson’s triangular externalism in relation both to his coherence theory of justification and to his anti-skeptical argument. On the one hand, she considers the claim that content externalism cannot be compatible with epistemic internalism and argues against the notion that there is no tension between triangular externalism and Davidson’s epistemic internalism. On the other, she agrees that triangular externalism supports the thesis that “belief is in its nature veridical” but also points out that it is probably not enough to satisfactorily answer the skeptic. The difficulties of triangulation in accounting for the two elements that Davidson considers essential to propositional thought—empirical content

20

and the concept of objectivity—are evaluated by Kirk Ludwig in “Triangulation Triangulated” and by Adina L. Roskies in “Triangulation and Objectivity: Squaring the Circle?” According to Ludwig, Davidson fails to show that the idea of triangulation is necessary for both the above elements. First, it is not required for the concept of objective truth because communication is not essential for developing the idea of contrasting perspectives necessary for the concept of objectivity. With respect to the problem of determining the empirical content, Ludwig shows that appeal to this sort of triangulation does not help with the problem of underdetermination of thought and meaning by the patterns of causal relations we stand in to the environment. Roskies concentrates her critiques on the concept of objectivity. According to her, triangulation cannot provide the logical ground for the construction of this concept because, in order for triangulation to get off the ground, the parties involved in the triangle must already have a notion that the world is objective, or mind-independent, namely, they must already have a concept of objectivity. As a consequence, it is not clear that a second person, or language, is needed to fix mental content. Moreover, Roskies concludes, triangulation fails to show that language is necessary for thought. Fredrik Stjernberg, in “Knowing Me, Knowing You: Triangulation and Its Discontents”, also claims that Davidson’s arguments derived from triangulation do not work as intended: the second person does not have to be invoked to ensure objectivity of thought or mental content. Moreover, Davidson’s conclusions seem to conflict with another argument from triangulation that, showing how objective, subjective, and intersubjective knowledge would be interacting in a concrete interpretational situation, aims to demonstrate that there is no privileged type of knowledge and that one must have all three types or none. According to Stjernberg, this emphasis on the interwoven nature of knowledge has the merit of locating knowledge where it actually should be located. We have said that triangulation, in Davidson’s view, is taken to be a necessary but not a sufficient condition for thought, as language is also necessary. This account is clearly circular and succumbs to the pivotal problem of explaining, in naturalistic and scientifically acceptable terms, the very emergence of thought (and language). Some attempts to fill the gap between the pre-cognitive realm and the full-cognitive one are pro-

21

posed by Dorit Bar-On and Matthew Priselac in “Triangulation and the Beasts”. They show that Davidson finds no room for a conceptual middleground between “pure” triangulation in the animal world (necessary but not sufficient for thought), and reflective triangulation between speakers. Bar-On and Priselac take a fresh look at Davidson’s “continuity skepticism” (as they label it) and offer some reflections on a distinct form of behavior that we humans share with the beasts: expressive behavior. They argue that the intersubjective interactions among creatures capable of expressive behavior can exemplify “intermediate triangulation” and support at least proto-objectivity. The relations between speaker, interpreter and the outside environment that are involved in the process of triangulation are analyzed by Sanford C. Goldberg, in “Interpreting Assertions”, from a perspective quite different from the above ones. More precisely, Goldberg explores the task of interpretation in cases involving sincere assertion and make two claims: first, that assertion has an epistemic norm; and second, that this has implications for the nature of what Donald Davidson called “the task of interpretation” in general and “triangulation” in particular. On the assumption that assertion has an epistemic norm (and that this is mutually familiar to all), Goldberg maintains we can justify a methodological injunction that requires the interpreter to exhibit a sort of “epistemological charity” in her rendering of the speech she is interpreting. The very nature of Davidson’s triangular externalism is then thoroughly analyzed and its common understanding partially revised by Mario De Caro in his “The Short Happy Life of the Swampman: Interpretation and Social Externalism in Davidson”. The author argues that even if in the beginning Davidson advocated a mixed form of externalism—one that, besides a strong physical variety of that view, also encompassed a peculiar form of social externalism, according to which interpretability offers necessary and sufficient conditions for the attribution of contents—he amended his own earlier view in his final years. More precisely, it is shown that because Davidson’s mixed form of externalism was unavoidably unstable, he finally decided to abandon the strong version of physical externalism he had defended. Davidson’s move from the truth-centered theory of radical interpretation to epistemology externalized is not a contingent one because in his

22

view there are no intensional entities, such as truth bearer, Frege-Sinne, propositions, and attributes. From the global third-person stance of radical interpretation comes the main question: how are we to individuate the entities of propositional attitudes that the speaker refers to? Following the interpretation of Davidson’s philosophy of Ernie Lepore and Kirk Ludwig (2005) as a keystone of the understanding of the unified theory of thought, meaning, action and evaluation, Gerhard Preyer, in “The Externalism of Triangulation, shows that charity (basic rationality) does not solve the task of interpretation of linguistic behavior. The constraints of interpretation also take effect in Davidson’s externalism of triangulation. Preyer resystemizes the perspectives of triangulation and shows that the common causes cannot close the triangle between speaker, interpreter, and the external environment within both are embedded. This leads him back to the constraints of interpretation. He concludes that both radical interpretation and the individuation of the content of our perceptual beliefs by directly caused objects and events are impossible. The epistemic restrictions of interpretation are not to eliminate by the truth-theory and its axioms because the truth-predicate cannot filter out the epistemic qualification in the procedure of making behavior intelligible. The notion of triangulation is also examined and criticized by Ingvald Fergestad and Bjørn Ramberg who, in their “Triangulation in Action: A Rationalizing Proposal”, suggest an alternative project in which triangulation may instead successfully serve. Their project is brought into view emphasizing those aspects of Davidson’s philosophy that resonate with Robert Brandom’s inferentialism, that is, triangulating the more traditional rendering of Davidson with Brandom’s inferentialist position. This move allows them to shed new light onto a “Radical Davidson”, an idealization of some tendencies that exist in Davidson’s work but that are overwhelmed by the naturalistic ones. More precisely, they conceive “Radical Davidson” as concerned with construing the conceptual as a certain kind of dynamic interaction and triangulation as a useful tool to make a point about what kind of dynamic interaction this might be. Finally, the contribution of Jess Malpas, “Triangulation and Philosophy: A Davidsonian Landscape”, is not focused mainly on triangulation itself but carefully explores some of the far-reaching connections— conceptual as well as historical—linking triangulation with a wide body of

23

philosophical ideas and ways of thinking that are not limited within the sole analytic tradition but also involve the pragmatist, phenomenological, idealist, and hermeneutic traditions. Hence, Malpas provides a sketch of the territory or region within which Davidson can be located—a territory or region that, in his own work, is encompassed within the idea of “philosophical topography” or “topology”. His aim is twofold: not only to shed new light on Davidson’s position as such but also to open up a different mode of philosophical proceeding than that which is common in the analytic tradition. The project was initiated and planned by M. Cristina Amoretti (University of Genova, Department of Philosophy, Italy)4 and Gerhard Preyer (ProtoSociology, Goethe-University Frankfurt am Main, Germany). The editors wish to thank all contributors and Frank Hüntelmann for his support in realizing the project. Our contributions show an open-minded view of a free intercourse of ideas between philosophers. The collection brings together contributions from philosophers with diverse points of view and, as such, it illuminates the lively and ongoing debate on mind, internalism, externalism, and epistemology with regard to adjoining disciplines such as, for instance, cognitive sciences and the philosophy of language. References Abbott, E.A. 1884, Flatland: A Romance of Many Dimensions. Oxford: Oxford University Press, 2006. Amoretti, M.C. 2009, “Comunicazione Preverbale e Razionalità”. In D. Gambarara and A. Givigliano (eds.), Origine e Sviluppo del Linguaggio, tra Teoria e Storia. Roma: Aracne Editrice, 291-301. Armstrong, D.M. 1968, A Materialist Theory of the Mind. London: Routledge. Block, N. 1978, “Troubles with Functionalism”. In C.W. Savage (ed.), Minnesota Studies in the Philosophy of Science, vol IX. Minneapolis: Univeristy of Minnesota Press.

4

M. Cristina Amoretti gratefully acknowledges financial support for her research from Grant POSDRU/89/1.5/S/63663.

24

Block, N. 1995, “The Mind as the Software of the Brain”. In D. Osherson, L. Gleitman, et al. (eds.), An Invitation to Cognitive Science. Cambridge: MIT Press. Briscoe, R.E. 2007, “Communication and Rational Responsiveness to the World”. Pacific Philosophical Quarterly, 88, 135-159. Brown, J. 2004, Anti-Individualism and Knowledge. Cambridge, Mass.: MIT Press. Burge, T. 1993, “Mind-Body Causation and Explanatory Practice”. In J. Heil and A.R. Mele (eds.), Mental Causation: Oxford University Press; repr. in Burge 2007 (from which we quote). Burge, T. 2006, “Postscript to ‘Mind-Body Causation and Explanatory Practice’”. In Burge 2007, 363-382. Burge, T. 2007, Foundations of Mind. Philosophical Essays vol. 2. Oxford: Oxford University Press. Child, W. 2001, “Triangulation: Davidson, Realism and Natural Kinds”. Dialectica, 55 (1), 29-49. Davidson, D. 1982, “Rational Animals”. Dialectica, 36, 317-327; repr. in Davidson 2001 (from which we quote). Davidson, D. 1983, “A Coherence Theory of Truth and Knowledge”. In D. Henrich (ed.) Kant Oder Hegel. Stuttgart: Klett-Cotta, 423-438; repr. in Davidson 2001. Davidson, D. 1984, “First Person Authority”. Dialectica, 38, 101-111; repr. in Davidson 2001. Davidson, D. 1986, “A Nice Derangement of Epitaphs”. In R. Grandy and R. Warner (eds.), Philosophical Grounds of Rationality. Oxford: Oxford University Press, 156-174; repr. in Davidson 2005. Davidson, D. 1987, “Knowing One’s Own Mind”. Proceedings and Addresses of the American Philosophical Association, 60, 441-458; repr. in Davidson 2001. Davidson, D. 1988, “The Myth of Subjective”. In M. Krausz (ed.) Relativism. Interpretation and Confrontation. Notre Dame: Universtiy of Notre Dame Press, 159-172; repr. in Davidson 2001.

25

Davidson, D. 1989, “What Is Present to the Mind?”. In J. Brand and W.L. Gombocz (eds.), The Mind of Donald Davidson. Amsterdam: Rodopi, 3-18; repr. in Davidson 2001. Davidson, D. 1990, “Epistemology Externalised”. Análisis filósofico, 10, 1-13; repr. in Davidson 2001. Davidson, D. 1991, “Three Varieties of Knowledge”. Philosophy, 66, 156-166; repr. in Davidson 2001. Davidson, D. 1992, “The Second Person”. In P. French, T. Uehling, et al. (eds.), The Wittgenstein Legacy, Midwest Studies in Philosophy, vol 17. Minneapolis: University of Minnesota Press, 255-267; repr. in Davidson 2001. Davidson, D. 1994, “The Social Aspect of Language”. In B. Mcguinness (ed.) The Philosophy of Michael Dummett. Dordrecht: Kluwer, 1-16; repr. in Davidson 2005. Davidson, D. 1999a, “The Emergence of Thought”. Erkenntnis, 51, 7-17; repr. in Davidson 2001. Davidson, D. 1999b, “Reply to Genova”. In L.E. Hahn (ed.) The Philosophy of Donald Davidson. Chicago: Open Court, 192-194. Davidson, D. 1999c, “Reply to Nagel”. In L.E. Hahn (ed.) The Philosophy of Donald Davidson. Chicago: Open Court, 207-210. Davidson, D. 1999d, “Reply to Stroud”. In L.E. Hahn (ed.) The Philosophy of Donald Davidson. Chicago: Opern Court, 162-166. De Caro, M. 1998, Dal Punto di Vista dell’interprete. Roma: Carocci. Dummett, M. 1986, “‘A Nice Derangement of Epitaphs’: Some Comments on Davidson and Hacking”. In E. Lepore (ed.) Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson. Oxford: Blackwell, 459-476. Edwards, S.D. 1994, Externalism in the Philosophy of Mind. Brookfield, Vt.: Ashgate. Engel, P. 2001, “The Norms of Thought: Are They Social?”. Mind & Society, 2, 129-148.

26

Fennell, J. 2000, “Davidson on Meaning Normativity: Public or Social”. European Journal of Philosophy, 8 (2), 139-154. Fodor, J. 1981, Representation. Cambridge, Mass.: MIT Press. Føllesdal, D. 1999, “Triangulation”. In L.E. Hahn (ed.) The Philosophy of Donald Davidson. Chicago: Open Court, 719-728. Hahn, L.E. 2003, “When Swampmen Get Arthritis: Externalism in Burge and Davidson”. In M. Hahn and B.T. Ramberg (eds.), Reflections and Replies: Essays on the Philosophy of Tyler Burge. Cambridge, Mass.: MIT Press, 29-58. Harman, G. 1990, “The Intrinsic Quality of Experience”. Philosophical Perspectives, 4, 31-52. Heil, J. 1992, The Nature of True Minds. Cambridge, Mass.: Cambridge University Press. Heil, J. and Mele, A.R. (eds.) 1993, Mental Causation. Oxford: Oxford University Press. Jackson, F. 1982, “Epiphenomenal Qualia”. Philosophical Quarterly, 32, 127136. Kim, J. 1993, Supervenience and Mind. Cambridge: Cambridge University Press. Kim, J. 1998, Philosophy of Mind. Boulder, Col.: Westview Press. Lepore, E. and Ludwig, K. 2005, Donald Davidson. Meaning, Truth, Language, and Reality. Oxford: Clarendon Press. Lewis, D. 1972, “Psychophysical and Theoretical Identification”. Australasian Journal of Philosophy, 50, 249-258. Ludlow, P. and Martin, N. 1998, Externalism and Self-Knowledge. Stanford: CSLI. Macdonald, C. 1989, Mind-Body Identity Theories. London: Routledge. Macdonald, C. 1990, “Weak Externalism and Mind-Body Identity”. Mind, 99 (395), 387-404.

27

Macdonald, C. 1992, “Weak Externalism and Psychological Reduction”. In D. Owain, M. Charles, et al. (eds.), Reduction, Explanation and Realism. Oxford: Oxford University Press. McGinn, C. 1982a, The Character of Mind: An Introduction to the Philosophy of Mind. Oxford: Oxford University Press. McGinn, C. 1982b, “The Structure of Content”. In A. Woodfield (ed.), Thought and Object. Oxford: Clarendon Press, 207-258. McGinn, C. 1983, The Subjective View: Secondary Qualities and Indexical Thoughts. Oxford: Oxford University Press. McGinn, C. 1989, Mental Content. New York: Blackwell. Montminy, M. 2003, “Triangulation, Objectivity and the Ambiguity Problem”. Critica, 35, 25-48. Nagel, T. 1974, “What Is It Like to Be a Bat?”. Philosophical Review, 83, 435456. Pagin, P. 2001, “Semantic Triangulation”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 199-212. Putnam, H. 1967, “The Nature of Mental States”. In Putnam 1975b, 429-440. Putnam, H. 1975a, “The Meaning of ‘Meaning’”. In Putnam 1975b, 215-271. Putnam, H. 1975b, Mind, Language, and Reality. Cambridge: Cambridge University Press. Rogler, E. and Preyer, G. 2001, Materialismus, Anomaler Monismus und Mentale Kausalität. Frankfurt a.M.: Humanities Online. Engl. transl. Anomalous Monism and Mental Causality. On the Debate on Doland Davidson’s Philosophy of the Mental. Frankfurt a.M.: Humanities Online, 2001. Searle, J. 1980, “Minds, Brains and Programs”. Behavioral and Brain Sciences, 3, 417-457. Sellars, W. 1956, “Empiricism and the Philosophy of Mind”. In H. Feigl and M. Scriven (eds.), Minnesota Studies in the Philosophy of Science. Minneapolois: University of Minnesota Press. Shoemaker, S. 1984, “Functionalism and Qualia”. In Identity, Cause and Mind. Cambridge: Cambridge University Press.

28

Sinclair, R. 2005, “The Philosophical Significance of Triangulation: Locating Davidson’s Non-Reductive Naturalism”. Metaphilosophy, 36 (5), 708-727. Stjernberg, F. 2002, “Sulla Combinazione di Olismo ed Esternalismo”. In M. Dell’Utri (ed.) Olismi. Macerata: Quodlibet, 231-248. Verheggen, C. 1997, “Davidson’s Second Person”. The Philosophical Quarterly, 47 (188), 361-369. Wright, C., Smith, B.C., et al. (eds.) 1998, Knowing Our Own Minds. Oxford: Clarendon Press.

PART I THE EPISTEMOLOGICAL TURN

TRIANGULATION AND PHILOSOPHICAL SKEPTICISM Claudine VERHEGGEN (York University, Toronto, Canada)

1. Could I be mistaken in thinking that I am now writing this on a white sheet of paper, that my watch says it is 3:40 in the afternoon, that my coffee is getting cold, that the leaves are falling from the trees outside my window and that it has started raining again? The Cartesian skeptic that I will be momentarily impersonating would resolutely answer that I could. More than that, she would assert that I could have all the beliefs I currently have about the world around me, as well as all my current experiences seemingly of the world around me, even if all of those beliefs were false and none of those experiences were veridical. I could, after all, be dreaming in the middle of a beautiful starry night. You may retort that dreaming would not prevent me from having at least general true beliefs and even some knowledge about the objects and events that surround me. But that would be assuming that the distinction I am making between dreaming and waking experiences is not itself the product of my imagination, that I have reason to think that my experiences sometimes are waking experiences. But the Cartesian skeptic that I have now become claims to have no such reason. For what could it be? I could not appeal to any aspect of my sensory experiences, since their reliability is precisely what is in question. Besides, those experiences as well as all of my thoughts and beliefs could have been caused by an evil demon in such a way that they may in fact correspond to nothing in the world around me, indeed, in such a way that this world may not even exist. Thus the hypothesis from which stems the Cartesian skeptical challenge: all of my beliefs about the world around me could be false. It is noteworthy that Descartes, when he was himself entertaining this hypothesis and its consequent doubts, never paused to ask what might make it possible for him to have all those beliefs seemingly about things beyond him—for all he tells us, they might literally have popped out of nowhere or at best (or worst) have been directly implanted in us by some supernatural being. But this question, which first exercised Kant, was the

32

central philosophical preoccupation of Donald Davidson. How can we think in the way we do about the world around us? Davidson asked this question, initially, not because he hoped, as Kant did, that a proper answer to it would block Cartesian skepticism, but because his primary philosophical interest was in the explanation of human intentional behavior, and the possession of thoughts and propositional attitudes is at the centre of this. Only later did he come to realize that the account of intentional content he had given has anti-skeptical consequences. The account is in fact such that, if it is right, the skeptic’s hypothesis that I could have all the beliefs I have about the world around me even if they were false is mistaken. Such beliefs can only be by and large true beliefs; otherwise I would not have them. As Davidson writes, «as long as we adhere to the basic intuition that in the simplest cases words and thoughts refer to what causes them, it is clear that it cannot happen that most of our plainest beliefs about what exists in the world are false» (1990: 196). «Belief is in its nature veridical» (1983: 146). How did he reach such a strong conclusion?1 It is a conclusion which Davidson thought he had first reached on the basis of his considerations concerning radical interpretation, the interpretation of another’s utterances and propositional attitudes from scratch, that is, assuming no prior understanding of the interpretee’s language and no (detailed) knowledge of her beliefs and other propositional attitudes (see Davidson 1973). But I shall start by rehearsing the later and more fully developed supporting argument, namely, the triangulation argument to the effect that only a person who has interacted with at least another and the world they share could be a thinker and a speaker. My primary purpose in this paper is not, however, to assess the triangulation argument itself but to investigate what precise anti-skeptical conclusion the argument warrants. For it has been argued, by Barry Stroud, that the only claim Davidson can make on its basis is that «belief-attribution is in its nature largely truthascribing» (Stroud 1999: 155), and not that belief is in its nature veridical. Therefore, according to Stroud, Davidson’s attempt to deal with the skeptic by denying the skeptical hypothesis does not succeed. Still, Stroud further argues, the weaker claim is strong enough to block philosophical skepticism. For, even though it does not allow us to tell the skeptic that our be1

Maybe too strong to be true, according to, e.g., Thomas Nagel (1999).

33

liefs about the external world are largely true, it does allow us to prevent the skeptic from asking the question whether they are.2 I shall argue that Stroud pays insufficient attention to the fact that Davidson’s aim is to provide not just conditions of belief-attribution but also conditions of belief-possession. This is what allows Davidson to make a stronger anti-skeptical claim, and hence to draw a stronger anti-skeptical conclusion, than those allowed by Stroud. Properly understanding the goal of the triangulation argument also enables us to address two additional complaints Stroud has about the stronger claim. One is that drawing it would involve adopting a standpoint on our knowledge of the world that Davidson has shown does not exist. The other is that, even if we could draw the stronger claim, it would not imply that we have any reason to believe anything about the world, and so it would not be sufficient to refute the skeptic. What, at bottom, these complaints amount to, I shall argue, is a rejection of Davidson's externalist account of content. 2. The triangulation argument starts with semantic externalism which Davidson declares in 1990 he has endorsed for some thirty years by insisting that the contents of our earliest learned and most basic sentences (‘Mama’, ‘Red’, ‘Fire’, ‘Gavagai’) must be determined by what it is in the world that causes us to hold them true. It is here, I have long claimed, that the ties between language and the world are established and that central constraints on meaning are fixed; and given the close connections between thought and language, analogous remarks go for the contents of the attitudes (Davidson 1990: 200).

Hence the general externalist statement: «the contents of our thoughts and sayings are partly determined by the history of causal interactions with the 2

Where I talk of two possible anti-skeptical claims, one weaker and one stronger, Stroud talks of two readings, one weaker and one stronger, of “belief is in its nature veridical”, which is the way Davidson expresses his anti-skeptical claim. Stroud wonders whether this should not be read as saying something weaker than it appears to. But the fact is that, as Davidson makes it clear himself (in 1999a: 162), he means it in a strong way. Hence my talk of two possible anti-skeptical claims and the question which is warranted by Davidson’s argument.

34

environment». They are partly determined by «what has typically caused similar thoughts» and sayings (1990: 200-1). One might immediately protest here that assuming this kind of externalism, according to which objects and events in one’s environment play a constitutive role in determining the contents of one’s thoughts and utterances, is really begging the question against the skeptic. My response to this will be brief. Davidson’s externalism was initially prompted by the failure of intensionalist accounts of meaning to say anything philosophically interesting about the nature of meaning (see Davidson 1967) and by his conviction that the semantic features of language and hence of thought are public features (see Davidson 1974a). I believe that the triangulation argument, if it is sound, goes a long way towards vindicating this conviction and, by vindicating it, goes a long way towards answering the skeptic without begging the question against him. It will always be possible, of course, for the skeptic simply to reject Davidson’s account of content and to invoke, say, a divine being as the cause of our thoughts. Ultimately nothing can be done against such a skeptic, except to insist that the Davidsonian story makes for a much more compelling explanation of the content of our thoughts, an explanation which is moreover in line with much else we believe about how the world functions. Now, if what determines the contents of our thoughts and utterances is what typically causes them, the question is, what are these typical causes? According to Davidson, triangulation is needed here to provide the answer. More precisely, what is needed for anyone to have a language and thoughts is a history of causal interactions with at least another person and their common environment, that is, a history of responses they make to each other and to their surroundings. These shared responses are needed, to begin with, to isolate distal causes rather than proximal ones. By herself, a person could not “choose” among all the possible causes of her responses—between, say, the stimulation of her nerve endings and the original big bang (Davidson 2001b: 4)—and, as a result, she could not mean or think anything. A second person helps to isolate distal causes because these are whatever is situated at the intersection of the two lines that one could draw between two interacting interlocutors and what in the world they are responding to. But this is not the end of the story. The isolated distal causes are still ambiguous. The question «how much of the total cause

35

[of an utterance or thought] is relevant [to its content]» (Davidson 1997: 129) still needs to be answered. Which is why, I take it, Davidson goes on to say that the interaction must also matter to the creatures involved. «Unless the creatures concerned can be said to react to the interaction there is no way they can take cognitive advantage of the three-way relation which gives content to our idea that they are reacting to one thing rather than another» (1992: 120). Davidson’s thought here is not exactly transparent, but I believe it is this. No matter how regular two people’s reactions to each other and to their environment might be, they could always be reactions of one kind rather than another. We observing them could not tell which kind it is. It is for the people involved to determine which it is, that is, it is for them to determine what it is they are reacting to, which causes are the same as which, and thus, ultimately, what causes their utterances and thoughts and fixes their content. The causes of our utterances and thoughts do not impose themselves on us, as it were; they are the causes, and hence provide their content, because we take them to be the causes.3 Davidson then draws the anti-skeptical implication: If anything is systematically causing certain experiences (or verbal responses), that is what the thoughts and utterances are about. This rules out systematic error. If nothing is systematically causing the experiences, there is no content to be mistaken about. To quote myself: ‘what stands in the way of global skepticism of the senses is, in my view, the fact that we must, in the plainest and methodologically most basic cases, take the objects of a belief to be the causes of that belief’ (Davidson 1990: 201).

Thus, «belief can be seen to be veridical by considering what determines the existence and contents of a belief» (Davidson 1983: 146). As for the skeptic’s question “How do I know my beliefs are generally true?”, «it answers itself», Davidson says, «simply because beliefs are by nature generally true» (1983: 153).

3

Incidentally, this is partly why I think that, for Davidson, the concept of objectivity is also needed for language possession, i.e., content determination. See Verheggen (2007).

36

3. Stroud, however, does not think that these anti-skeptical claims are warranted. He does not think, to begin with, that Davidson is in a position to tell the skeptic that his beliefs about the world around him are by and large true. As we shall see presently, according to Stroud, all that Davidson has established is a claim about belief-attribution, viz., that it is largely truthascribing. What follows from this, Stroud argues, is simply that the skeptic cannot ask whether his beliefs about the world around him are by and large true, for he cannot coherently entertain the thought that all or most of those beliefs could be false since from the conditions of interpretation it follows that he can think of himself as having the beliefs he has only while thinking that most of these beliefs are true. Not only, Stroud further argues, does the claim that belief is in its nature veridical not follow from the claim that belief-attribution is in its nature largely truth-ascribing, making the stronger claim would in fact involve adopting a standpoint on our knowledge of the world that Davidson has shown does not exist. Moreover, Stroud continues, the stronger claim would not imply that anyone knows or has any good reason to believe anything about the world. Even if it were the case that most of our beliefs about the world are true, we might still lack any good reason to believe this. And so the stronger claim would not refute the skeptic in so far as he precisely denies that we have any knowledge of the world (Stroud 1999: 154). As I said earlier, Davidson takes his anti-skeptical claim already to follow from considerations about the nature of interpretation, considerations he advanced before developing the triangulation argument which I rehearsed above. Stroud is well aware of the triangulation argument but nonetheless focuses on radical interpretation and thus on the question what it takes for someone fully equipped with a language and thoughts to understand from scratch another person’s utterances and attribute to her beliefs and other propositional attitudes. The answer, to put it briefly, is that, since the interpreter could not get interpretation started unless she assumed that the speaker has by and large the beliefs she, the interpreter, would have in the circumstances, the interpreter is forced to attribute to the speaker beliefs that she herself takes to be true. Thus one could not first attribute beliefs and then ask whether these beliefs are by and large true. As Stroud writes, if Davidson «is right about the conditions of interpretation, no interpreter will be in a position to find all or most of the beliefs he ascribes to

37

people false […] one’s finding most or all of those attributed beliefs to be false is inconsistent with one’s having found the people in question to have those very beliefs» (1999: 146). But this means that «our having [the beliefs we now have] and their being all or mostly false is not a possibility we could consistently believe to be actual, so it is not a possibility we could be pressed to explain how we know is not actual» (1999: 157). Thus, according to Stroud, the weaker claim, according to which beliefattribution is in its nature largely truth-ascribing, is sufficient to keep the skeptic quiet, but not to negate his contention that we lack knowledge of the world. This would require, to begin with, that the stronger claim, according to which belief is in its nature veridical, be established.4 I think we must agree with Stroud that the stronger claim does not immediately follow from the conditions of interpretation. But Davidson does not mean to provide only conditions of interpretation but also conditions of the possession of language and thoughts, something perhaps made clearer by the triangulation argument and something Stroud is at least to some extent aware of. After all, he does write that «on Davidson’s view, thoughts with propositional content are present only if there is communication or reciprocal interpretation» (1999: 153). (As we shall see, though, Stroud’s talk of “reciprocal interpretation” may somewhat have misled him.) Now, does focusing on triangulation as an account of the possession of language and thoughts make any difference to the anti-skeptical claim that can be drawn? Does triangulation add anything to the conditions of interpretation? Stroud himself asks this question and considers specifically the role of causality in triangulation. But he concludes that this cannot strengthen the anti-skeptical claim, for, as I emphasized above, ultimately the causes of triangulating interlocutors’ utterances are what they take these causes to be. And Stroud takes this to imply that these causes could be other than what they take them to be. Only, he writes, if there were «no possible difference between what in fact cause verbal responses and what an inter4

Note that Stroud does consider, and then rejects, the omniscient interpreter argument, to the effect that even an omniscient interpreter would have to find his interpretees by and large in agreement with himself. But in this case what they agree about would by hypothesis also have to be true (Davidson 1977: 201). Since Davidson himself later disavowed the omniscient interpreter (in Davidson 1999b: 192), I shall not discuss it here.

38

preter takes their causes to be» (Stroud 1999: 151) could the stronger claim be supported. But now, on the face of it, Stroud seems to be expressing a flat contradiction here. If the causes of interlocutors’ utterances are what they take these causes to be, how could the causes of their utterances be different? What could Stroud have in mind here? In what follows, I shall argue that his (at least apparently) contradictory claim belies a failure to recognize that triangulation is supposed to provide conditions of the possession of language and thoughts. Stroud quotes Davidson as saying, in support of the stronger claim, that «we do not first form concepts and then discover what they apply to; rather, in the basic cases the application determines the content of the concept». But he chooses to focus on what Davidson says next: «An interpreter who starts from scratch […] cannot independently discover what an agent’s beliefs are about, and then ask whether they are true». However, Stroud then argues, the truth of most of an agent’s beliefs is not needed to explain that. «That is explained by the fact that interpreters attribute beliefs only by correlating the utterances they interpret with conditions which they believe to hold in the world they share with their speakers; they cannot discover what those speakers’ beliefs are without also believing certain things to be the causes of those utterances». But it does not follow from this «that interpreters must take the utterances to mean whatever in fact causes them» (Stroud 1999: 151-2). I suspect that Stroud sees no significant difference between radical interpretation and triangulation because Davidson’s account of triangulation, as I have indicated, is ultimately an account of language and thought within language and thought. That is, it does not provide an explanation of how specific utterances or thoughts get connected with specific causes; indeed, as long as these connections are not recognized by interlocutors as connections of given kinds, they cannot be content-determining. Thus we do not get a reductive account of how they come to have those thoughts. We might say that Davidson’s triangulating interlocutors are just as full of thoughts as are the radical interpreters. But saying this is slightly misleading, and I believe it has misled Stroud. True, we cannot think of triangulating interlocutors as speakers and thinkers unless their utterances and thoughts have determinate causes. But this should not blind us to the fact that triangulation is also the process through which they get to determine

39

what the causes of their utterances and thoughts are. If we are so blinded, we may again, like Stroud, miss Davidson’s claim that triangulation provides conditions of the possession of language and thoughts. 4. Is Stroud right, then, in maintaining that considering triangulation in addition to the conditions of interpretation makes no difference to how we treat the skeptical problem? Bjorn Ramberg, for one, claims that significant progress is made when we move from the considerations of radical interpretation to those of triangulation (Ramberg 2001: 228). This is because he takes radical interpretation to be an account of thought-attribution exclusively, where the question what constitutes the content of thoughts does not arise, and thus it makes sense to think of all the attributed thoughts as possibly mistaken. If, however, triangulation is constitutive of the thoughts of interlocutors, «we must conclude that we can no longer make sense of the idea of a radically mistaken radical interpreter» (Ramberg 2001: 231), the very claim Stroud seems to deny. For reasons I shall not spell out here, I believe that interpreting Davidson as changing his mind when he moves from radical interpretation to triangulation is mistaken.5 And, as I have suggested several times, this is not a mistake Stroud makes. Still, it does seem to me that, once we acknowledge that Davidson aims to provide not just conditions of interpretation but also conditions of the possession of language and thoughts, we are in a stronger position against the skeptic. For suppose Stroud is right in thinking that we could be wrong about the causes of our beliefs. What becomes clear is that this possibility does not entail simply the possibility that our beliefs are false, but it entails the possibility that our beliefs have a content different from what 5

See, for one thing, my quotation (here § 2—Davidson 1990: 200) about thirty years of endorsing the idea that the contents of our basic sentences are determined by what in the world causes us to hold them true—from the context, it is obvious that it is the “determined” of content constitution and not simply of content attribution that Davidson has in mind. For another, Davidson has long maintained that only someone who is not just interpretable but who has actually been interpreted by another could have a language and thoughts, something even Ramberg (2001) notices in Davidson (1975) but chooses to ignore.

40

we think they have. Thus the thought the skeptic cannot consistently entertain is not just, “I have all these beliefs and these beliefs are all false”, but “I have all these beliefs and I am wrong about their content, which is to say, in effect, I do not have all these beliefs”. This suggests two ways of dealing with the skeptic that are different from Stroud’s. Either we tell the skeptic that he must produce an alternative account of what does provide our thoughts with content, if it is not what we take to be their causes.6 Or we push skepticism one step further and become skeptical, as Nagel has put it, «about whether we are really capable of significant thought, while at the same time admitting that [such skepticism] is inexpressible and strictly unthinkable, since it is equivalent to saying, ‘Perhaps the sentence I am uttering right now means nothing at all’» (1999: 205). A further way to see that the considerations of triangulation cut more deeply against the skeptic than Stroud maintains is suggested by an observation of Ernest Sosa’s. Sosa points out that from the fact that “Not p but I believe p” cannot be coherently believed, it does not follow that “It is possible that both: not p and I believe p” cannot be coherently believed. (2003: 171).7 Now, when we focus on the content of beliefs rather than on their truth-value, what cannot be coherently believed is “I believe p but I do not believe p”. But in this case neither can “It is possible that both: I believe p but I do not believe p” be coherently believed. And this, it seems to me, shows again that the skeptic is more deeply challenged by the considerations regarding the possession of language and thoughts. After all, we, philosophers, discussing philosophical skepticism may observe, as Stroud repeatedly does, when reflecting on the conditions of interpretation, that there may be a gap between what we take to be the causes of our utterances and what these causes in fact are. This may not be something the skeptic could consistently assert, but it is something that could be true. (Hence, in effect, Sosa’s disappointment with Stroud’s conclusion.) But triangulation 6

This is the answer proposed by Ramberg (2001: 234). Sosa goes on to conclude that Stroud’s reading is not much of an answer to the skeptic. For «even if, only for the sake of the argument, it is granted that I cannot coherently think or say the following: ‘although I believe that p, it is, compatibly with that, possible that not p’, nevertheless it does not follow from that, nor is it true, that I cannot coherently think or say this: ‘I might be massively wrong in my beliefs’» (2003: 171). In other words, for Sosa, the skeptical hypothesis remains alive and well.

7

41

offers the Cartesian skeptic something that, as mentioned at the outset, he has simply neglected to give, viz., an answer to the question what makes it possible for us to have the thoughts we have. Now, when discussing philosophical skepticism, what may the skeptic observe at this stage? That he may not have the beliefs he thinks he has? This does not seem to be an acceptable starting point for Descartes (or anyone, for that matter!). And this indicates that what the skeptic owes us is indeed an alternative account of the content of our thoughts. As Davidson reminds us, following Wittgenstein, «it is only after belief has a content that it can be doubted» (Davidson 1999b: 165). Stroud is in fact still conceding too much to the Cartesian skeptic by taking content for granted, an assumption which, ultimately, is revealed by his willingness to conceive of a possible gap between what triangulating interlocutors take the causes of their utterances and thoughts to be and what these causes actually are. It is time to reconsider this possible gap. When I initially introduced this contention of Stroud’s, I noticed that it was at the very least puzzling that Stroud could conceive of such a gap but I then proceeded to grant it to Stroud and ask how granting it would affect how we reflect, not just on the truth-value of our beliefs, but on their content. As we have seen, the answer has been twofold. Either the skeptic is left with no starting point; if he cannot be sure of what it is he believes, he a fortiori cannot discuss the possibility that most or all of his beliefs could be false. Or the skeptic insists that he has the beliefs he thinks he has but owes us an account of their content that is different from that provided by triangulation. Either way, it has now become plain that what the skeptic, and Stroud with him, is doing is simply denying Davidson’s externalist account of content. Allowing for there to be a possible gap between what interlocutors take the causes of their utterances to be and what these causes actually are is in effect denying Davidson’s externalism. Throughout, Stroud simply fails to ascertain what anti-skeptical implications would follow if Davidson’s account, as an account of belief-possession, and not just as an account of belief-attribution, was right. One last way to realize this is to recall Stroud’s choosing to focus on Davidson’s claim that «An interpreter who starts from scratch […] cannot independently discover what an agent’s beliefs are about, and ask whether they are true», and then arguing that the truth of an agent’s beliefs is not needed to explain that, hence the

42

possible gap. Let us now look at the other claim Stroud quotes but then chooses to ignore: «we do not first form concepts and then discover what they apply to; rather, in the basic cases the application determines the content of the concept» (1990: 196). If Davidson’s externalist account of content is right, the correctness of an agent’s applications, that is, the truth of her beliefs, is needed to explain that. If the application determines the content of the concept, there is no possible gap.8 5. What precise anti-skeptical claim is Davidson then in a position to draw? I would put it this way: belief, as we understand it, is in its nature veridical. What I mean will become clear in what follows. As Davidson himself has insisted, I believe that he is in a position to tell the skeptic that the hypothesis that gives rise to the skeptical challenge is untenable: it is not possible for all or most of our beliefs about the world around us to be false. If our beliefs are contentful, that is, if we do have the beliefs we have about the world around us, then they are by and large true beliefs. Contra Stroud, however, I do not think that this way to address the skeptic is too quick or unsound. Stroud claims that it violates «the abstract possibility of a set of beliefs’ being all or mostly false in the minimal sense that the truth of all or most or even any of them does not follow simply from their being held […] It would be to deny that, considered all together, the truth or falsity of the things we believe is independent of their being believed to be so» (1999: 155). But, as Davidson has argued, «the assumption that the truth of what we believe is logically independent of what we believe is […] ambiguous» (1991: 214). Thus, the truth of any one belief about the world does not follow from its being held. But the truth of a large number of them does so follow. Again, this is grounded on Davidson’s account of content. Beliefs could not by and large be false since what makes many of them true is what makes many of them contentful to begin with. And, as Davidson has further argued, this does not threaten the objectivity 8

Note that the unintelligibly of such a gap is of a piece with the unintelligibility of the distinction between scheme and content, an unintelligibility Davidson has famously argued for. In each case, what makes no sense is the idea that different worlds could contribute in the same way to our system of beliefs; it is the idea of the world standing completely apart from our beliefs (see Davidson 1974b).

43

of truth in the sense that what we believe to be true is so independently of our believing it. Davidson’s view does not commit him to the anti-realist claim according to which what the world is like depends on what we believe it is. It is true, though, that Davidson also does not think that there is no connection whatsoever between our beliefs and the world they are about and so between beliefs and truth. But the connection goes the other way around, so to speak. It is, rather, that what we believe about the world depends on the way the world is. (This is why he eventually preferred to call his view neither realist nor anti-realist— Davidson 1988: 185.) But Stroud has two other complaints about drawing the stronger claim, which I mentioned earlier. The first is that drawing it involves adopting a standpoint on our knowledge of the world that Davidson himself has shown does not exist. What Davidson has shown is that we cannot step outside our body of beliefs in order to assess them. As he writes, «There is no going outside this standard [provided by communication] to check whether we have things right, any more than we can check whether the platinum-iridium standard kept at the International Bureau of Weights and Standards in Sevres, France, weighs a kilogram» (Davidson 1991: 217-8). But the skeptic wants us to reach our conclusion without appealing to any of the beliefs in question. This complaint of Stroud’s is of course a variation on the idea that, if Davidson is right, the skeptical question cannot be raised. And, if it cannot be raised, it cannot be addressed. Davidson objects to this, however, noting that from the fact that it makes no sense to question the adequacy of a standard, it does not follow that it also makes no sense to assert its adequacy: «when the standard for the kilogram was a weight kept in Sevres, there was no point in checking whether that weight weighed a kilogram. It does not follow that that weight was not the standard» (Davidson 1999a: 164). I think Stroud’s complaint belies again a failure to acknowledge that the stronger claim is grounded on an account of the content of our beliefs, on our best understanding of what it takes to have the beliefs we have about the world around us. This precludes us from asking whether all or most could be false. But it does not preclude us from asserting that many of them must be true. The other complaint Stroud has about the stronger claim is that, if we could draw it, it would not refute skepticism because it would not imply that we know or have any good reason to believe anything about the world.

44

This, in effect, is yet another way of saying that we could not step outside all of our beliefs in order to assess them. If Davidson is right, there are many things we know about the world since how our knowledge-claims come to have the content they have both ensures the truth of many of them and explains how they are justified (at least to the extent that we have some good reason to believe that many of them are true). But, again, if Davidson is right, there is no way we can know that we have that knowledge (if this does involve stepping outside all of our beliefs). And this seems to be what the skeptic is seeking. Well, if he is, as Davidson puts it, he is «demanding more of knowledge than is justified» (1999a: 162). More accurately, I would add, he is demanding more of knowledge than is intelligible. If we take seriously the claim that triangulation provides conditions of the possession of language and thoughts and not just conditions of interpretation, we can no longer point out, from a philosophically skeptical point of view, that all or most of our beliefs could be false after all, or that we could be wrong about all or most of the causes of our beliefs. Rather, we have to remain silent. Not being able to assess our beliefs from outside means not being able to discuss them from outside either. The skeptic cannot have it both ways, which is, in effect, what Stroud tries to do, acknowledging on the one hand that the skeptic has been quieted, but then insisting that the skeptic has not been addressed properly. To end, let me reiterate that the aim of this paper has not been to assess Davidson’s account of content but to ask what precise anti-skeptical conclusion his account warrants. Now the skeptic might also enquire into the status of this account in order to assess the status of the anti-skeptical conclusion. In response, I shall quote Davidson one last time: There cannot be proof of [the sort of externalism I embrace]. Its plausibility depends on a conviction which can seem either empirical or a priori; a conviction that it is a fact about what sort of creatures we are. Empirical if you think it just happens to be true of us that this is how we come to be able to speak and think about the world; a priori if you think, as I tend to, that this is part of what we mean when we talk of thinking and speaking. After all, the notions of speaking and thinking are ours (2001c: 294).

45

Hence my revised version of Davidson’s anti-skeptical claim: belief, as we understand it, is by nature veridical. But what other way to understand it is there? References Davidson, D. 1967, “Truth and Meaning”, repr. in Davidson 1984. Davidson, D. 1973, “Radical Interpretation”, repr. in Davidson 1984. Davidson, D. 1974a, “Belief and the Basis of Meaning”, repr. in Davidson 1984. Davidson, D. 1974b, “On the very Idea of a Conceptual Scheme”, repr. in Davidson 1984. Davidson, D. 1975, “Thought and Talk”, repr. in Davidson 1984. Davidson, D. 1977, “The Method of Truth in Metaphysics”, repr. in Davidson 1984. Davidson, D. 1983, “A Coherence Theory of Truth and Knowledge”, repr. in Davidson 2001a. Davidson, D. 1984, Inquiries into Truth and Interpretation, Oxford: Clarendon Press. Davidson, D. 1988, “Epistemology and Truth”, repr. in Davidson 2001a. Davidson, D. 1990, “Epistemology Externalized”, repr. in Davidson 2001a. Davidson, D. 1991, “Three Varieties of Knowledge”, repr. in Davidson 2001a. Davidson, D. 1992, “The Second Person”, repr. in Davidson 2001a. Davidson, D. 1997, “The Emergence of Thought”, repr. in Davidson 2001a. Davidson, D. 1999a, “Reply to Barry Stroud”, in Hahn 1999. Davidson, D. 1999b, “Reply to A.C. Genova”, in Hahn 1999. Davidson, D. 2001a, Subjective, Intersubjective, Objective. Oxford: Clarendon Press. Davidson, D. 2001b, “Externalisms”, in Kotatko et al. 2001.

46

Davidson, D. 2001c, “Comments on Karlovy Vary Papers”, in Kotatko et al. 2001. Hahn, L.E. 1999, The Philosophy of Donald Davidson. Chicago and La Salle, Ill.: Open Court. Kotatko, P., Segal, G. and Pagin, P. 2001, Interpreting Davidson. Stanford, Cal.: CSLI. Nagel, T. 1999, “Davidson’s New Cogito”, in Hahn 1999. Ramberg, B. 2001, “What Davidson Said to the Skeptic Or: AntiRepresentationalism, Triangulation and the Naturalization of the Subjective”, in Kotatko et al. 2001. Sosa, E. 2003, “Knowledge of Self, Others, and World”, in K. Ludwig (ed.), Donald Davidson. Cambridge, Mass.: Cambridge University Press. Stroud, B. 1999, “Radical Interpretation and Philosophical Skepticism”, in Hahn 1999. Verheggen, C. 2007, “Triangulating with Davidson”, The Philosophical Quarterly, 57 (226), 96-103.

TRIANGULATION BETWEEN EXTERNALISM AND INTERNALISM M. Cristina AMORETTI (University of Genoa, Italy)*

The disputes between externalism and internalism in philosophy of mind, on the one hand, and epistemology, on the other, do not seem at least prima facie to be deeply intertwined.1 I believe that this is not really the case, however, and that the theory of triangulation—developed by Donald Davidson in the last two decades of his life—is a clear example of how considerations about the nature of mental content, justification, and knowledge may be interdependent. In this paper I shall analyze Davidson’s “triangular” externalism about mental content in relation both to his coherence theory of justification (which may be regarded as a kind of epistemic internalism), and to his attempt to offer an answer to the skeptic. First of all, I shall consider the thesis that content externalism cannot be compatible with epistemic internalism, and argue against it that there is no tension between triangular externalism and Davidson’s internalism of justification. Secondly, I shall not only demonstrate that triangular externalism supports Davidson’s thesis that belief is in its nature veridical, but also claim that such a result is probably not enough to offer a satisfactory argument against the skeptic about the external world. 1. On Davidson’s Coherence Theory of Justification It is well known that Davidson strongly denies foundationalism and defends a coherence theory of epistemic justification according to which *

I gratefully acknowledge financial support from Grant POSDRU/89/1.5/S/63663. I also thank Nicla Vassallo for reading a previous version of this paper and providing thoughtful comments. 1 For an exhaustive survey on the relations between the two disputes see Goldberg 2007.

48

«empirical knowledge has no epistemological foundation, and needs none» (Davidson 2001b: xiv). More precisely, he points out that «what distinguishes a coherence theory is simply the claim that nothing can count as a reason for holding a belief except another belief. Its partisan rejects as unintelligible the request for a ground or source of justification of another ilk», (Davidson 1983: 141). In particular, he thinks that perceptions play no epistemic role (1983: 141-146). Since perceptions obviously cause some of our beliefs (namely empirical beliefs)2 in a certain sense they may be viewed as the “basis” or “ground” of those same beliefs. We should not, however, confuse a mere causal dependence with an epistemic one: the point is that perceptions cannot be considered as the justifying ground of any belief at all. Only other beliefs of the subject may justify one of her beliefs, including the empirical ones.3 According to a coherence theory of justification, generally speaking, a belief that p is justified when it coheres with a vast array of other beliefs.4 As Davidson puts it, «there is a presumption in favor of the truth of a belief that coheres with a significant mass of beliefs. Every belief in a coherent total set of beliefs is justified in the light of this presumption» (1983: 138139). He also adds however that such a claim is not enough to guarantee genuine justification. In fact, if the belief that p coheres with a significant group of false beliefs, we could not say that it is a justified belief. Hence, a more plausible coherence theory of justification should specify that the belief that p is justified when it coheres with a vast array of true (or at least 2

More precisely, empirical beliefs are those contingent beliefs directly caused by sensory or perceptual experience. As we shall see, they play a pivotal role in the definition of Davidson’s content externalism, because they “anchor” thought to the world (Davidson 1988: 43-45). 3 Davidson argues against the existence of “epistemic intermediaries” in several papers (see for instance his 1982a, 1983, 1988, 1991); unfortunately, in the present paper it is not possible to discuss this issue further. 4 According to Davidson «coherence is nothing but consistency» (1990a: 155). Such a characterization is noticeably problematic for a coherence theory of justification, but I will not discuss this difficulty here. Actually, Davidson’s emphasis on coherence is just a way of denying foundationalism and stressing the idea that «all that counts as evidence or justification for a belief must come from the same totality of belief to which it belongs» (1983: 153).

49

largely true) beliefs. Since coherence alone cannot guarantee that the overall set of a subject’s beliefs is true or even largely true, however, it seems that coherence alone cannot provide any justification for a belief. The above considerations may easily yield to a certain kind of philosophical skepticism. If we wish to defend—as Davidson actually does—a weak form of realism (thinking that the external reality is logically independent from what a person believes about it) and the idea that beliefs are objective (namely, that they are true or false independent of the existence of the belief or the believer—see for instance Davidson 1999a: 129), then it would always make sense to wonder whether a coherent given set of beliefs is true. It seems that the possibility that the whole set may be comprehensively false about the actual world—even though it is coherent and held to be true by the subject—is still open to question.5 If these considerations are right, however, then without a reason to believe that a coherent set of beliefs cannot be totally false coherence cannot even provide justification. In short, the outcome would seem to be that we must accept the idea that we do not have any genuine knowledge of the external world.6 A possible way to avoid this skeptical conclusion without withdrawing a coherence theory of justification would be to insure that a coherent set of beliefs must be true or at least largely true, despite the fact that each particular belief, being objective, could be singularly false (even when it coheres with other beliefs). The problem, as Davidson puts it, is that «the partisan of a coherence theory can’t allow assurance to come from outside the system of beliefs, while nothing inside can produce support except as it can be shown to rest, finally or at once, on something independently trustworthy» (1983: 140). According to him, then, «what is needed to answer the skeptic is to show that someone with a (more or less) coherent set of beliefs has a reason to suppose his beliefs are not mistaken in the main», a reason «that is not a form of evidence» (1983: 145). More precisely, such a reason may be found by reflecting on the very nature of our beliefs about 5

Prima facie, this conclusion seems to work also in a Davidsonian framework, where there are no epistemic barriers between knowledge of the external world, our own mental states, and the mental states of others. 6 If coherence is not able to provide justification, we may in fact have a substantial number of coherent true beliefs that, nevertheless, do not amount to knowledge.

50

the external world and on the way their contents are determined. It is precisely content externalism that should give us a reason to believe that our cohering beliefs must be largely true. 2. On Davidson’s Triangular Externalism about Mental Content To put it generally, in philosophy of mind externalism about mental content is the thesis according to which the content of (some) propositional attitudes constitutively depends, at least in part, on objects and events in the outside world. Unfortunately, it is not possible to describe and discuss Davidson’s externalism exhaustively —“triangular” externalism, as I shall call it from now on—but for the aim of this paper, sketching some of its pivotal features will still be sufficient.7 First of all, it should be noted that triangular externalism is quite different from “orthodox” externalisms, such as those defended by Hilary Putnam (1975) or Tyler Burge (1979, 1986). It is certainly a form of causal (or perceptual) externalism to the extent that the contents of (some) propositional attitudes constitutively depend, at least in part, on objects and events in the outside world, or rather on what has “typically” caused them, that is, on the causal history of the subject’s relations with those objects and events.8 Triangular externalism, however, is not merely causal (or perceptual), but can be considered as an attempt to combine both causal (or perceptual) and social externalism. According to Davidson, in fact, the mere history of causal relations between the subject and the outside world is not at all sufficient to determine the content of her propositional attitudes. A third in7

For further discussion on triangular externalism, see for instance Amoretti 2007, 2008a, 2008b; Amoretti and Vassallo 2008; Fennell 2000; Føllesdal 1999; Glüer 2001, 2006; Lepore and Ludwig 2005; Pagin 2001; Ramberg 2001; Verheggen 1997, 2007. 8 In this sense, triangular externalism may be better seen as a form of diachronic, rather than synchronic, externalism: «There must be a distinction between the role the actual presence of water plays when it causes me to think this is water, and the role the history of my relations to water play in making a false thought that I am seeing water nevertheless a thought about water. Clearly it is the latter thesis that is essential to externalism» (Davidson 1990b: 199-200). See also Amoretti 2008b; Lepore and Ludwig 2005.

51

gredient, the “second person” (as Davidson labels it), is also necessary, and represents the social element that he wants to introduce in his own kind of externalism. More precisely, mental content also depends on the fact that there are other creatures (at least one) whose perceptual apparatus, concept-forming abilities and other biological dispositions are innately sufficiently similar to those of the subject to enable “triangulation” on the background of a shared public world. It is important to stress that within Davidson’s externalism—unlike Burge’s—the term “social” peculiarly refers to an intersubjective space and not to a community of speakers who are bound by a common language and conventional shared norms, practices or rules (see for instance Davidson 1986, 1994). In any case, the role of the second creature is precisely the social element that Davidson is looking for and it constitutes one of the distinctive marks of triangular externalism: «I would introduce the social factor in a way that connects it directly with perceptual externalism, thus locating the role of society within the causal nexus that includes the interplay between persons and the rest of nature» (Davidson 1990b: 200). Moreover, according to Davidson externalism is “global”, that is, it involves (directly or indirectly) the whole thought and not only some special beliefs concerning natural kinds or indexicals. Since he actually believes that a subject’s mental contents are holistically interconnected, it is possible to affirm that all of our mental states depend, at least indirectly, on external objects and events. Given holism, triangular externalism should also be considered as just a weak kind of externalism. In fact, even though mental content depends on the history of causal interactions between the subject and the outside world on the background of an intersubjective framework, this dependence is not its only component, given that holistic factors are equally crucial. In other words, the content of a belief is also determined by its inferential relations with a vast and coherent array of other propositional attitudes. As a result, we may say that there are two basic elements in Davidson’s notion of mental content: a referential element (introduced by the subject’s causal history) and an inferential one (introduced by some holistic constraints). The former connects content to the external world, whereas the latter connects a single belief to other beliefs entertained by the subject (Amoretti 2007).

52

3. Epistemic Internalism vs. Content Externalism: Some Insights Before examining whether the combination of Davidson’s coherence theory of justification and triangular externalism represents a good strategy to «tell the skeptic to get lost» (Davidson 1990a: 157), we should first make sure that these theories fit well together.9 One primary step, then, would be to determine whether Davidson’s coherentism may be regarded as a kind of epistemic internalism or as a kind of epistemic externalism.10 Although there are many different varieties of epistemic internalism about justification, one of its most prominent tenets is the idea that the justification for a belief that p is completely determined by some elements “internal” to the subject’s mind.11 According to this general idea, Davidson’s coherence theory—whose central rationale is that justification consists in having a reason in the form of another belief12—may well be considered as a kind of internalism about justification. Let us examine it.

9

When talking about epistemic internalism/externalism I will only refer to justification. A different matter is whether Davidson may be considered as an epistemic internalist/externalist about knowledge. On the one hand, since he believes that holding a true belief is not enough for knowledge, he cannot be considered as an externalist about knowledge. On the other, because he regards truth as a necessary condition for knowledge, and also endorses a moderate kind of realism (according to which truth is objective and independent of both our actual beliefs and justification), then it is easy to conclude that he cannot accept a strong internalism about knowledge. If, however, we understand epistemic internalism about knowledge in a weaker sense, as the thesis that knowledge is at least in part determined by some elements “internal” to the subject’s mind, then Davidson’s coherentism may well be considered as a form of epistemic internalism about knowledge. 10 Of course coherentism and foundationalism are traditionally regarded as paradigmatic internalist theories; nevertheless it is still possible, at least in principle, to conceive them as externalist. 11 Another way of stating this idea is through the notion of supervenience: whether a subject is justified in believing that p supervenes on some elements internal to the subject’s mind. 12 A subject is justified in believing that p when she has good reasons for believing that p, and—according to Davidson—the reasons for believing that p may be only some other beliefs cohering with p. That is to say, one’s own beliefs completely determine the justificatory status of one’s belief that p.

53

If we understand the dichotomy between epistemic internalism and epistemic externalism as a dichotomy about the nature of what can actually justify a subject’s belief that p and we identify the “internal” with the “mental”, then Davidson’s coherence theory of justification may be easily identified as a form of epistemic internalism. Let us call “mentalism” the thesis that justification is completely determined by one’s own mental states or, to put it another way, that one’s own mental states completely determine the justificatory status of one’s particular belief that p. Since Davidson thinks that justification must be completely determined by one’s own beliefs, and beliefs are obviously a special kind of mental state, then he should also accept epistemic internalism as mentalism. If the basic tenet of epistemic internalism is merely that justification is completely determined by one’s own mental states, then it is clearly compatible with triangular externalism (and with content externalism in general), because beliefs are definitely still mental states, however their content is determined. As Earl Conee points out, if internalism «is the thesis that for epistemic purposes the ‘internal’ is the mental», then «since content externalism expands the factors that fix the mental, content externalism expands the supervenience base for justification according to mentalism» (2007: 51). The most usual way of understanding the dichotomy between internalism and externalism in epistemology, however, is to characterize it in terms of “accessibility” or “special access”, taking these two notions to refer to what is accessible by reflection alone or a priori, without inference from observation of one’s behavior, speech or environment (Audi 2010; Bernecker and Dretske 2000). Let us call “accessibilism” the thesis that justification is completely determined by some elements to which the subject has special cognitive access or, to put it another way, that some reflexively accessible elements completely determine the justificatory status of one’s particular belief that p. Generally speaking, a coherence theory of justification may be easily construed as a form of accessibilism. In fact, a coherentist maintains that justification is a matter of inferential relations among beliefs; and beliefs are commonly considered as something to which the subject (at least potentially) has special cognitive access, namely something which the subject may allegedly know or be aware of via mere reflection, without recourse to any evidence or observation.

54

Davidson has never explicitly described his coherence theory of justification in terms of accessibility or special access, nor overtly talked about a priori in relation to one’s access to one’s own beliefs. Nevertheless, it should be recognized that he still believes there is a manifest asymmetry between self-knowledge and other kinds of knowledge (i.e. knowledge of the external world and other minds), as first person access to one’s own mental states is not a kind of access one may have to the external world and to the mind of others. If we understand the notion of special access to be the notion of what is accessible “from the armchair”, without inference from observation of one’s behavior, speech or environment, then we may say that Davidson thinks one potentially has special cognitive access to one’s own beliefs. In his own words: «It is seldom the case that I need or appeal to evidence or observation in order to find out what I believe; normally I know what I think before I speak or act» (1987: 15), and again «People generally know without recourse to inference from evidence, and so in a way that others do not, what they themselves think, want, and intend» (1988: 48). Against the thesis that Davidson’s coherence theory of justification may be construed as a form of accessibilism it may be objected that beliefs are something that the subject may not actually be aware of. This is certainly true, but it should also be recognized that in normal cases the subject could become aware of her own beliefs by reflection alone, by simply turning her attention to them. Some exceptions are unconscious or repressed beliefs, to which the subject does not have special cognitive access, not being able to know or become aware of them by reflection alone. In order to establish whether or not Davidson’s coherentism may be seen as a kind of accessibilism, then, we should determine if some non-reflexively accessible beliefs (such as unconscious or repressed beliefs) could be one’s reasons for believing that p. Since a reason should be something that a subject is able to exhibit, however, I think that Davidson would not accept this possibility. A non-reflexively accessible belief (such as an unconscious or repressed belief) may be the cause for believing that p, but it does not rep-

55

resent a reason for believing it.13 If it is true, then the reasons for believing that p may be only some (actual or dispositional) reflexively accessible beliefs that cohere with p. Then justification is completely determined by reflexively accessible beliefs or, more precisely, in logical relations among those reflexively accessible beliefs, and whether or not one’s belief that p coheres with some other beliefs is also reflexively accessible by the subject. Since accessibilism holds that justification is completely determined by reflexively accessible elements and both non-repressed beliefs as well as logical relations among them are reflexively accessible by the subject, Davidson’s coherentism may be seen as a form of accessibilism. The alleged incompatibility between accessibility internalism and triangular externalism depends on the widespread opinion that content externalism undermines a subject’s ability to know the content of her own beliefs by reflection alone, without inference from observation of her own behavior, speech or environment.14 The objection may be summarized as follows: according to externalism, content is determined, at least in part, by external factors which, being external, may be inaccessible by the subject; if this is so, the subject may also ignore the content of her own beliefs and be mistaken about what she herself believes. This can be shown by means of a slow switch example. Let us suppose we live on Earth and believe that water is wet. If we were switched to Twin Earth, then we would come to believe that twater is wet (since on Twin Earth there is no water, but twater), without being able to know that by reflection alone. In fact, we would not be able to distinguish between water-beliefs and twater-beliefs by reflection alone, without making some empirical investigations. This implies that given externalism one lacks special access to (some) mental contents and thus to (some of) the factors which determine the justificatory status of (some of) one’s beliefs. Accordingly, accessibilism is false. A general reply, also endorsed by Davidson, is that what determines the content of a first-order thought—whatever it is—also determines the 13

See, for instance, the following passage: «Mental phenomena may cause other mental phenomena without being reasons for them, then, and still keep their character as mental, provided cause and effect are adequately segregated» (Davidson 1982b: 181). 14 For a rich survey on this issue see for instance Brown 2004; Ludlow and Martin 1998; Nuccetelli 2003; Wright et al. 1998.

56

content of the corresponding second-order thought, leaving no room for error. Moreover, it is worth recalling that according to triangular externalism the subject’s causal history is what actually determines content (together with holistic constraints). Hence, if I had always lived on Earth, I would have never interacted with water, I would not have any twater-belief (not even after the slow switch), and so there would be nothing to compare my water-beliefs with: «What I see before me I believe to be water; I am in no danger of thinking it is twater, since I do not know what twater is. If I am on Earth, I also believe I think I am seeing water, and in this I am right. If I were without my knowledge transferred to [Twin Earth] I would believe twater was water—a mistake. In both cases I would know what I believed» (Davidson 1989: 61).15 Finally, there is a stronger kind of epistemic internalism about justification that needs to be examined—especially because it is the one that, according to many scholars, best exemplifies our deep intuitions about internalism. The idea is that to be justified in believing that p, a subject must have special access to her own justificatory status. Again, the notion of special access should be taken to refer to what is accessible by reflection alone or a priori, without inference from observation of one’s behavior, speech or environment.16 Following what I said in § 1, it seems that Davidson endorses this kind of epistemic internalism as well. In fact, he maintains that coherence alone cannot provide justification for the belief that p (it is not enough that a subject happens to have many true beliefs cohering with p). In order to be justified in believing that p, the subject must be aware by reflection alone not only that p coheres with many other coherent beliefs, but also that cohering beliefs cannot be totally false. I will extensively discuss this kind of epistemic internalism in the next sections,17 but 15

On this point see also Amoretti 2008a. Of course, the notion of special access should not be considered as referring to direct or infallible knowledge, since coming to know one’s own justificatory status certainly requires reasoning. 17 One may object that this kind of epistemic internalism is not compatible with content externalism, as long as content externalism implies that a subject with two occurring beliefs p1 and p2 may not be able to establish by reflection alone whether p1 and p2 have the same (or a different) content or not. I have extensively argued elsewhere that triangular externalism does not have this implication (see Amoretti 2007). 16

57

from now on let us label it as epistemic “internalism” tout court and its denial as epistemic “externalism”. 4. Davidson’s Anti-Skeptical Argument: Does It Really Work? According to Davidson, in order to convince the skeptic that a (more or less) coherent set of beliefs must be largely true, we should show her that someone with such a set of beliefs has a reason to consider that their beliefs are not totally false. Without such a reason, one would not be justified in believing that p, even if the belief that p is actually coherent with a large set of other beliefs. Since this reason, for a coherentist, cannot be a form of evidence, we need an epistemological proof, namely a deductively valid argument whose premises are true and known by us.18 Let us suppose that one of our premises, P, is true. If we consider the traditional definition of knowledge as a justified true belief, in order to know that P, we also need to believe that P and to have a justification to believe that P. As we have just seen, the justification cannot be a piece of empirical evidence, since in Davidson’s view there is no non-doxastic form of evidence (our sensations do not justify but simply cause our beliefs).19 Thus, the justification for the premises of our deductively valid argument must consist in other beliefs, which would simply enlarge the set of our cohering beliefs. It seems, however, that in order to satisfy the skeptic our justification cannot consist in empirical beliefs directly caused by perception and thus, in a sense, “based” on perception. Our justification should be something like an a priori reasoning on the nature of thought and language. According to Davidson, in fact, a subject has only to reflect (i.e. to reason from the armchair) on what a belief is and how it is acquired and 18

Of course, we do not need to prove each single premise, because we would be committed to an infinite regress: let us suppose that we find a proof for a premise; then we should find a proof for the proof we have just found and so on ad infinitum. If the skeptic is going to insist on that, skepticism is obviously inevitable, but we must admit that her request cannot be accepted, since human beings are finite beings (Vassallo 2003). 19 Appealing to empirical evidence to argue against a radical skeptic would simply beg the question.

58

determined to appreciate that most of one’s basic beliefs about the external world are actually true. As he explicitly puts it, «we can dismiss a priori the chance of massive error» (1975: 169). Davidson’s anti-skeptical argument (A) can be summarized as follows: (P1) If triangular externalism is true, than our cohering beliefs cannot be totally false; (P2) Triangular externalism is true; (C) Then, our cohering beliefs cannot be totally false. The above reasoning is clearly a deductively valid argument of the form “If p then q”, “p”, then “q”, that is a Modus Ponens. But to have an epistemological proof against the skeptic, the subject must also know its two premises: it is at the least necessary that the subject believes them, they are true, and the subject has a justification for believing them. Let us analyze (P1). According to Davidson’s triangular externalism, what really matters for the content determination of our basic beliefs about the external world is the history of the actual interactions between the subject and the outside world in the space of triangulation. Hence, if triangular externalism is true, then in the most basic cases our propositional attitudes derive their content from the external causes which have typically caused them, namely from the subject’s causal history (the natural history of what is in her head). That is to say that in the most basic cases the content of our beliefs depends on those objects and events that have typically caused those same beliefs. Moreover, we have seen how beliefs are connected in a holistic framework from the very beginning: acquiring beliefs means acquiring a cohering set of beliefs. This implies however that our cohering sets of beliefs are by their nature veridical (they are about what has actually caused them) and that we cannot be systematically deceived as in the skeptical hypothesis. Therefore, if we have any belief at all, then triangular externalism guarantees that our cohering beliefs cannot be totally false, i.e. that the external world is more or less as we think it is. (P1) seems to be not only true but also justifiable by mere reflection, because it is directly implied by triangular externalism that our cohering beliefs cannot be totally false. As Davidson puts it, «If you accept the steps that lead to my version of externalism, then you cannot, I think, be a skep-

59

tic about the existence of an external world much like the one we all believe we share, nor about the existence of other people with minds like ours» (1999b: 194). If we reflect on triangular externalism and on the very nature of beliefs, from the simple fact that we have thoughts and other propositional attitudes we can infer by mere reflection that there is an external world, which is more or less as we think it is, and there are also other rational creatures sufficiently similar to us. Given externalism, it is a condition of thought and language that we are not brains in a vat.20 We can say, then, that (P1) is justified by an a priori reasoning about the very theory of triangulation. Let us see now if we also know (P2), namely that triangular externalism is true. We have seen that, in order to have an epistemological proof, the premises of Davidson’s argument must be justified by mere reflection, without the support of any empirical belief. Let us suppose, for the sake of argument, that triangular externalism is true and that the subject believes it is. What is her justification for believing that triangular externalism is true? The problem is whether triangular externalism can be established by mere reflection or not. In other words, we need to determine whether Davidson’s externalism may be derived exclusively by philosophical arguments (i.e. from the armchair), without appealing to any evidence. First of all, it should be noted that Davidson rejects any thought experiment and wishes to derive triangular externalism exclusively from what happens to be the actual practice of communication between two or more speakers: «I have a general distrust of thought experiment that pretends to reveal what we would say under conditions that in fact never arise. My version of externalism depends on what I think to be our actual practice» (Davidson 1990b: 197), and again «the considerations in favor of […] [triangular externalism] seem to depend in part not on purely a priori considerations but rather on a view of the way people are» (Davidson 1999b: 194). Since, on this account, triangular externalism seems to de20

In fact, the general picture of mind presented by Davidson connects subjective knowledge (knowledge of our own mental states), objective knowledge (of the external world), and intersubjective knowledge (of others’ mental states): none of these three kinds of knowledge can actually stand alone, nor be reduced to the others (Davidson 1991).

60

pend on empirical considerations, a skeptic would probably think that the argument simply begs the question and thus is flawed. However, there is also a sense in which triangular externalism may depend exclusively on philosophical arguments, that is, a priori and not empirically. The view that a history of encounters with some of the things we speak about and have beliefs about is necessary if we are to refer to and form attitudes towards those objects is at the heart of the sort of externalism I embrace. On this view, aspects of our interactions with others and the world are partially constitutive of that we mean and think. There can not be said to be proof of this claim. Its plausibility depends on a conviction which can seem either empirical or a priori; a conviction that this is a fact about what sort of creatures we are. Empirical if you think it just happens to be true of us that this is how we come to be able to speak and think about the world; a priori if you think, as I tend to, that this is part of what we mean when we talk of thinking and speaking. After all, the notions of speaking and thinking are ours (Davidson 2001a: 293-294).

If Davidson’s idea is plausible, then (P2) may be justifiable a priori by merely reasoning from the armchair. Hence, the skeptic cannot simply say that the argument (A) begs the question, but she should provide herself with a different account of the nature of thought and language, an account that would be more realistic and convincing than triangular externalism. To put it another way, the burden of the proof is now on the skeptic. Meanwhile, (A) can be considered a good epistemological proof to exhibit against her argument. I doubt that triangular externalism could be derived solely a priori, but even putting aside this difficulty there is still another problem clearly stated by Ernest Sosa (2003). Since the efficacy of our epistemological proof depends on the capacity to derive triangular externalism merely a priori, only a few individuals would be able to do that, and the majority of people would not have any reason to believe that their cohering beliefs must be largely true. Thus, they would also lack any reason to suppose that a belief cohering with a vast array of other beliefs is prima facie justified, there being a presumption in favor of its truth. The conclusion is that those people—not being capable of deriving triangular externalism a priori— would not be able to have any special access to their justificatory status,

61

and thus they could at best have a set of true beliefs but no genuine knowledge of the external world, as the skeptic actually maintains. Even if some people lack epistemological proof to dismiss the radical skeptic and have no reason to presume that a belief cohering with a vast array of other beliefs is prima facie justified, there may be another way to resist the conclusion that those people cannot have any genuine knowledge of the external world. In fact it has been argued that if triangular externalism actually implies that our beliefs cannot be generally wrong, then such a result can still be considered as a justification conceived as external or, better, it «is not a reason, inasmuch as it is a source of justification that epistemically favors even those who have no belief in any Davidsonian theory about how our beliefs and sayings acquire content. Nor need one have any reason at all for beliefs that are nonetheless justified» (Sosa 2003: 178). To put it another way, if triangular externalism is true, then there is a source of justification conceived as external, in the sense that a subject may be unable to gain special access to her own justificatory status, being nevertheless justified in holding a particular belief. Since triangular externalism—following (P1)—guarantees that one’s cohering beliefs must be largely true, then one may be justified in holding a belief that coheres with a vast array of other beliefs of hers, even if one is unable to tell whether she is justified or not, since this would require reflective access to Davidson’s argument (A). It seems that Davidson himself has changed his mind, arguing for just such an externalist account of justification: The right thing to say is rather this: we are justified in taking our perceptual beliefs to be true, even when they are not and so when they are true, they constitute knowledge (this is what I meant by saying our perceptual beliefs are veridical). But since our only reasons for holding them true are the support they get from further perceptual beliefs and general coherence with how we think things are, the underlying source of justification is not itself a reason (Davidson 1999c: 208).

Following this reasoning, if triangular externalism is true and has the antiskeptical consequences sketched above, then it epistemically favors even those people who are not able to understand it reflectively, namely those

62

people who do not have any reason to suppose their beliefs are not mistaken in the main and thus do not have any special access to their justificatory status. Hence, the skeptic would be wrong in claiming that we do not have any genuine knowledge of the world simply because we lack such a reason. Coherence may secure our knowledge that p just because triangular externalism independently grants that cohering beliefs cannot be completely wrong. Of course, this conclusion may be maintained only if one renounces to defend epistemic internalism in its stronger version, the one that best exemplifies our deep intuitions about epistemic internalism itself. 5. A Weaker Result? We have seen that according to triangular externalism the content of our beliefs is determined by the causal history of the subject, i.e. by those causal interactions that have typically tied up the subject with the external relevant cause of her beliefs within an intersubjective space. We have also pointed out that Davidson is right in holding that triangular externalism rules out radical skeptical scenarios, like Descartes’s deceiving God or the possibility of a permanent brain envatment, such as the one described by Putnam’s famous thought experiment (1981). Ruling out those skeptical hypotheses, therefore, we see that triangular externalism may provide an argument against the skeptic and thus a justification—at least conceived as external—to guarantee our knowledge of the outside world. There are still other skeptical scenarios, however, which seem to be perfectly coherent with triangular externalism and Davidson’s account of thought and language. Let us imagine the following situation. Emily was born and lived on Earth for a long time, triangulating external objects and events with other rational creatures like herself. Then, Emily acquired lots of beliefs that, according to triangular externalism, must be largely true about Earth. At a certain point in her life, however, Emily’s brain is envatted by a mad scientist. Now, Emily still has beliefs about Earth, but they are largely false about her actual vat world. Such a skeptical situation is perfectly compatible with triangular externalism (Nagel 1999). To put it another way, the skeptic may argue that triangular externalism is not able to rule out the possibility of a Recent Envatment (RE), an envatment carried out after an

63

otherwise normal life made up of causal systematic interactions with external objects and events within an intersubjective social framework, such as the one required by Davidson’s theory of triangulation. In a RE situation Emily had triangulated external objects and events with other creatures like herself, she acquired beliefs about the outside world and, for a certain period, her beliefs were largely true about Earth. Now (after a RE), however, her cohering beliefs are totally false about Earth. An externalist like Davidson can obviously reply that sooner or later new and different causal relations with the vat world will be established. Thus, such a deception cannot last forever. Yet the skeptic has still another move. Again, let us assume that Emily was born and lived on Earth for a long time, triangulating external objects and events with other rational creatures like herself. Then Emily acquired lots of beliefs that, according to triangular externalism, must be largely true about Earth. At a certain point in her life, however, Emily’s brain starts to be envatted and unenvatted by a mad scientist, and each envatment is not long enough to allow Emily to establish new causal relations with her vat world and to become aware of her actual situation. Now, Emily still has beliefs about Earth, but they may be largely false if her actual world happens to be the vat. This skeptical situation is also compatible with triangular externalism, which is not able to rule out the possibility of an Enduring Envatment (EE), a series of envatments and unenvatments carried out after an otherwise normal life and such that each envatment is not long enough to allow the subject to establish new causal relations with the vat world. Thus, the most we can infer from triangular externalism is that our cohering beliefs cannot always have been totally false, or that our cohering beliefs are very probably not totally false. Triangular externalism, however, does not imply the stronger thesis that our cohering beliefs can never be totally false (Amoretti 2008a). Even granting that triangular externalism may be justifiable a priori (which in my opinion is still very dubious), is this weaker consequence enough to repel the skeptic? If we assume the skeptic’s own epistemic standards, Davidson’s anti-skeptical argument is probably just a blank shot, because it is not able to provide the skeptic with a strong and compelling reason to accept the idea that a coherent set of beliefs held true by the subject must be largely true about the world.

64

We may still wonder however whether the above conclusions are still sufficient to guarantee that we have a justification—at least conceived as external, since the subject may not be able to have reflective access to her own justificatory status—for our empirical beliefs, and thus genuine knowledge about the external world. I believe that the answer, again, depends on the epistemic standards we wish to adopt. A skeptic would presumably tend to embrace absolute standards of knowledge. In this case, saying that triangular externalism implies that our cohering beliefs are very probably largely true would not be sufficient to guarantee that we have a justification, not even conceived as external, for our empirical beliefs, and thus genuine knowledge about the external world. If, however, we think (as I am inclined to) that in this case ordinary standards of knowledge would be enough, then we would have a justification for our empirical beliefs that—even if it is conceived as external, because the subject may not be able to have reflective access to her own justificatory status—may still guarantee that a belief cohering with a vast array of other beliefs is prima facie justified, having in fact a presumption in favor of its truth. Again, this conclusion may be asserted only if one renounces to defend epistemic internalism in its stronger version, the one that best represents our deep intuitions about epistemic internalism itself. References Amoretti, M.C. 2007, “Triangulation and Rationality”. Epistemologia, 30 (2), 269-288. Amoretti, M.C. 2008a, “Davidson, Self-Knowledge, and Skepticism”. In M.C. Amoretti and N. Vassallo (eds.), Knowledge, Language, and Interpretation: On the Philosophy of Donald Davidson. Frankfurt: Ontos Verlag. Amoretti, M.C. 2008b, Il Triangolo dell’interpretazione: Sull’epistemologia Di Donald Davidson. Milano: FrancoAngeli. Amoretti, M.C. and Vassallo, N. 2008, “Introduction”. In M.C. Amoretti and N. Vassallo (eds.), Knowledge, Language, and Interpretation: On the Philosophy of Donald Davidson. Frankfurt: Ontos Verlag.

65

Audi, R. 2010, Epistemology: A Contemporary Introduction to the Theory of Knowledge, 3rd Edition. London: Routledge. Bernecker, S. and Dretske, F.I. 2000, Knowledge: Readings in Contemporary Epistemology. Oxford: Oxford University Press. Brown, J. 2004, Anti-Individualism and Knowledge. Cambridge, Mass.: MIT Press. Burge, T. 1979, “Individualism and the Mental”. Midwest Studies in Philosophy, 4, 73-122. Burge, T. 1986, “Individualism and Psychology”. The Philosophical Review, 95 (1), 3-45. Conee, E. 2007, “Externally Enhanced Internalism”. In S.C. Goldberg (ed.), Internalism and Externalism in Semantics and Epistemology. Oxford: Oxford University Press, 51-67. Davidson, D. 1982a, “Empirical Content”. Grazer Philosophische Studien, 1617, 471-489; repr. in Davidson 2001c (from which I quote). Davidson, D. 1982b, “Paradoxes of Irrationality”. In R. Wollheim and J. Hopkins (eds.), Philosophical Essays on Freud. Cambridge: Cambridge University Press, 289-305; repr. in Davidson 2004 (from which I quote). Davidson, D. 1983, “A Coherence Theory of Truth and Knowledge”. In D. Henrich (ed.), Kant Oder Hegel. Stuttgart: Klett-Cotta, 423-438; repr. in Davidson 2001c (from which I quote). Davidson, D. 1986, “A Nice Derangement of Epitaphs”. In R. Grandy and R. Warner (eds.), Philosophical Grounds of Rationality. Oxford: Oxford University Press, 156-174; repr. in Davidson 2005 (from which I quote). Davidson, D. 1987, “Knowing One’s Own Mind”. Proceedings and Addresses of the American Philosophical Association, 60, 441-458; repr. in Davidson 2001c (from which I quote). Davidson, D. 1988, “The Myth of Subjective”. In M. Krausz (ed.), Relativism. Interpretation and Confrontation. Notre Dame: Universtiy of Notre Dame Press, 159-172; repr. in Davidson 2001c (from which I quote). Davidson, D. 1989, “What Is Present to the Mind?”. In J. Brand and W.L. Gombocz (eds.), The Mind of Donald Davidson. Amsterdam: Rodopi, 318; repr. in Davidson 2001c (from which I quote).

66

Davidson, D. 1990a, “Afterthoughts, 1987”. In A. Malachowski (ed.), Reading Rorty. Oxford: Blackwell, 134-137; repr. in Davidson 2001c (from which I quote). Davidson, D. 1990b, “Epistemology Externalised”. Análisis filósofico, 10, 1-13; repr. in Davidson 2001c (from which I quote). Davidson, D. 1991, “Three Varieties of Knowledge”. Philosophy, 66, 156-166; repr. in Davidson 2001c (from which I quote). Davidson, D. 1994, “The Social Aspect of Language”. In B. Mcguinness (ed.), The Philosophy of Michael Dummett. Dordrecht: Kluwer, 1-16; repr. in Davidson 2005 (from which I quote). Davidson, D. 1999a, “The Emergence of Thought”. Erkenntnis, 51, 7-17; repr. in Davidson 2001c (from which I quote). Davidson, D. 1999b, “Reply to Genova”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. Chicago: Open Court, 192-194. Davidson, D. 1999c, “Reply to Nagel”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. Chicago: Open Court, 207-210. Davidson, D. 2001a, “Comments on Karlovy Vary Papers”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 285-308. Davidson, D. 2001b, “Introduction”, in Davidson 2001c, xiii-xviii. Davidson, D. 2001c, Subjective, Intersubjective, Objective. Oxford: Oxford University Press. Davidson, D. 2004, Problems of Rationality. Oxford: Oxford University Press. Davidson, D. 2005, Truth, Language, and History. Oxford: Oxford University Press. Fennell, J. 2000, “Davidson on Meaning Normativity: Public or Social”. European Journal of Philosophy, 8 (2), 139-154. Føllesdal, D. 1999, “Triangulation”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. Chicago: Open Court, 719-728. Glüer, K. 2001, “Dreams and Nightmares: Conventions, Norms and Meaning in Davidson’s Philosophy of Language”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 53-74.

67

Glüer, K. 2006, “Triangulation”. In E. Lepore and B. Smith (eds.), Oxford Handbook of Philosophy of Language. Oxford: Oxford University Press. Goldberg, S.C. (ed.) 2007, Internalism and Externalism in Semantics and Epistemology. Oxford: Oxford University Press. Lepore, E. and Ludwig, K. 2005, Donald Davidson. Meaning, Truth, Language, and Reality. Oxford: Clarendon Press. Ludlow, P. and Martin, N. 1998, Externalism and Self-Knowledge. Stanford: CSLI. Nagel, T. 1999, “Davidson’s New Cogito”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. Chicago: Open Court, 195-206. Nuccetelli, S. (ed.) 2003, New Essays on Semantic Externalism and SelfKnowledge. Cambridge, Mass: MIT Press. Pagin, P. 2001, “Semantic Triangulation”. In P.Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 199-212. Putnam, H. 1975, “The Meaning of ‘Meaning’”. In K. Gunderson (ed.), Language, Mind and Knowledge. Minneapolis: University of Minnesota Press, 215-271. Putnam, H. 1981, “Brains in a Vat”, Reason, Truth, and History. Cambridge, Mass.: Cambridge University Press, 1-21. Ramberg, B. 2001, “What Davidson Said to the Skeptic. Or: AntiRepresentationalism, Triangulation and the Naturalization of the Subjective”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 213-236. Sosa, E. 2003, “Knowledge of Self, Others, and World”. In K. Ludwig (ed.), Donald Davidson. Cambridge, Mass: Cambridge University Press, 163182. Vassallo, N. 2003, Teoria della Conoscenza. Roma-Bari: Laterza. Verheggen, C. 1997, “Davidson’s Second Person”. The Philosophical Quarterly, 47 (188), 361-369. Verheggen, C. 2007, “Triangulating with Davidson”. The Philosophical Quarterly, 57 (226), 96-103.

68

Wright, C., Smith, B.C., et al. (eds.) 1998, Knowing Our Own Minds. Oxford: Clarendon Press.

TRIANGULATION TRIANGULATED Kirk LUDWIG (Indiana University, Bloomington, USA) 1. Introduction Appeal to triangulation occurs in two different contexts in Davidson’s work.1 In the first, triangulation—in the trigonometric sense—is used as an analogy to help explain the central idea of a transcendental argument designed to show that we can have the concept of objective truth only in the context of communication with another speaker. In the second, the triangulation of two speakers responding to each other and to a common cause of similar responses is invoked as a solution to the problem of underdetermination of thought and meaning by the patterns of causal relations we stand in to the environment. I examine both of these uses of the idea of triangulation. In section 2, I take up the use of triangulation as an analogy in connection with Davidson transcendental argument to establish that communication is essential for the concept of objectivity. I argue that it is unsuccessful because the case has not been made that scope for deploying the idea of contrasting perspectives, which is needed for the concept of objectivity, is available only in the context of communication. In section 3, I take up the idea that triangulation on a common cause of common responses of two creatures interacting with each other provides the additional constraint needed to assign objective content to our thoughts and words. I show that appeal to this sort of triangulation provides minimal help in responding to the problem it is intended to solve. Section 4 provides a brief summary and conclusion. 2. Triangulation as an Analogy Triangulation is a technique for determining indirectly a feature of something, its distance from a baseline, by measuring something systematically related to it. In the general case, one determines the angle of an object from 1

I cite the original publication date of Davidson’s papers; page numbers, however, unless otherwise indicated, will be to reprints in (Davidson, 2001).

70

one’s position at the two end points of a baseline of known length. The angles and baseline determine uniquely the height of the triangle formed and can be calculated using trigonometric functions. Davidson first invokes triangulation in this sense as an analogy in “Rational Animals” (1982). In “Rational Animals”, he argued that it is necessary and sufficient for having propositional attitudes that one have the capacity to speak a language and to interpreter other speakers. The argument has three premises (1982: 102). (1) [A]ll propositional attitudes require a background of beliefs. (2) [I]n order to have a belief, it is necessary to have the concept of belief. (3) [I]n order to have the concept of belief one must have language. The analogy with triangulation arises in connection with the third premise, which I concentrate on here. The argument depends on the idea that there must be an appropriate sort of ground for attributing a concept to a creature. This comes out in passages [a] and [b]. [a] Much of the point of the concept of belief is that it is the concept of a state of an organism which can be true or false, correct or incorrect. To have the concept of belief is therefore to have the concept of objective truth. If I believe there is a coin in my pocket, I may be right or wrong; I’m right only if there is a coin in my pocket. If I am surprised to find there is no coin in my pocket, I come to believe that my former belief did not correspond with the state of my finances. I have the idea of an objective reality which is independent of my belief (1982: 104).

Davidson follows this with the observation that complex interaction with the world, the possibility—described in a behaviorist vocabulary—of discriminating properties and generalizing, in the sense of reacting to new stimuli in the same ways as to prior stimuli, is not sufficient to attribute to something the concept of belief. This is followed by the claim that linguistic communication would suffice. [b] What would show command of this contrast? Clearly linguistic communication suffices. Communication depends on each communicator having, and correctly thinking that the other has, the concept of a shared world, an

71

intersubjective word. But the concept of an intersubjective world is the concept of an objective world, a world about which each communicator can have beliefs (1982: 105).

When we put [a] and [b] together, it is clear that the idea is that to have the concept of belief there must be a point to having it for its possessor, that is to say, there must be a scope for its application within its experience. It is clear why, on Davidson’s view, there should be scope for its application in the context of linguistic communication. In interpreting others we are guided by the principle of charity, which involves two components that are in tension with each other. One Davidson has called the principle correspondence, and the other the principle of coherence.2 The principle of correspondence tells us to find the speaker mostly right about her environment. The principle of coherence tells us to find her largely rational. The former principle is needed to solve the interdependence of belief and meaning, which emerges in considering how to get from the identification of a correlation between a hold true attitude toward a sentence and certain circumstances in the environment to the meaning of the sentence and content of the belief. If we know that it is a lawlike regularity that [L] ceteris paribus, Karla holds true s iff p we are not yet in a position to say either what Karla believes or what she means. Her hold true attitude is the result of what she believes and what (she knows) her sentence to mean: if she believes that p and (knows) her sentence s means that p, then she holds true s. Thus, to solve for either belief or meaning, knowing only what she holds true, we must bring to bear some additional constraint. The principle of correspondence holds that what the speaker believes about her environment is true, with the goal of allowing us then to infer that the conditions under which she holds true s as identified in [L] give both the content of her belief and the meaning of the sentence she holds true on its basis.3 The principle of coherence is a holistic constraint. It holds that the speaker, being an agent, is largely rational, 2

See “Three Varieties of Knowledge” (1991b: 211). See Lepore and Ludwig (2005: ch. 12-15) for an extended criticism of this solution to the problem.

3

72

and, hence, that her attitudes must be attributed in largely rational patterns. Since perspectives on the world differ, in interpretation there will inevitably be circumstances in which we may find a tension between making the subject of interpretation more rational by dint of finding her mistaken in her beliefs here and there, or less rational by finding less error. It follows that the concept of the contrast between how things are believed to be and how things are is an essential component of the conceptual scheme of the interpreter. In “Rational Animals”, Davidson notes that the sufficiency of having a language for having the concept of belief is not adequate for his argument. He needs to show that it is necessary. He offers the analogy with triangulation as a substitute, designed to help persuade in the absence of a proof, as expressed in [c]. [c] If I were bolted to the earth, I would have no way of determining the distance from me of many objects. I would only know they were on some line drawn from me towards them. I might interact successfully with objects, but I could have no way of giving content to the question where they were. Not being bolted down, I am free to triangulate. Our sense of objectivity is the consequence of another sort of triangulation, one that requires two creatures. Each interacts with an object, but what gives each the concept of the way things are objectively is the base line formed between the creatures by language. The fact that they share a concept of truth alone makes sense of the claim that they have beliefs, that they are able to assign objects a place in the public world (1982: 105).

The analogy supports the interpretation given of Davidson’s argument above. If I could not move, I would (Davidson says) not be able to determine the distance of objects from me. I cannot measure it directly, but neither can I measure a baseline to determine it indirectly by triangulation. And this is not just an epistemic limitation: “I could have no way of giving content to the question where they were” in respect of distance from me. He thus links the idea of being able to think a thing a certain distance with a method of determining it, a procedure for the (correct) application of the concept, and so to conditions in which aspects of experience would stand as ground for the application of the concept.

73

Triangulation determines the distance of an object by way of fixing a baseline and the two angles that the lines from its ends to the object form with it. The concept of an object’s distance has a place in a network of concepts in which it is related systematically to the angles from which it is seen and the distance between the viewing locations. When we are in a position to make use of the entire scheme that specifies these interrelations, we can give content to the idea of distance. Davidson’s idea is that the concept of objectivity likewise has its place in a network of concepts that includes the concepts of the propositional attitudes, and centrally of belief, truth, falsity, evidence, and error, and that the entire scheme is bound up with the capacity for speech because it puts us in contact with the potentially differing thoughts of others. The idea is that I alone, like the man bound to a single position who cannot give content to the idea of distance, could not give content to the idea that the world does not correspond to my image of it. Without that, I could not give content to the idea of the contrast between truth and falsity, and so could not give content to the idea of belief, as it depends on that. What is needed is something akin to the capacity for movement, which gives me the two different perspectives on an object crucial to using triangulation to determine its distance, and so a use for the scheme in which the concept has a role. In the case of the concept of objectivity, it is the possibility of identifying a perspective which potentially contrasts with my own that gives content to the idea of error. The two perspectives minimally needed are my own and that of another whose thoughts are focused on the same world. The possibility of a contrast between the two gives scope for the concept of error, and so of objectivity, both being bound up with the idea of misrepresentation. Thus, if Davidson is right, to have the idea of a world that is independent of the way one represents it, i.e., objective, is at the same time to have the idea of a world which is intersubjective, for to be able to think it objective requires thinking of it in the context of a contrast with another’s perspective on it. To have this idea, in turn, requires (on the assumption that to have a concept there must be scope for the correct application of it in one’s experience) that one to be able to identify another together with his perspective. At this point the idea of language enters. For to complete the analogy, we must suppose that the identification of another perspective that contrasts

74

with one’s own requires one to be in communication with another. If we grant this, then the emergence of the two perspectives is made possible by a shared language. This is the sense in which a shared language (and shared concept of truth therefore) is the baseline for the concept of objectivity. For just as a shared language makes possible the two perspectives that give content to the idea of objectivity, the baseline in triangulation makes possible the two perspectives on an object that gives content to the idea of its distance from us. Davidson does not offer this analogy as a proof that only in the context of communication can the concept of objectivity, and so of belief, arise. But he assumes this in later work, as shown in [d] from “Epistemology Externalized” (1991a).4 [d] I do not mean […] that one creature observing another provides either creature with the concept of objectivity; the presence of two or more creatures interacting with each other and with a common environment is at best a necessary condition for such a concept. Only communication can provide the concept, for to have the concept of objectivity, the concepts of objects and events that occupy a shared world, of objects and events whose properties and existence is independent of our thought, requires that we are aware of the fact that we share thoughts and world with others (Davidson 1991a: 202).

The argument, which we can all the argument from error, is summarized here.5 (1) To have the concept of a belief, one must have the concept of error, i.e., of objective truth, of a way things are independent of how one believes them to be. (2) That a creature possesses the concept of error or objectivity stands in need of grounding, that is, specifically, we must be able to make sense of there being scope for the (correct) application of the concept in the creature’s experience and behavior. 4

See also, “Indeterminism and Realism”: «If we were not in communication with others, there would be nothing on which to base the idea of being wrong, or, therefore, of being right, either in what we say or in what we think» (Davidson 1997: 83). 5 See Lepore and Ludwig (2005: 397-9) for related discussion.

75

(3) There is scope for the application of the concept of error or objectivity in a creature’s behavior if but only if it shares a language with another with whom it is (or has been) in communication. (4) A creature possesses the concept of error or objectivity only if shares a language with another with whom it is (or has been) in communication (from 2 and 3). (5) A creature possesses the concept of belief only if it shares a language with another with whom it is (or has been) in communication (from 1 and 4). With the assumption that to have belief one must have the concept of belief this delivers the conclusion that language is essential for thought. The two crucial premises are (2) and (3). Premise (2) is an instance of a more general requirement on the possession of a concept: that there be scope within a thinker’s experience for it to manifest its grasp of the application conditions of the concept by deploying it on the basis of the appropriate ground for it. The idea goes back at least to Kant’s thought that the possibility of self-conscious experience requires the application in experience of certain general concepts, this being at the same time a condition on the determinate possession of the concepts themselves.6 Two issues come up in connection with this. The first is whether what we should require is the potential for application of the concept in relation to appropriate experience as opposed to actual application in relation to appropriate experience. The second is whether we should require subjective or object deployment of the concept. Since to possess a concept is to have a disposition to deploy it in appropriate circumstances, it is unclear why the actual application to a course of experience is needed for a creature to possess a concept, as opposed to its being able to apply it in appropriate conditions, even in the case of basic concepts. Though we acquire our basic concepts in conditions in which there is scope for their application, what matters for the possession of the concepts is the state we end up in, not how we end up in them. For when we check to see what concepts someone has, it is not his history that mat6

See in this connection Strawson (1959, 1966).

76

ters but what he can do with it. This is recommended by Davidson’s own view that the most fundamental stance is on the nature of meaning and thought is that of the radical interpreter, whose evidence ultimately consists of a creature’s behavior in interaction with its environment (including others of its kind). The theory formed on this basis is a theory seeks to fit a scheme of interpretation onto the creature’s dispositions to interact with its environment, because it aims to explain both observed behavior and what it will do and say in a variety of counterfactual circumstances. If the radical interpreter could have exhaustive knowledge of the creature’s dispositions, he would already be in a position to interpret it, given knowledge of the environment relative to which he was to be interpreted. If concept possession, even for basic concepts, does not require actual application, but only the ability to apply them when presented with appropriate circumstances, then actual communication with another would not be required in order to have the concept of belief, even granting the rest of the argument. The second issue is whether the requirement that to possess a concept, a creature have scope in its experience and behavior for its application, is to be read as requiring correct application or only that from the subjective standpoint of the creature itself it appears so. This raises a familiar difficulty for transcendental arguments from conditions on concept possession to objective application of the concepts (Stroud 1968). For illustration of the general difficulty, consider an argument for the necessity of reidentification of objects for the possession of the concept of an object. Plausibly, if there were no scope in experience for the reidentification specific objects, at least in the sense that one’s experience was not rich enough to support such reidentification, one could not have the concept of an object at all. But why isn’t it enough that one’s subjective experience be rich enough to provide evidence of sameness? Why should not a brain in a vat, for example, whose brain states are type identical to one of ours and whose experience is subjectively qualitatively the same, not be in as good a position to have the concept of an object as we are? To bridge the gap, as Stroud noted, we must in effect rely on a verification principle. If concept possession, even for basic concepts, does not require correct application, but only application (actual or potential) on the basis of subjective conditions appropriate for their deployment, then even granting the

77

rest of the argument, it would not follow that one had to be in communication with others to have the concept of belief. In raising these two issues about premise (2), I mean not to settle them in favor of the skeptic, but only to identify what would be required to complete the argument, and to place it in its historical context. However, even if we are only able to endorse a weaker principle, if the rest of the argument is correct, we would still be able to conclude language is necessary for thought. There are two readings of each of two aspects of the principle. There is the dispositional versus actual reading of concept application. Then there is the subjective versus objective reading of concept application. This gives us four interpretations altogether: dispositional objective (DO); dispositional subjective (DS); actual subjective (AS); actual objective (AO). Davidson assumes the last. If we reject AO, we are left with three readings. (DS) is the weakest reading, but it still gives us the conclusion that to possess the concept of objectivity one must possess a language, i.e., be in a position to, in appropriate circumstances, communicate with another who shares a language with one. For it requires one be in a position, granting the rest of the argument, to be able, in response to appropriate subjective experience, to respond properly as if one were in communication with another. The same conclusion then follows from each of the others. Granting that the concept of belief is required for belief, we are still able to reach the main conclusion of “Rational Animals”, that language, if not actual communication, is necessary for thought. This highlights the importance of premise (3). The left to right direction of the biconditional can be granted. The question is whether language is necessary for the concept of error. If we take the analogy with triangulation to indicate what the basic requirement is, it is that we be able to make sense of different perspectives on the same world, and so a contrast between getting it right and getting it wrong. The question then is whether we can make sense of distinct perspectives without admitting communication with another whose point of view we identify with the second perspective. There are two possibilities which would have to be ruled out for premise (3) to be established. The first is that the different perspectives are provided by a single agent thinking about the same situation at different times. The second is that the different perspectives are provided by distinct indi-

78

viduals capable both of thought and of thought about the thoughts of others, but who do not possess language. The first suggestion is that the first person perspective itself provides adequate scope for the concept of error by appeal to a difference in perspective on the same world from different times. Suppose that at time t I think that there is a man standing on a hillside. This prompts me to approach. At t + two minutes, when I am closer, I look again and there appears to be only a small, withered tree where I believed a man was standing, and I come so to believe. These beliefs at t and at t + two minutes are not inconsistent with each other, but they form an inconsistent set with my belief that two objects cannot occupy the same place at the same time, that small withered trees do not come into existence rooted into the ground in ten minutes, and that a man cannot be transformed into a small withered tree. To maintain consistency, I must give up one of my beliefs. Nothing here requires that I have the concept of belief, but if I do, then there is clearly scope for the application of the concept of false belief. Just as in interpretation one achieves a better understanding of another by sometimes attributing false belief to him, so in one’s own case, as time goes on, one achieves a better understanding of oneself and the world by attributing to oneself false beliefs in the past. The second suggestion is that another’s perspective on the world may be identified independently of sharing a language with it. This depends upon the possibility of identifying another as a thinker and as thinking about particular things without sharing a language with it. This is as plausible as that a non-linguistic being can be attributed propositional attitudes to explain its behavior. There is no doubt that we routinely explain the behavior of nonlinguistic animals using the framework of propositional attitude psychology, as in the example Norman Malcom gives of a dog barking up the wrong tree because it things mistakenly the squirrel it is after is in it,7 and that we have no better or more accurate way of explaining and anticipating their behavior. There is therefore a prima facie case to be made for the possibility of identifying another as a thinking being without sharing a language with it. And if this is possible, then we can make sense 7

Davidson discusses this case of Malcolm’s in “Rational Animals” (1982: 96-7); Malcolm’s discussion appears in “Thoughtless Brutes” (1973).

79

of another perspective on the same circumstances which we may want to see as involving a mistake to make better sense of the other as a rational being, as we do in attributing to a dog the mistaken belief that the squirrel is in the tree up which it is barking. Davidson raises doubts about the adequacy of our practice, which focus on the question whether the behavior of nonlinguistic animals supports the dense interconnections between the concepts expressible in natural languages (Davidson 1982: 97-101). This is rather a doubt about whether the concepts we perforce use in attributing attitudes to nonlinguistic animals are too fine-grained, rather than a doubt about the applicability of the framework of propositional attitude psychology as such. We may here invoke Davidson’s own analogy with measurement theory to make sense of our practices with nonlinguistic animals (1989: 59-60). There are many adequate ways of mapping our concepts onto theirs because the structures in which ours stand are richer than those in which theirs stand. So just as there are different adequate mappings, relative to some arbitrary starting choices, in assigning numbers to temperature (the assignment of 0 and an interval), so there are many adequate mappings of our concepts on to those of, e.g., dogs, relative to some arbitrary starting choices. And it will do no good here to point out that this conflicts with the claim that there can be thought without language, for the point of the argument in question is to establish it, and if it has to assume it at some point, then it begs the question. To sum up the discussion so far, premise (2) appears to be too strong in two ways. First, insofar as we think of concept possession as a dispositional trait, actual deployment is not necessary for possession, but only the capacity to correctly deploy it as circumstances warrant. Second, it is unclear that concept possession requires more than that there be point to the deployment of a concept from the subjective point of view of the agent. Despite this, the argument for the necessity of language for thought will still go through even on the assumption that concept possession requires only a disposition to deploy a concept in response to subjective experience in a way that expresses grasp of its application conditions, if the remaining premises are correct. This shows that premise (3) is the crucial premise. However, there are strong grounds for rejecting premise (3). The thought underlying (3) is that (i) the concept of objectivity requires making sense of differing perspectives on the world and (ii) to make sense of different

80

perspectives one must identify a distinct individual with thoughts and (iii) the only way to do that is by way of sharing a language and being in communication with her. The difficulty is that, first, an thinker’s own standpoint on the same circumstance at different times can provide the difference in perspective needed to make sense of a world independent of thought, and, second, even if a second person were required, the assumption that a distinct thinker could be made sense of only if she shared a language with one is tantamount to the intended conclusion of the argument, and so cannot be invoked in the face of the prima facie intelligibility of identifying nonlinguistic thinking beings by way of patterns in their behavior. 2. Triangulation as a Solution to Underdetermination Triangulation emerges as a solution to the problem of determining the objective content of thoughts in “Epistemology Externalized” (1991a) and “Three Varieties of Knowledge” (1991b). The problem is described in passage [e] from “Epistemology Externalized”:8 [e] the cause of certain mental states is relevant to the content of those states. […] one kind of case is especially important: an example is the way the fact that a certain mental state has been typically caused by seeing cows allows us to think ‘There’s a cow’ even when no cow is present. But here a problem arises. What determines the content of such basic thoughts […] is what has typically caused similar thoughts. But what has typically caused them? There are many choices, for example events that occurred before all cows, or events spatially closer to the thinker than any cow (Davidson 1991a: 201).

The trouble is that, even identifying a common response on different occasions of a creature to its environment, there will always be a variety of different common causes we could choose as the one that it is responding to. There is the cow, there are common causes of all cows, there are events between the cow and the observer, there are events at the observer’s sensory surfaces. Perhaps there will not be salient commonalities for us between 8

See also “The Second Person” (Davidson 1992: 118).

81

many of these events. But so far as the objective facts go, there is going to be some common pattern that can picked out (allowing for some false positives as we must in the case of the cow also). For there is a shared causal power, exercised against the rest of the background conditions that results in the common response. The question is at what common link in the causal chains leading up to a response one should locate the object of the thought, if any, that it expresses. There is even a problem about what to count as the same response to stimuli, as Davidson notes. For what a creature does on different occasions in response to its environment will be similar and different in endless ways. Which of the similarities in response across various occasions in which a creature causally interacts with its environment should we treat as the relevant one? Davidson accepts that if no answer can be given to these questions, no sense can be made of the response being an expression of a thought at all. Davidson’s general methodological stance on thought and meaning requires that the facts about them be recoverable from the third person standpoint. This is the basis for his externalism, for his claim that one cannot be mostly wrong about the world, and for his claim that one must know what one means and thinks and be able to tell in the case of others. From this standpoint, the question is what objective resources are available to solve the problem of the determination of thought content. There are two problems to be solved simultaneously. One is the problem of determining when a subject responds in a way similar to the way he has responded to the environment previously. The other is the problem of determining what he is responding to. The solution Davidson offers appeals to the perspective of the radical interpreter, of one subject communicating with another. The idea is that objective content can be assigned if, but only if, we can see the creature in question as in communication with another about its environment.9 What 9

As he says in “The Second Person”, «If we consider a single creature by itself, its responses, no matter how complex, cannot show that it is reacting to, or thinking about, events a certain distance away rather than, say, on its skin. The solipsist’s world can be of any size; which is to say, from the solipsist’s point of view it has no size, it is not a world» (Davidson 1992: 119).

82

this provides is a common object of thought, by way of a common cause of a similar response in each to the object. Where the causal chains leading to their responses overlap is where we locate the object of thought. This is illustrated in diagram 1.

Diagram 1

The inclusion of an additional subject provides an additional constraint. Now we look not just at the responses of a single subject on difference occasions to the environment, but minimally of two subjects interacting with each other and the environment, as is required for us to conceive of them in communication with one another. Then we require common responses in both of them and common causes of those responses, on occasions on which there is interaction between them that can be interpreted as communication. The objects of their thoughts, if any, are those that are the common causes of their common responses to the environment in communicative situations. Davidson puts it this way in “Three Varieties of Knowledge” (1991b):10 [g] It takes two points of view to give a location to the cause of a thought, and thus to define its content. We may think of it as a form of triangulation: each of two people is reacting differentially to sensory stimuli streaming in from a certain direction. Projecting the incoming lines outward, the com10

See also “Epistemology Externalized” (Davidson 1991a: 203), and “The Second Person”, (Davidson 1992: 119): «[w]here the lines from child to table and us to table converge, ‘the’ stimulus is located. Given our view of child and world, we can pick out ‘the’ cause of the child’s responses. It is the common cause of our response and the child’s response».

83

mon cause is at their intersection. If the two people now note each other’s reactions (in the case of language, verbal reactions), each can correlate these observed reactions with his or her stimuli from the world. A common cause has been determined. The triangle which gives content to thought and speech is complete. But it takes two to triangulate (Davidson 1991b: 213).

In evaluating this proposal, we need to keep in mind that the task Davidson has set himself is to describe what objective evidence suffices to interpret another speaker. The problem of thought content arises in this context. It does no good to ask for the common cause of common responses taking a creature by itself because there is nothing that will serve to distinguish a common response to a common cause. We cannot appeal to the creature’s point of view because that has to be constructed from the objective facts. But the idea is that something will present itself if we think about the subject as a speaker, for then there is not just the subject responding to his environment, but the subject responding to his environment and responding to another subject responding to her environment and to him in turn. For this to serve, the descriptions of the situations must be given in a vocabulary that does not presuppose that the subject speaks a language or has any thoughts. For the claim is that the non-intentional and non-semantic facts conceptually ground the application of the intentional and semantic concepts.11 We are to see that just the causal facts underlying episodes of communication in which there is a triangulation between object, two individuals, and each other is adequate for sufficiently determinant schemes of 11

This is not the same as saying that the concept of thought or meaning is reducible to something else, a claim that Davidson has repeatedly denied. What is at issue here is supervenience, not reduction. Supervenience is implied by Davidson’s basic stance on what the relevant evidence from which the facts about meaning and thought, which he regards as inextricably linked, can be recovered. As he has put it at one place: «The semantic features of language are public features. What no-one can, in the nature of the case, figure out from the totality of the relevant evidence cannot be part of meaning» (Davidson 1979: 235). This is a thought that Davidson takes from Quine. Where they differ is on whether the concepts of meaning and the propositional attitudes are fully legitimate. Quine offers replacements for these concepts constructed on the basis of the concept of stimulus meaning in Word and Object (1960). Davidson’s offers an explication in terms of the holistic fitting of a theory deploying the concepts to the totality of relevant evidence.

84

interpretation. This together with the inadequacy of the objective causal information when we take a creature by itself as interacting with its environment, then, is to show that the causal facts that ground the possibility of communication at the same time ground the possibility of thought.12 If, as Davidson assumes, communication requires knowledge of one’s own thoughts, and knowledge of the thoughts of others,13 then, since knowledge of the thoughts of others requires knowledge of the external world, it would follows that thought requires knowledge of one’s own mind, of the external world, and of the minds of others (Davidson 1991b: 213). These three varieties of knowledge would then fall out, on Davidson’s view, as the birthright of language, as language is the birthright of thought. A further claim follows, though Davidson does not mention it, namely, that, since knowledge of other minds presupposes generalizations from others’ behavior in projecting to their meanings and thoughts, we are guaranteed, if Davidson is right, that induction yields knowledge, if we are able to think at all. This is a transcendental argument, in the sense that it aims to establish knowledge of one’s own mind, of the external world, of the unobserved (via induction), and of other minds as a condition on the possibility of any thought at all. In ambition, it parallels Kant’s argument in The Critique of Pure Reason that knowledge of things in space and time, of the self and objective causal regularities is a condition on the possibility of selfconscious experience. It differs in aiming to secure knowledge of things in space and time as features of a completely objective, mind-independent reality, and not merely as of appearances from the transcendental standpoint. Before proceeding, let me summarize Davidson’s argument.

12

In “Indeterminism and Antirealism” (1997) Davidson connects this with Wittgenstein’s reflections concerning rule following: «Without a second person there is, as Wittgenstein powerfully suggests, no basis for a judgment that a reaction is wrong, or, therefore, right» (Davidson 1997: 83). 13 This is not immediate because of the gap between true belief and knowledge, and so this further inference requires further argument.

85

(1) The fundamental ground for the application of semantic concepts and the concepts of the propositional attitudes consists of the nonsemantic and non-intentional facts about the creature’s interaction with its environment (call these “the basic facts”). (2) To make sense of a creature having a thought about its environment, the basic facts must enable us to determine the object of its thought (from 1). (3) To determine the object of a creature’s thought on the basis of the basic facts we must be able to identify a common cause of a common response to stimuli that is a better candidate for what it is thinking about than any others. (4) Across any series of occasions on which a creature considered alone is interacting with its environment, given the basic facts, there will be many common causes of for each common response of the creature to its environment, and many common responses, which provided equally good candidates for what the creature is thinking about. (5) Therefore (from 2-4), the causal interaction of a single creature alone with its environment is not adequate to make sense of a creature having a thought about its environment. (6) There are objective circumstances in which, when two creatures interact with each other and with the environment, one can identify, on the basis of the basic facts, the common cause across a series of occasions of a common shared response each has to the environment (as in diagram 1). (7) Such circumstances, then (from 1-3, 5-6), are essential for the making sense of a creature having a thought about its environment. (8) Furthermore, the possibility of ascribing thoughts in such circumstances to either of the creatures depends on their interactions with each other and the environment being sufficiently complex. (9) Their interactions with each other and the environment are sufficiently complex in such circumstances to ascribe thoughts to them only if the basic facts license interpreting their interactions with each other as communication. (10) Therefore (from 7-9), a creature can have thoughts about its environment only if it is (or has been) in communication with another creature.

86

(11) A creature can have thoughts at all only if it is capable of thinking about its environment (from 1). Therefore, a creature can have thoughts at all only if it is (or has been) in communication. The underived premises are (1), (3), (4), (6), (8) and (9). I will concentrate on (6), and end with a few remarks on (1). As a preliminary remark, as the conclusion is supposed to express a necessary condition on the possibility of thought, each of the premises must be presented as having its ground in the nature of its subject matter, that is, as having an a priori ground in the nature of the concepts deployed, and in evaluating (6) and (1) we will take them as so intended.14 Premise (6) expresses the central idea: that in considering two creatures responding to each other and in a common way to a common object in the environment we find an objective feature of their interactions with the world that suffices to determine a candidate for a sufficiently unique object of their thought. This is supposed to be what provides the ground for thinking of them as responding to something further out than their own sensory surfaces. When we keep our focus on what we have called the basic facts, however, it is not clear that adding additional responders to the environment does provide an adequate objective ground for determining what they are thinking about. The trouble arises both with respect to what to classify as a common cause and what to count as a common response. First, suppose that we have settled on what is to count as a common response. Among the common causes of anyone’s response to a cow is one 14

Davidson has expressed support for Quine’s rejection of the analytic/synthetic distinction, which Quine allied with the rejection of the a priori/a posteriori distinction. Yet it is clear that Davidson conceives of himself as providing illumination of concepts and of their connections with the evidence on the basis of which they applied. This is explicit in the following passage from “Rational Animals”: «There are conceptual ties between the attitudes and behavior which are sufficient, given enough information about actual and potential behavior, to allow correct inference to the attitudes» (Davidson 1982: 100). My view is that much of the interest of Davidson’s project rests on seeing him as aiming to illuminate the structure of our psychological and semantic concepts. If he were only making the claim that as a matter of fact only creatures that communicate have thoughts, then (i) it would be of considerably less interest and (ii) the project would fall squarely in the empirical discipline of animal psychology, and the methods he uses would be poorly matched to the enterprise.

87

that Davidson mentions as a problem in the case of a single individual, namely, all the common causes of cows in the first place. If that is legitimate to cite in the case of an individual, it is equally reasonable to cite it in the case of two individuals. But the problem is more radical than this suggests. To see this, take an example in which it is relatively easy to see that there are many common causes of a response to something in the environment, two people watching news on television, who each make the same exclamation in response to a stimulus. There is a causal chain that leads from each of the viewer’s eyes to the screen. If it is a cathode ray tube, then there is an electron beam that is the cause of the light emissions at the surface of the screen, which is in turn driven by a signal, which is traceable through radio waves or a cable to a television studio, and, thence, to various events around the world present and past. In this case, it is obvious that there are multiple common causes of the common response. It is easy for us to trace a particular causal chain backwards in this case because we are responsible for setting it up. However, this merely illustrates something that is always present. Wherever there is a common cause of a common response in two individuals, there will be multiple common causes because there are a variety of causal chains that can be traced back from that common cause. The particular series of causes we pick out are not, of course, sufficient for the generation of the responses. Many other conditions have to be in place for those responses to occur. In general, we have to take into account everything in the past light cone of the event.15 The past light cone of an event is the region of space-time such that from any point in it a ray of light can reach the event. Since no information can be transmitted faster than the speed of light, this determines the region from which causal influence may be transmitted to the point or region at which the event takes place. The light cone can be illustrated for a world with two spatial dimen15

This discussion is couched in the framework of special relativity. The point, though, holds in a classical framework and in general relativity. So far as I can see, bringing quantum mechanics into the picture would not change anything relevant to the main point. It would require shifting to a framework for talking about probabilistic dependence. But the point that there are many paths of dependence we can trace back through time leading up to any current event depending on what we treat as background conditions is unaffected.

88

sions in diagram 2, where the vertical axis is the temporal dimension. For the three dimensional case, the past light cone forms an expanding sphere as one goes back in time; at any past time, events outside the sphere at that time are causally irrelevant to the event in question.

Diagram 2

When we trace back a causal chain of events or conditions from an event, we hold fixed everything else in the light cone, as background conditions relevant to those causes having the effects that we are interested in. This may be represented by a line traced back through the light cone in past time. For any event, there will be multiple lines of causal influence we can trace back through the light cone, when we hold the rest fixed. Therefore, there will never be just one common cause of a common response in two individuals. It is natural to suggest that if we look for what is consistently the common cause of a common response in two or more individuals then we will be able to identify a unique cause. But it is hard to see why we should suppose this. When two people sit down in front of the television each night to watch the news, where there are similar causes of similar re-

89

sponses, there are similar causal chains leading up to them. And there is a more radical difficulty, which Davidson recognizes in the case of a creature interacting alone with its environment, namely, the question of how we are to type causes. In the case of a single individual, there might be features of the causes of its responses which we find saliently similar. But this would not settle that it was responding to something of that type. So far as the basic facts go, it could be treated as responding to (what seem to us to be) a complex disjunctive type (being A or B or C …), as long as some (perhaps rough) law connects things falling under that type with the response. But the more general point is that if we have to construct the categories of a creature’s thoughts out of the basic facts, we cannot privilege the types that seem natural to us. If we cannot do this in the case of an single creature alone interacting with its environment, we cannot do it in the case of a pair of individuals interacting with their environment, and if any types are allowable, there is little hope that adding another individual responding to the environment over time will help to identify a unique type of cause of a particular type of response. A similar problem attends the identification of a common response. We want to identify a type of response. However, just as in the case of an individual interacting alone with his environment there will be many different types of responses we can identify as in common on the various occasions on which it interacts with its environment, so too in the case of a pair of individuals, if we do not limit the categories we can appeal to, there will be many different types of responses we can identify as common on the various occasions on which they interact with each other and the environment.16 16

Davidson of course holds that there is indeterminacy in interpretation. Why not bite the bullet here and say that this is just an example of indeterminacy? One reason is that the same could be said about the creature considered by itself: it has thoughts as well, so communication is not necessary to fix objectively the content of thought, it is just that thought exhibits radical indeterminacy. If this is not a response Davidson is willing to accept in the case of a single individual, he should not be willing to accept it in the case of two, when the grounds for denying the basic facts are adequate are of basically the same kind. Beyond this, there has to be a principled basis for invoking indeterminacy. Davidson’s reason is that semantic matters, and those conceptually tied to them, as he thinks propositional attitude psychology is, must be in principle publically

90

When we ask how it is that we do manage to interpret others, there is a natural response. It is one that Davidson appears to give in his preamble to his identification of triangulation as what solves the problem of the common cause. It is that it «is we humans for whom [certain] classifications are complicated and impossible to articulate» (Davidson 1991a: 202) and others natural. It is we who class cow appearances together, more or less naturally, or with minimal learning. And even so, another classification is required to complete the point, for the class of relevant causes is in turn defined by similarity of responses: we group together the causes of someone’s responses, verbal and otherwise, because we find the response similar. What makes these the relevant similarities? The answer again is obvious; it is we, because of the way we are constructed (evolution had something to do with this), who find these responses natural and easy to class together. If we did not, we would have no reason to claim that others were responding to the same objects and events (i.e. causes ) that we are (1991a: 202).

The solution is this: we identify as the common causes of the responses of others to the environment what we find or notice to be the similarities in the environment and in the responses. That is, we use the categories with which we think about the world, the saliencies we see in the environment, and in the responses of others to it, in interpreting them. Is this a resource that Davidson can appeal to, however? It is not, if it is part of his project to show how to construct an account of what another thinks and means from the basic facts, the non-intentional and nonsemantic facts about a creature’s interactions with its environment. For the appeal to saliences we see or notice is an appeal to psychological facts available, in a strong sense. The second part of this depends, of course, on his claim that language is required for thought, which is in dispute. In addition, the claim itself is clearly hostage to there being some limits on acceptable interpretations, which is why Davidson is sensitive to the problem of the underdetermination of the objects of thought by the structure of our causal interactions with the environment. If anything goes, the appeal to indeterminacy to defend the thesis that the basic facts provided an adequate ground for attributing semantic and psychological facts looks entirely ad hoc: a desperate attempt to defend an untenable thesis by inoculating it in advance against any untoward results that investigation may turn up, no matter how absurd.

91

about us. If this is essential for interpreting another, for imposing enough constraints on the evidence so that it yields a reasonably unique answer to the question what another thinks and means by his utterances, then we must admit that the facts available from the third person standpoint alone do not suffice for interpretation. This would show that premise (1) of the argument is incorrect. For the facts about our own psychological states which we appeal to are not facts that we have learned about from the basic facts, and to appeal to them as a constraint in interpreting others is to admit that we could not in principle arrive at an account of those facts about ourselves from the basic facts. Thus the first person standpoint on our own thoughts turns out to be essential for adequate interpretation. This is not compatible, however, with Davidson’s basic stance on meaning and the propositional attitudes, which holds that the facts of the matter are exhaustively determined by the public, non-intentional, non-semantic facts, that is, the facts available from the standpoint of the radical interpreter, who presupposes, ultimately, nothing about what another thinks or means. Furthermore, to appeal to what we find salient in causes and responses is essentially to assume that the creature we are interpreting finds the same things salient. This is an empirical assumption, well grounded in the case of conspecifics, but not something constitutive of thought or meaning. To suppose otherwise would require an argument to show that thinkers and speakers as such both share with us the same concepts and share with us the same similarity space. There is no a priori reason to suppose this.17 If life can evolve in environments with significantly different physical characteristics than ours, where tracking, for example, electromagnetic radiation in the spectrum visible to us would not be useful, thinking creatures would not evolve to find the same saliences in their environment as we do. Even in our environment, it turns out that there are many creatures which 17

Davidson’s argument in “On the Very Idea of a Conceptual Scheme” (1974) assumes at a crucial point that another possesses concepts only if we can determine that to be so, and that we can do that only if we share a language with him (see Lepore and Ludwig 2005: 305-321). It rests, thus, on a kind of verificationist principle. There is a general, if not universal, consensus that such general verificationist principles cannot be established a priori. Whether such a principle restricted to matters connected with language can be established depends on the outcome of just such arguments as we are now considering. See the following remarks about premise (1) in this connection.

92

find different saliences (olfactory and auditory, for example) than we do. Even color saliencies differ significantly between species. Of course, there is a great deal of overlap in object saliences, but extensions of these sorts of differences in perceptual saliences (and even more radical ones, such as the ability to detect and track changes in magnetic fields) together with differences in what is important in an environment for survival could prima facie lead to very different object saliences as well. It is very difficult to see how an a priori limit could be put on what things in its environment could be salient to a creature. In short, once we have made essential appeal to what we find salient as a constraint on interpretation, we have abandoned the project of showing how the basic facts alone can ground a theory of interpretation, we have given up on the claim that the mental facts can be shown to supervene conceptually on the non-mental facts, and we have adopted as a background assumption the empirical claim that those that we aim to interpret share with us a similarity space, and so share with us a psychology which we understand from our own case. I end with two remarks about premise (1). First, premise (1) claims that the fundamental ground for the attribution of propositional attitudes and meanings rests on the basic facts about a creature’s interaction with its environment. This rests in turn on the claim that the possibility of thought at all depends on the possibility of confirming from the third person standpoint an interpretation theory for another, and the impossibility of that absent seeing it as thinking about its environment. Taking the third person standpoint as basic is motivated in Davidson’s work by thinking of it as the basic standpoint from which to investigate language, on the grounds that language is by its nature a tool for enabling us to communicate with others. Its nature, then, requires that it be decipherable by others. For this, however, to ground the claim that the third person standpoint is basic for the investigation of thought as such, not just thought that is bound up with language, requires the assumption that language is necessary for thought. Part of the grounding for the present argument can be seen, then, as resting on the conclusion of Davidson’s earlier argument that the concept of belief is necessary for belief, that belief is necessary for thought, and that having the concept of belief requires having a language. Second, while premise (1) is a background assumption of the argument, it is clear that success in find-

93

ing an objective ground for identifying the objects of thought is essential if it is to be maintained (see note 16 in connection with this). In light of this, the case against (6) counts against (1) as well. 4. Summary and Conclusion The idea of triangulation enters into Davidson work in two different connections. The first is in connection with Davidson’s argument in “Rational Animals” that language is necessary for thought. Trigonometric triangulation is used as an analogy to illustrate the thesis that only in the context of communication can a creature have the concept of belief. Together with the assumption that having the concept of belief is necessary for belief, and that belief is necessary for any thought, this yields the conclusion that language is necessary for thought. The argument depends crucially on the claim that only where there is scope for the correct application of a concept does it make sense to attribute it to a creature, and that only in the context of communication is there scope for the correct application of the concept of belief. We found the first assumption to be doubtful both because it is unclear that concept possession requires concept deployment and because it is unclear that if it does, it requires objective rather than subjective deployment. But even if we weaken the assumption, the argument for the necessity of language for thought, if not of communication, will go through accepting the rest of the argument. This shows the assumption that only in the context of communication is there scope for the application of the concept to be the central assumption. However, it is far from clear that it is only in the context of communication that one can find scope for the application of the concept of belief, with its attendant distinction between truth and error. Prima facie the possibility of the two contrasting perspectives needed to make sense of error can be provided either by a single individual’s reflections on the same thing at different times or on reflection on another creature’s thoughts as revealed by its non-linguistic behavior (as is prima facie possible) in contrast to what one believes oneself. The second is in connection with the problem of the underdetermination of thought content by a creature’s objective causal relations with its environment. Considering a creature by itself interacting with its environ-

94

ment, there are two intractable problems in trying to make sense of what it is thinking about, according to Davidson. The first is that even taking into account various occasions on which it is responding to its environment, there will always be multiple common causes of its reactions. The second is that to identify a common cause, one must also identify a common reaction, and there will be many different ways of deciding what counts as a common response. Davidson argues that the only solution is to see what determines the objects of thoughts about the environment as emerging from two speakers interacting with each other and objects in the environment which cause common responses in them (as illustrated in diagram 1). The central problem is that the addition of another creature does not suffice to solve the difficulties raised in the case of a single creature interacting with its environment. Wherever there is a common cause of a common response, there will be many; and a common type of cause will typically have common types of causes in its etiology. This is particularly evident if we put no constraints on the categories under which we type causes, a point Davidson makes in connection with the difficulty for a creature considered alone. This last difficulty arises also for identifying a common response to a common cause. There is nothing in the basic facts to tell us what are the relevant similarities across different occasions. The appeal to our categories solves both problems, but at the cost of undermining Davidson’s goal of showing that there are adequate grounds from the third person standpoint to determine an adequate interpretation theory for another. The difficulties that arise rather suggest that there is not, and that it is therefore a mistake to attempt to understand psychological and semantic categories from a purely third person standpoint. References Davidson, D. 1974, “On the Very Idea of a Conceptual Scheme”, Proceedings and Addresses of the American Philosophical Association, 47, 5-20; repr. in Davidson 1984 (from which I quote). Davidson, D. 1979, “The Inscrutability of Reference”, The Southwest Journal of Philosophy, 10, 7-20; repr. in Davidson 1984 (from which I quote).

95

Davidson, D. 1982, “Rational Animals”, Dialectica, 36, 317-328; repr. in Davidson 2001 (from which I quote). Davidson, D. 1989, “What is present to the mind?”, in J. Brand and W.L. Gombocz (eds.), The mind of Donald Davidson. Amsterdam: Rodopi; repr. in Davidson 2001 (from which I quote). Davidson, D. 1984, Inquiries into Truth and Interpretation. Oxford: Clarendon Press. Davidson, D. 1991a, “Epistemology Externalized”, Dialectica, 45 (2-3), 191202; repr. in Davidson 2001 (from which I quote). Davidson, D. 1991b, “Three Varieties of Knowledge”, Philosophy, 30 (Suppl.), 153-166; repr. in Davidson 2001 (from which I quote). Davidson, D. 1992, “The Second Person”, Midwest Studies in Philosophy, 17, 255-267; repr. in Davidson 2001 (from which I quote). Davidson, D. 1997, “Indeterminism and Antirealism”. in C.B. Kulp (ed.), Realism/Antirealism and Epistemology. Lanham, MD: Rowman and Littlefield; repr. in Davidson 2001 (from which I quote). Davidson, D. 2001, Subjective, Intersubjective, Objective. Oxford: Clarendon Press. Lepore, E. and Ludwig, K. 2005, Donald Davidson: Meaning, Truth, Language, and Reality. Oxford: Clarendon Press. Malcolm, N. 1973, “Thoughtless Brutes”, Proceedings of the American Philosophical Association, 46, 5-20. Quine, W.V.O. 1960, Word and Object. Cambridge, Mass.: MIT Press. Strawson, P.F. 1959, Individuals: An Essay in Descriptive Metaphysics. London: Methuen. Strawson, P.F. 1966, The Bounds of Sense. London: Methuen. Stroud, B. 1968. “Transcendental Arguments”, Journal of Philosophy, 65 (9), 241-56.

TRIANGULATION AND OBJECTIVITY: SQUARING THE CIRCLE? Adina L. ROSKIES (Dartmouth College, Hanover, USA)

Donald Davidson famously argues that language is necessary for thought, and that to have propositional attitudes (which he takes to be mental states with determinate content) one must be the speaker of a language. Triangulation, an interaction between two (or more) creatures with each other and the external world, is meant to illustrate the dependence of thought on language. Davidson claims triangulation is essential to the determination of mental content, and to the development of a concept of objective truth, itself a necessary condition for propositional attitudes. The role claimed for triangulation in grounding thought is considerable. One is justified in asking whether triangulation is as important to thought as Davidson seems to imply. I will argue that although triangulation is an evocative and compelling metaphor, it neither depends upon the participation of other linguistic agents as Davidson claims, nor can it possibly do the conceptual work Davidson requires of it. In recognizing the failure of triangulation to ground fundamental aspects of mental states, we can begin to break the bonds that seem to render thought dependent on language. 1. Triangulation and Language Triangulation involves directed action by two (or more) creatures toward an object in a common world. Moreover, it involves recognition that there is directed action of this sort: to have the concept of a table or a bell is to recognize the existence of a triangle, one apex of which is oneself, the second apex another creature similar to oneself, and the third an object (table or bell) located in a space thus made common (Davidson 1992: 121).

98

In order for the mutual recognition of the existence of such a triangle, Davidson says, we need to be in communication with the other creature, via language: The only way of knowing that the 2nd apex of the triangle – the second creature or person – is reacting to the same object as oneself is to know that the other person has the same object in mind. But then the second person must also know that the first person constitutes an apex of the same triangle another apex of which the second person occupies. For two people to know of each other that they are so related, that their thoughts are so related, requires that they be in communication. Each of them must speak to the other and be understood by the other (Davidson 1992: 121).

With a causal theory of language, there are chains of causes, each of which is a potential referent of a term. As Quine so effectively argues, reference appears to be underdetermined. Davidson sketches in some detail how triangulation serves to pick out the necessary relation between language and the world that determines content. Namely, triangulation trades on the innate similarity spaces we have, and our tendency to generalize in accordance with these. He provides an example of content fixation in a language learning situation. His example is one of a child learning the word for table: Involved in our picture there are now not two but three similarity patterns. The child finds tables similar; we find tables similar; and we find the child’s responses in the presence of tables similar. It now makes sense for us to call the responses of the child responses to tables. Given these three patterns of response we can assign a location to the stimuli that elicit the child’s responses […] Where the lines from child to table and us to table converge, ‘the’ stimulus is located. Given our view of child and world, we can pick out ‘the’ cause of the child’s responses. It is the common cause of our response and the child’s response (Davidson 1992: 119).

Although this does not do away with Quinean indeterminacy, it does appear to enable us to pick out a unique spatiotemporal location as the referent of our and the child’s thought. But we must ask whether Davidson is correct about the role of language in the triangulation relation. When ex-

99

amined more closely, language seems inessential to this function. Consider ONE-BOX: ONE-BOX: substitute for the child a grey box with a red light that blinks reliably when placed within one meter of a table. Again, we have three similarity patterns. We find tables similar, the box “finds” tables similar (i.e. responds in a similar fashion in the presence of tables), and we find the box’s responses in the presence of tables similar. In this case, we can again assign a location to the stimulus that leads to the box’s response. The temptation is to assign a content to the flashing, namely that the box is responding to tables. This seems to suggest that in order to fix content we require neither a second person, nor language, nor, even more to the point, a communicative intention.1 That is perhaps a little too fast. Davidson would of course object that we fall short of fixing content in this case; there is no determinate meaning to be assigned to the flashes, just a causal explanation offered for the regularities. Let us keep this in mind and pursue the operationalization of the triangulation scenario further. Consider 2-BOX: 2-BOX: Keep the first box and replace the 2nd agent (ourselves) with another box that flashes blue in response to tables, and green in response to other boxes when they flash red. 2-BOX involves two “agents” responding to the same similarity space, and to other agents doing likewise. However, a crucial connection between the patterns of signaling has not been made: there is no connection between the 2nd box’s signal and its “perception” of the first box. Perhaps in the case of 2-BOX we would again agree with Davidson that there is no determinate meaning to be assigned to the flashes, just a causal explanation offered for the regularities.

1

To be sure, there are additional theoretical reasons to think that the similarity spaces of two members of the same species are likely to be more similar than the similarity spaces of wildly different creatures, but those considerations have nothing to do with the importance of language to thought.

100

2-BOX*: Replace the 2nd box with another box that flashes blue in response to tables, green in response to other boxes when they flash red, and yellow when both its blue and green lights flash simultaneously. In 2-BOX* might we interpret the yellow light as a recognition of the triangulation relation? If so, it seems that most of Davidson’s requirements for triangulation have been met, but they have been met without language: We have two creatures that interact with a common object; each reliably indicates the presence of the object and there is a “recognition” that there are responses to a common object. We may however, still deny that there is thought here at all. Certainly Davidson would. The intuition that there is still no thought here might seem to ratify his view that there is no thought without language. To determine whether this is the case, however, we must first diagnose what is missing from this account. 2. Triangulation and Objectivity One thing that is noticeably absent from this scenario is any explicit sense in which the representations of the box-“creatures” are objective, or of an objective world. It is perhaps this that underwrites the intuition that there is no thought in the BOX cases. This is not lost on Davidson: having the concept of objective truth is essential, according to Davidson, for having propositional attitudes at all: Belief, intention, and the other propositional attitudes are all social in that they are states a creature cannot be in without having the concept of intersubjective truth, and this is a concept one cannot have without sharing, and knowing that one shares, a world, and a way of thinking about the world, with someone else (Davidson 1992: 121).

So the recognition that one shares an objective world with another is necessary for having a concept of objective truth. The power of triangulation in Davidson’s philosophy becomes evident, for he argues that the notion of objectivity is itself a result of triangulation: Our sense of objectivity is the consequence of another sort of triangulation, one that requires two creatures. Each interacts with an object, but what gives each the concept of the way things are objectively is the base line

101

formed between the creatures by language. The fact that they share a concept of truth alone makes sense of the claim that they have beliefs, that they are able to assign objects a place in the public world (Davidson 1982: 105).

Thus, not only does triangulation fix content, but it provides creatures with a means for constructing a concept of objectivity, which itself is required to endow that content with the requisite objectivity to make it count as representational thought, rather than mere stimulus response. Can triangulation accomplish both jobs at once, and if so, does it require language? A concept of objectivity is supposed to arise from triangulation, but this is surely wrong: objectivity cannot be the consequence of triangulation. For imagine what a supposed triangulation event would be like for a creature that lived in an objective world with other creatures, but lacked a notion objectivity: that creature’s experience would be as of a “world” of subjective objects (a series of coherent sense impressions). It would include sense-impressions of other creature-like things which would interact with the object-like things in regular ways; a box might represent other boxes with lights flashing contingent upon other object-like things in the sense-impressions. The sense-impressions of other box-creatures might even involve them making invariant sounds (i.e. linguistic utterances) in the presence of certain object-like things. In short, the entire world, the other creatures, and their communicative attempts would merely be taken as more subjective states. Triangulation offers no way to break out of the subjective world to achieve a notion of objectivity. Thus, in order for triangulation to get off the ground, an agent must first see others as part of the external world as opposed to an element in its mentality. As long as the other is merely a part of the way it takes things to be, it cannot fulfill the role of the second person. Triangulation cannot confer objectivity. This point is seldom noted. For example, in her defense of Davidson’s arguments about the importance of language for thought, Verheggen affirms that triangulation is required for objectivity: «if I am right, the solitary creature is in no position to do any of this precisely because it cannot have the concept of objectivity» (Verheggen 2007: 102). She goes on to say that the burden of proof is on those who claim that the solitary agent could have the concept of objectivity without triangulation. But the burden

102

is just the opposite: triangulation is not possible without presupposing a concept of objectivity. Thus, there are two fundamental problems with using triangulation as an argument for language being necessary for thought. First, there is nothing apparent about triangulation that requires spoken language as opposed to some other sort of joint interaction or nonlinguistic signaling. It is, indeed, difficult to see why anyone holds language as opposed to action operative in fixing content or developing a notion of a world external to ourselves. So triangulation fails to show that language is necessary for thought. Secondly, triangulation cannot provide the logical ground for the construction of a concept of objectivity or truth. In order for triangulation to get off the ground, an agent must already have a concept of objectivity. As long as the other is merely a part of the way an agent takes things to be, it cannot fulfill the role of the second person. So triangulation fails as a mechanism for constructing the concept of objectivity. To think otherwise is to try to square the circle. For triangulation to perform its function of helping to fix intersubjective content, the parties involved in the triangle must already have a notion that the world is objective, or mindindependent. And once there is that understanding, then it is not clear that a second person, or language, is needed to fix mental content at all. References Davidson, D. 1982, “Rational Animals”, Dialectica, 36, 317-328; repr. in Davidson 2001 (from which I quote). Davidson, D. 1992, “The Second Person”, Midwest Studies in Philosophy, 17, 255-267; repr. in Davidson 2001 (from which I quote). Davidson, D. 2001, Subjective, Intersubjective, Objective. Oxford: Clarendon Press. Verheggen, C. 2007, “Triangulating with Davidson”, The Philosophical Quarterly, 57 (226), 96-103.

PART II COMMUNICATION AND ENVIRONMENT

KNOWING ME, KNOWING YOU: TRIANGULATION AND ITS DISCONTENTS Fredrik STJERNBERG (University of Linköping, Sweden)

Davidson’s triangulation arguments present a bewildering terrain. There are several different conclusions of on the face of it similar arguments. Sometimes the arguments proposed seem like little more than gestures in a general direction. At times, Davidson himself refers to what is going on as “arguments” (i.e., with scare quotes): «This concludes the first part of my ‘argument’» (Davidson 1982: 104). There is also a curious mix of necessity claims being presented. The result of the argument is that a second speaker is necessary—in some sense of “necessary”—if a first speaker is to have thought at all, if that person is to be a thinker. And the claimed results of the arguments have struck many as unpersuasive or downright wrong. So there is little that is obvious here. My general attitude towards the argumentation concerning triangulation is that the stronger versions of the arguments fail, but that there is still something that is both non-obvious and worth holding on to. Some of the details of the arguments are worth looking more closely at, however. The results of Davidson’s triangulation arguments1 provide good material for lightbulb jokes. How many Davidsonians are needed to screw a lightbulb? Two—one to screw in the lightbulb, a second person to think that it is a lightbulb (I said the material was good, not that it made for good jokes). According to Davidson, at least two persons are needed, if one person is to have objective propositional thought. The second person is used and needed in making the first person a bona fide thinker. Without a second person, no first person can be aspiring to having thoughts with an objective content. This amounts to a kind of social externalism, holding that my possibility of having thoughts with content rests upon the existence of 1

The arguments are to be found scattered over several papers. Davidson (1982, 1992) are important, as well as the other papers collected in Davidson (2001a), especially section II, “Intersubjective”.

106

further persons.2 The parallels with, and perhaps inspiration from, Wittgenstein is clear, and acknowledged by Davidson: «Without a second person there is, as Wittgenstein powerfully suggests, no basis for a judgment that a reaction is wrong or, therefore, right» (Davidson 1997: 83). The claim about the need for a second person is not intended as an empirical claim. Davidson is not arguing that we in fact lean on other persons all the time in thinking about the world. Virtually no one denies that claim. It is also not the empirical claim that no human being, or something with a mind like that of human beings, would have developed a language without help from others, or even have been able to think much of any interest at all. There is ample evidence for this empirical claim as well, and few want to deny it. We should also not get too hung up on the appeal to a second person. The exact number of speakers is not that important for Davidson. Perhaps, as a matter of empirical fact, many more persons are needed. Davidson’s claims are stronger than empirical claims, and more philosophically controversial, perhaps even to the extent of being seen as obviously wrong by many. The claim is a necessity claim: further speakers are necessary for there to be thinkers at all. But this appeal to necessity still leaves ample room for many questions—what kind of necessity is intended here, and how is it shown that this holds?

2

There is, however, an important difference between the triangulation view and the social externalism of the Burgean type (Burge 1979, 2007.) The Burgean social externalist can very well allow that solitary, isolated, thinkers are possible. This kind of social externalist just says that for some concepts, the concepts we should be attributing to the single subject is dependent on the nature of the concept in that subject’s social surroundings. In certain surroundings, the single speaker should have been held to have thought that he had arthritis, in others, that he had some other, relevantly similar, disease. This kind of social externalist would say that given that the subject’s surroundings are of such and such a kind, then there will be some concepts that have to be determined and individuated by referring to the concepts used in the surrounding community. This kind of social externalist will then not be saying that other speakers are needed for a single speaker to acquire concepts at all, just that in view of the existence of communities and the phenomenon of incomplete understanding of concepts used for communication, we are forced at times to attribute concepts through referring to a concept being used in the surrounding community. Davidson doesn’t agree with this view; see for instance Davidson (2003).

107

1. A Brief Sketch of the Arguments Two early papers employing the triangulation arguments are Davidson (1982 and 1992).3 In the first, Davidson is presenting an argument that will show that communication is necessary for rationality and thought: «rationality is a social trait. Only communicators have it» (Davidson 1982: 105). There are two steps to this conclusion: First, I argue that in order to have a belief, it is necessary to have the concept of belief. Second, I argue that in order to have the concept of belief one must have language (Davidson 1982: 102).

If it then can be shown that language requires other speakers—that there is no private language, in one prominent sense of “private”, then we get the conclusion that we have to be in communication with others to be rational, to have thoughts. The triangulation arguments all rest on the application of Davidson’s well-known, and often used, principle of charity. The principle of charity states that someone has to be rational in order to have any thoughts at all. Rationality is not an extra attribute of the thinking creature, it is a requirement for there to be any thinking going on at all (see for instance Davidson 1970, 1974). Therefore, if a person has thought at all, she is rational. So we can put these intermediate steps together, and say that belief requires other believers, and other linguistic creatures. On the closing page of “Rational Animals” (1982), Davidson sums up his argument, and sets out two conclusions. One conclusion is that the concept of intersubjective truth suffices as a basis for belief and hence for thoughts generally. And perhaps it is plausible enough that having the concept of intersubjective truth depends on communication in the full linguistic sense (Davidson 1982: 105).

I think this statement of the conclusion is weaker than what Davidson intended. As it stands, we want to reply that of course if someone has the concept of intersubjective truth, then such a creature has beliefs. A concept 3

I have benefited greatly from Glüer (2006), which provides a useful overview and assessment of the strands in the argument.

108

of intersubjective truth is quite demanding. It would be obvious that having a concept of intersubjective truth is sufficient for having beliefs and having thoughts generally. The claim Davidson is usually arguing for is the more contentious claim that having a concept of intersubjective truth is necessary for having thoughts in general.4 Second-order belief is required for having first-order belief. Davidson has often argued that we need this stricter requirement in order to avoid having to say that just about anything has a concept. If we for instance tie the possession of concepts with an ability to discriminate two kinds of things from each other—things that fall under the concept and things that don’t fall under the concept, respectively—we find ourselves forced to admit that tomato-plants and sunflowers have concepts, since they track sunlight in appropriate ways.5 Such tracking is not sufficient for Davidson. He connects the use of concepts with the possibility of saying that mistakes are made, and that some creature can be surprised. Earthworms cannot be mistaken or surprised (Davidson 1995: 8-9). Empirical attempts have been made to surprise bees, and they have failed.6 No matter what a bee does, its reactions should not be classified as reactions of surprise. Perhaps we shouldn’t have bothered with trying to surprise bees; it should have been clear to us all along that nothing the bees would get up to would indicate surprise. We will return to the issue of surprise below. The second conclusion Davidson makes in the quotation above is presented in a weaker way here; it is just said to be “plausible enough” that we need language for the concept of the intersubjective to take hold. In “Rational Animals” (1982), he confesses that he doesn’t have a knock-down

4

This is for instance argued in Davidson (1995: 8-9, 2001a: 138-140). Hurford has noted that at least some biologists actually are willing to say things like “trees think”. One biologist is reported as saying «excluding plants from having concepts was perhaps too precipitate» (Hurford 2007: 287). It seems that we can safely leave that use of terms like “concept” to biologists. 6 Arrangements were made to present the bees with pieces of information they already knew were misleading. Another bee would come back to the hive, and present information—with the bee dance—indicating that there was food at a place the bees should know was without food (in the middle of a nearby lake). In such situations, the bees just didn’t do anything at all. 5

109

argument for this claim, but sketches a way to make the claim more reasonable. This sketch is the first installation of the triangulation argument: To complete the “argument”, however, I need to show that the only way to show that one could come to have the belief-truth contrast is through having the concept of intersubjective truth. I confess I do not know how to show this. But neither do I have any idea how else one could arrive at the concept of an objective truth. In place of an argument for the first step, I offer the following analogy. If I were bolted to the earth, I would have no way of determining the distance from me of many objects. I would only know they were on some line from me towards them. I might interact successfully with objects, but I could have no way of giving content to the question where they were. Not being bolted down, I am free to triangulate. Our sense of objectivity is the consequence of another sort of triangulation, one that requires two creatures. Each interacts with an object, but what gives each the concept of the way things are objectively is the base line formed between the creatures by language. The fact that they share a concept of truth alone makes sense of the claim that they have beliefs, that they are able to assign objects a place in the public world (Davidson 1982: 105).

So what we have first is not an argument, but an analogy. The benefits of triangulation in the proper sense are to be had by the benefits of communication, and in this way, we can get a grip on objective truth. Objective thought will thus require intersubjectivity, as a later passage indicates: The ultimate source (not ground) of objectivity is, in my opinion, intersubjectivity. If we were not in communication with others, there would be nothing on which to base the idea of being wrong, or, therefore, of being right, either in what we say or in what we think. The possibility of thought as well as of communication depends, in my view, on the fact that two or more creatures are responding, more or less simultaneously, to input from a shared world, and from each other. We are apt to say that someone responds in “the same way” to, say, wolves. But of course, “same” here means “similar”. This prompts the next question: what makes the reactions similar? The only answer is, someone else finds both wolves and the reactions of the first person similar. This of course only put the basic question off once more. Nevertheless, it is this triangular nexus of causal relations involving the reactions of two (or more) creatures to each other and to shared stimuli in the

110

world that supplies the conditions necessary for the concept of truth to have application. Without a second person there is, as Wittgenstein powerfully suggests, no basis for a judgement that a reaction is wrong or, therefore, right (Davidson 1997: 83).

Without a grip on objectivity, no thinking, and without other speakers, no grip on objectivity. Hence we are forced to triangulate with other speakers, if thinking is to be taking place: «[T]riangulation is a necessary condition of thought and language […] social interaction, of which triangulation is the most basic form, is necessary to the success of ostension» (Davidson 2003: 694). This is then the first conclusion of the triangulation considerations. This is what Glüer calls “the argument from objectivity”. This is clearly Wittgensteinian in spirit, and Davidson has acknowledged the parallels. There is also another strand we can find in the triangulation arguments. This is the idea that there is no determining the contents of a single speaker’s mental states, of her thoughts, without triangulation, without bringing in a second speaker. Glüer calls this “the argument from content determination” (Glüer 2006). This argument is more Quinean, and more specific, perhaps also resting on a more literal understanding of “triangulation”. This proceeds from the claim that, when determining content, we at least in the simplest cases have to look at what it is that causes the subject’s beliefs. It is then held that this process cannot be accomplished without a second speaker. Without a second speaker, we get too many potential causes for the belief. Is the subject reacting to the object in the surroundings, or to the proximal irritation of her nerves? It is argued that the only way to settle this is by bringing in a second person: Social interaction, triangulation, also gives us the only account of how experience gives a specific content to our thoughts. Without other people with whom to share responses to a mutual environment, there is no answer to the question what it is in the world to which we are responding. The reason has to do with the ambiguity of the concept of cause. It is essential to resolve these ambiguities, since it is, in the simplest cases, what causes a belief that gives it its content. In the present case, the cause is double indeterminate: with respect to width, and with respect to distance. The first ambiguity concerns how much of the total cause of a belief is relevant to content. The

111

brief answer is that it is the part or aspect of the total cause that typically causes relevantly similar responses. What makes the responses relevantly similar in turn is the fact that others find those responses similar; once more it is the social sharing of reactions that makes the objectivity of content available. The second problem has to do with the ambiguity of the relevant stimulus, whether it is proximal (at the skin, say) or distal. What makes the distal stimulus the relevant determiner of content is again the social character; it is the cause that is shared. The stimulus is thus triangulated; it is where causes converge (Davidson 1997: 129f).

Focus points to determine content are the causes of the content (again, as Davidson stresses, in the simplest cases, but we will go along with this). Triangulation as a social process is needed, because the single subject faces too many potential causes. It is only in a social setting that we can pare down the number of potential causes to numbers we can handle. Even if we have two speakers, triangulating a common cause, there will be a number of potential causes, but now they have this set of possible causes in common. Given a common setting, with a second person triangulating my thinking, it can be determined whether I am thinking about the nearby rabbit, or whether the content is the rabbit-shaped stimulation of my retina. Without such common focuses, no determination of content. And this is not just a question of a difficulty to determine the content—it is held that in such cases, there will be no content to determine. Here Davidson is making more or less tacit use of the first strand in the triangulation argument. Davidson (1992) makes the argument that the single subject is not really a thinker at all, without the process of triangulation: [T]riangulation […] is necessary if there is to be any answer at all to the question what [a creature’s] concepts are concepts of. If we consider a single creature by itself, its responses, no matter how complex, cannot show that it is reacting to, or thinking about, events a certain distance away rather than, say, on its skin. The solipsist’s world can be any size; which is to say, from the solipsist’s point of view it has no size, it is not a world. […] The problem is not, I should stress, one of verifying what objects or events a creature is responding to; the point is that without a second creature responding to the first, there can be no answer to the question. […] before anyone can speak a language, there must be another creature interacting with the speaker (Davidson 1992: 119f; cf. Davidson 2000: 1059f).

112

So, according to the triangulation reasoning, a second person is needed to make a single subject’s activities—“no matter how complex”—into acts of thinking, language use, or into the exercise of rationality. 2. Some Problems It should be no surprise that these Davidsonian claims have been hotly contested and debated ever since they were first presented. I will not rehearse all the various types of argument presented, but concentrate on a few points of interest, and then move on to a third strand in the triangulation considerations. Concerning the first strand—the argument from objectivity of thought, Davidson’s handling of animals is unsatisfying. This strand of the argument has it that single-creature objective thought is impossible, because having concepts—necessary for being called a thinker—requires that we make room for saying that the creature is mistaken or surprised. And nothing can be the ground for saying this, unless we have reason to say that it is a language-user. This excludes animals, since they lack a language. It is hard to know exactly how we should view the minds of languageless animals, but Davidson is here not sufficiently sensitive to the issues. In earlier works, Davidson tended to see the options in handling animal cognition in terms of a stark dichotomy: either we must say that they think in the way we think, or else our use of the terminology of thinking just masks that it is a useful way to handle their activities, useful because we don’t have all the required information to describe their cognition in the way we describe and explain the “cognition” if heat-seeking missiles. It is useful for us to characterize the missile’s activities as “wanting” to find the enemy airplane, but this just masks our meagre knowledge of what is really going on here (Davidson 1982: 101). This is surely a false dilemma. The research problems of animal cognition are real enough, but nothing is gained by seeing the options as constrained in this way. There is much research in animal cognition, showing us new things about what animals are capable of. Davidson is not denying that animals are capable of many things, but he is contesting how we should characterize their abilities. Recent research in animal cognition often employs the very criterion which Davidson suggested for having concepts; the ability to be surprised,

113

or to be mistaken.7 It turns out that at least some primates actually do show that they can be surprised at times. Baboons can for instance display surprise—reacting in a way that shows that something is going against their expectations—and this surprise can be elicited by information from other baboons presenting information that goes against what they think that baboon knows or thinks. The baboons react differently to surprising information—information that the baboons had reason to believe was false. If a baboon in the group in some way indicates that something unexpected has happened, such as new information about ranking in the hierarchy, or information that baboons A and B are feuding, or that they have made up, while no other baboons think that there was a feud, or that it is too early for the feuding baboons to make up, they react differently, when compared with what they do when they receive information that was expected: they look at the baboon bearing the surprising news, but they don’t act on it.8 Cheney and Seyfarth (2007: ch. 11) want to describe baboon cognition as sufficiently like the cognition of us linguistic creatures, holding that their cognition employs a language of thought. This is probably overstating the existing evidence somewhat. It is perhaps better to say as Camp, that the baboons «employ a representational format that differs importantly from language» (2009: 113). Other research into primate cognition, notably chimpanzees, stresses that the question whether for instance chimpanzees have a theory of mind is not readily settled as a yes-no question. We might want to say that they don’t have a theory of mind in the full sense of a human theory of mind, but they seem to have a conception of other chimpanzees being up to things, understanding their surroundings in suitably, though at times misguided, ways (Call and Tomasello 2008). These questions are by no means settled, but recent research at least shows that Davidson’s way of putting the differences between language-using humans and languageless animals is much too crude.

7

We should here of course distinguish between being startled and being surprised in the sense of something that goes against one’s expectations. Virtually all animals can be startled—shocked into action by something novel. 8 These violation of expectation tests were first made on human infants, and are now being carried out on several animals. See Cheney and Seyfarth (2007: 92ff).

114

In his (2003), Davidson acknowledges some of these objections, and retreats somewhat. He says that there may after all be good reasons to use many of the terms connected with cognition for characterizing animal activity—so use of cognition-related terms is not just a way to get around our lack of the mechanical workings of animals. But he still wants to reserve some terms for us language-using humans: We know what [animals] can do, and that certainly includes all sorts of things right-thinking, or right-talking people, call thinking. I’m caving in on the words ‘think’, ‘represent’, ‘cognition’, and so on. I hold out on propositional contents […] for the data do not seem to me to compel us to say that only propositional thought can explain those data (Davidson 2003: 697).

This is a significant weakening of the original thesis, which held that triangulation was necessary for thought and rationality. It could also be seen as almost trivializing the original claim—perhaps language is trivially needed precisely for propositional attitudes, given a suitable definition of “language” and of “propositional attitudes”. Related to these observation from studies of animal cognition, it appears that the triangulation argument makes little sense from an evolutionary perspective. People writing about the evolution of language have gravitated towards seeing things the other way round, i.e. seeing the linguistic ability as the result of an ability to interact with other creatures in the group, and then specifically using their social intelligence, their ability to find out and understand what is going on with other members of the group. The ability to use a language is connected with an ability for recursion, the ability to construct an infinite number of new sentences by putting the familiar building blocks together in new ways. The use of grammar, and the comprehension of new sentences, relies upon this. Many of the other abilities of human beings have clear parallels in animal cognition, but it appears that there is nothing like a recursive ability to be found in the animal kingdom (Fitch, Hauser and Chomsky 2005. See also Hurford 2007, and Pinker and Jackendoff 2005). How and when this recursive ability developed is still something of a mystery, but one hypothesis is that the recursive ability is an offspring from a development in the theory of mind ability. The idea behind this hypothesis is roughly as follows. In complicated social settings (as those we

115

know that our ancestors lived and evolved in), mindreading would be a good thing. But such an ability to read other members’ minds would be even better, even more advantageous, if it could take into consideration what the other thinks about me, and what the other thinks that I think about her, and so on. The opportunities for deception and for correct understanding of what is going on are vastly enhanced. So a kind of recursion would here be a good thing, evolutionarily speaking. This is one aspect of the socalled social intelligence hypothesis. It is by no means the sole contender in this area, but it has some empirical backing, and is definitely a serious contender in the attempts to explain how we came to end up being language-users, a linguistic and symbolic species. This hypothesis would then indicate that it puts the cart before the horse to insist that our ability to use language is needed for us to have a theory of mind. It would still leave something for a non-trivial emphasis on triangulation, however: the claim that our knowledge of ourselves is inextricably interwoven with the ability to have some kind of knowledge about other speakers, an issue I return to below. Let us turn to he content-determination argument. It appears to be question-begging. Even if we would go along with saying that the contents of the single speaker’s mental states is underdetermined, solving this problem doesn’t require bringing in other speakers. Why couldn’t this problem be handled by looking at how the single speaker treats her surroundings over time? This will be sufficient to distinguish distal from proximal stimuli. We could determine the cause of the beliefs of the single subject by for instance checking how the subject moves her head over time—fixating on the object, or fixating on the appearance of the object, will lead to different activities on the part of the subject. There is ample evidence for this kind of triangulation in many animals, in their perceptual representation of the surroundings. The single subject would be reacting to the rabbits in the distance, and not to the rabbit-shaped proximal stimuli, and this could be judged by considering the single subject’s reactions over time. It is only if it is assumed that no single speaker on her own can have determinate mental contents that this argument works. So what is it that remains of the triangulation arguments? The objectivity argument cannot show the need for other language users, since there will be animals meeting criteria for objective thought (they pass the sur-

116

prise test). Here I am in general agreement with Burge’s assessment in Burge (2003).9 Burge is negative to the appeal to triangulation, doubting that there is any need for extra interpreters. Burge thinks that the claim that we need a conception of belief to have any beliefs at all is mistaken, and he thinks that there is ample empirical evidence to show that the triangulation problem has been solved already, in nature, by languageless organisms. These objections are mainly objections against the triangulation argument from objective thought. Objective thought is possible without triangulation. The argument from content determination will not in itself work to show that other speakers are necessary—it would be possible for single subjects to meet the requirements by reacting differentially to stimuli over time. This option is at least available if we don’t from the start assume that a single subject is powerless to have objective thought.10 So these two strands in the argument don’t really work as Davidson intended. But there is another prominent theme that Davidson stresses in papers where the triangulation arguments are presented, and that is the interrelatedness of our different ways of knowing. This strand in the argument strikes me as more promising. The next section develops this claim. 3. A Third Strand in the Arguments In the two first strands of the triangulation arguments, Davidson argued that a second person is needed to make it the case that the first subject is having objective thought. In “Three varieties of knowledge” (1991), there is another argument. The three kinds of knowledge mentioned in the title of the paper are knowledge about the world, knowledge about one’s own mental states, and knowledge about the mental states of other speakers. The main claim in that paper is that there is no privileged type of knowledge—a subject must have all three types or none. The other types of 9

A more extended case, with further examples from animal cognition studies, is made in Burge (2009). 10 These are just two issues I want to highlight in the triangulation arguments. Glüer (2006) and Pagin (2001) give a useful overview of several other problems with the arguments. Amoretti (2008) is also useful.

117

knowledge cannot be reduced to one or the other. This argument is in some tension with the other strands mentioned above: «There are compelling reasons for accepting the view that none of the three forms of knowledge is reducible to one or both of the others» (Davidson 1991: 206). This quotation only gives the first part of the claim—the non-reducibility claim. Davidson also argues that we cannot have one form of knowledge without the other: If I did not know what others think, I would have no thoughts of my own and so would not know what I think. If I did not know what I think, I would lack the ability to gauge the thoughts of others. Gauging the thoughts of others requires that I live in the same world with them, sharing many reactions to its major features, including its values. So there is no danger that in viewing the world objectively we will lose touch with ourselves. The three sorts of knowledge form a tripod: if any leg were lost, no part would stand (Davidson 1991: 219f).

This seems to me to be the right thing to say. This conception of how our varieties of knowledge are interrelated is in conflict with the two earlier strands discussed above. In these strands, intersubjectivity was needed for subjects to arrive at thinking in general. This is not in itself a reducibility claim, but it does suggest that subjects can have intersubjective awareness without having the other kinds of knowledge. And this seems quite wrong—how could I know that the other person is reacting to rabbits in the vicinity, unless I know that I am reacting to rabbits in the vicinity, and that it is rabbits that I am reacting to? This question shows how the three kinds of knowledge would be interacting in a concrete interpretational situation. Now we find that subjects cannot have one type of knowledge without having the others, and as I said, this seems to be precisely how it should be. But the earlier strands in the triangulation arguments seemed to say something else. Perhaps Davidson had in mind some weak conception of intersubjectivity—hence my talk of “intersubjective awareness”—meaning that different creatures react in similar ways to similar stimuli in common surroundings, and that these common differential reactions are what is needed for triangulation. But if that was what Davidson intended when talking of triangulation, this cannot give us concepts in the sense Davidson wants and needs. Fish in a school of fish can all react in the same way to

118

an intruder, and some of the fish may react because they perceive that their fellow fish are reacting in such and such a way, but this does not mean that they will have a concept of scuba diver. So it seems that Davidson cannot have it both ways here. If all forms of knowledge stand or fall together, then rationality cannot be grounded on intersubjectivity. If something less than intersubjective knowledge—i.e. some form of common intersubjective awareness of common surroundings—is meant instead, then what we get will not give us concepts. Then we would still not be better off than the earthworms. But the weaker conclusion of the triangulation arguments is still not trivial or uninteresting. Even if second persons don’t have to be invoked to ensure objectivity of thought or guarantee mental content, the emphasis on the interwoven nature of knowledge will locate knowledge where it should be located, not in a presumed Cartesian theatre of infallible knowledge, not in a purportedly objective world out there. I cannot know fully what I believe, unless I have some grasp of what it is that others are up to, and I cannot be said to really know what concepts I have unless I know that my concept will be graspable by others, or that it will be conforming with something that is out there in the world. Therefore, knowing me is knowing you, and both are tied in with knowing the world. So the words of ABBA, making up the title of this paper, apply: Knowing me, knowing you, It’s the best I can do

This much of triangulation will remain, even if the more ambitious philosophical claims and arguments concerning triangulation fail. Basically, Davidson got things right here, even if the strong arguments from triangulation don’t work as intended. References Amoretti, M.C. 2008, “Davidson, Self-Knowledge, and Skepticism”, in Amoretti and Vasallo 2008.

119

Amoretti, M.C. and Vassallo N. 2008 (eds.), Knowledge, Language, and Interpretation. On the Philosophy of Donald Davidson. Frankfurt: Ontos Verlag. Burge, T. 1979, “Individualism and the mental”, repr. in Burge 2007. Burge, T. 2003, “Social Anti-Individualism, Objective Reference”, Philosophy and Phenomenological Research, 67 (3), 682-690. Burge, T. 2007, Foundations of Mind. Oxford: Clarendon Press. Burge, T. 2009, “Perceptual Objectivity”, Philosophical Review, 118 (3), 285324. Call, J. and Tomasello M. 2008, “Does the Chimpanzee Have a Theory of Mind? 30 years later”, Trends in Cognitive Sciences, 12 (5), 187-192. Camp, E. 2009, “A Language of Baboon Thought?”, in R.W. Lurz (ed.), The Philosophy of Animal Minds. Cambridge: Cambridge University Press. Cheney, D. and Seyfarth R. 2007, Baboon Metaphysics. The Evolution of a Social Mind. Chicago: University of Chicago Press. Davidson, D. 1970, “Mental Events”, repri. in Davidson 1980. Davidson, D. 1974, “Psychology as Philosophy”, repr. in Davidson 1980. Davidson, D. 1980, Essays on Actions and Events. Oxford: Clarendon Press. Davidson, D. 1982, “Rational Animals”, repr. in Davidson 2001a. Davidson, D. 1991, “Three Varieties of Knowledge”, repr. in Davidson 2001a. Davidson, D. 1995, “The problem of objectivity”, repr. in Davidson 2004. Davidson, D. 1997, “Indeterminism and antirealism”, repr. in Davidson 2001a. Davidson, D. 2000, “The Perils and Pleasures of Interpretation”, repr. in Lepore and Smith 2006. Davidson, D. 2001a, Subjective, Intersubjective, Objective. Oxford: Oxford University Press. Davidson, D. 2001b, “What thought requires”, repr. in Davidson 2004. Davidson, D. 2003, “Responses to Barry Stroud, John McDowell, and Tyler Burge”, Philosophy and Phenomenological Research, 67 (3), 691-699.

120

Davidson, D. 2004, Problems of Rationality. Oxford: Oxford University Press. Fitch, T., Hauser, M. and Chomsky N. 2005, “The Evolution of the Language Faculty: Clarifications and Implications”, Cognition, 97, 179-210. Glüer, K. 2006, “Triangulation”, in Lepore and Smith 2006. Hurford, J. 2007, The Origins of Meaning. Language in the light of evolution. Oxford: Oxford University Press. Lepore, E. and Smith, B.C. 2006 (eds.), The Oxford Handbook of Philosophy of Language. Oxford: Oxford University Press. McDowell, J. 2003, “Subjective, Intersubjective, Objective”, Philosophy and Phenomenological Research, 67 (3), 675-681. Pagin, P. 2001, “Semantic Triangulation”, in P. Kotatko, P. Pagin and G. Segal (eds.), Interpreting Davidson. Stanford: CSLI. Pinker, S. and Jackendoff R. 2005, “The Faculty of Language: What’s Special about It?”, Cognition, 95 (2), 201-236.

TRIANGULATION AND THE BEASTS Dorit BAR-ON (University of North Carolina, Chapel Hill, USA) Matthew PRISELAC (University of North Carolina, Chapel Hill, USA) 1. Introduction Philosophical debates about the mental life of non-human animals provide an especially vivid illustration of how radically philosophers’ intuitions concerning other minds can diverge. Do animals have mental states? Of what sort? Do any of the beasts have minds that overlap with ours? Is there any significant continuity between their minds and ours? Davidson is well known for arguing that, for conceptual reasons, at least when it comes to beliefs and other propositional attitudes, non-human animals differ from us in having none. For example, he has argued that having beliefs requires having the concept of belief, which in turn requires language (Davidson 2001a, 2001b). He has also argued that possession of any propositional attitude presupposes possession of belief, and that attributing any propositional attitude to a creature requires crediting them with the concepts that figure in the specification of the attitude’s content (Davidson 2001a, 2001b, 2004). Davidson’s arguments are tantalizing, but also puzzling, and far from explicit; and they have been subjected to sharp scrutiny and powerful objections from a number of authors in recent years. Animal mentality, as such, is not our direct concern in this paper. Thus, we will here set aside those of Davidson’s arguments that try to establish directly the dependence of propositional thought on language. Instead, we are interested in a theme that became prominent in Davidson’s more recent discussions of the subject, namely: triangulation. This is the idea that contentful thought about an objective world, as well as meaningful linguistic communication, require «the existence of a triangle, one apex of which is oneself, another a creature similar to oneself, and the third an object […] located in a space thus made common» (Davidson 2001a: 121). In Davidson’s hands, triangulation culminates in a specific kind of skepticism regarding the possibility of a philosophical account of the emergence

122

of propositional thought and linguistic communication among creatures like us, with a certain natural history, capacities, and setting. If Davidson is right, the intersubjective relations that are necessary for the very possibility of objective thought could never be sufficient for its emergence. To be sufficient, Davidson claims, the line connecting the two subjects at the base of his envisaged triangle would itself have to involve mutual linguistic interpretation. Since linguistic interpretation, in turn, presupposes objective thought, this means that there is no hope for an illuminating philosophical story about the emergence of objective thought in the natural world. Davidson takes this skepticism to be of a piece with his general anti-reductionism in the philosophy of mind and language. He finds an unbridgeable chasm separating the intentional, rational, rule- or normgoverned behavior of human agents as manifested in language use, on the one hand, from the merely responsive, passion-driven, pattern-governed, discriminatory behavior of the beasts, on the other. As we shall see, Davidson recognizes “pure” triangulation among the brutes, which he takes to be insufficient for objective thought, and he recognizes “reflective” triangulation among linguistic creatures like us, which already presupposes objective thought; but he finds no conceptual middle-ground between the two. We will argue that this sort of anti-reductionist “continuity skepticism” is more radical than the claim (mentioned at the outset) that non-linguistic animals cannot be said to have any propositional thought. Thus one might naively be more sanguine about the possibility of explaining the gradual emergence of objective thought than about claiming that non-linguistic animals are already endowed with propositional thought, or communicate in ways that already resemble human language in all significant respects.1

1

Davidson therefore aligns himself historically with a rationalist tradition including Aristotle, Descartes, Leibniz and Kant that, in its most extreme form treats animals as mere mechanical automata. With regard to more contemporary figures, Davidson’s view is akin to those of Sellars, Rosenberg, McDowell and Brandom for whom genuine thought requires operating within the linguistically structured “space of reasons”. Our attempt below to forge an intermediate form of triangulation between reflective, linguistic triangulation and pure triangulation, then, is an attempt to show how a Davidsonian view is reconcilable with views which locate thought in a continuous natural world.

123

In what follows, we will accept the claim that triangulation of some sort is indeed necessary for the emergence of language and propositional thought. We will also accept that triangulation “in its pure state” as found among non-linguistic beasts is not sufficient to ground the sort of objectivity required for (and exemplified by) propositional thought and linguistic communication. Our aim will be a fairly modest one: to give sense to the idea of an intermediate triangulation that can be interposed between the pure and linguistic triangulations Davidson contrasts so sharply. The possibility of such intermediate triangulation, we will suggest, can be appreciated by attending to a distinct form of behavior that we humans share with the beasts—namely, expressive behavior. In Section 2, we offer a sympathetic interpretation of Davidson’s appeal to triangulation; we briefly explain how we understand his nonreductive naturalism and articulate his continuity skepticism. In section 3, we present expressive behavior as a special sub-category of intersubjective communicative behavior, which is not readily describable in purely causal, non-intentional terms. In Section 4, we explain how the expressive behavior of non-linguistic animals gives rise to the possibility of intermediate triangulation sufficient to support at least a sort of proto-objectivity. By itself this possibility may not suffice to rebut the direct Davidsonian arguments against animal thought, or forestall wholesale other-minds skepticism. But we believe it can help undermine Davidson’s triangulation-based continuity skepticism. 2. Triangulation, Non-Reductive Naturalism and Davidson’s Continuity Skepticism The idea of triangulation is invoked by Davidson to help account for what he takes to be essential characteristics of thought. First, thought has objective content, and second, thought can have determinate empirical content. Our focus here will be on the role Davidson takes triangulation to play in providing a foundation for the objectivity of thought. Our thought is objective, according to Davidson, in the sense that «it has a content which is true or false independent (with rare exceptions) of the existence of the thought or the thinker» (Davidson 2001a: 130). The discriminative behavior prevalent in the animal world, Davidson thinks, is importantly different from

124

linguistic behavior, which is informed by genuine conceptualization. The former is behavior that is merely in accord with rules; the latter involves following rules. Merely acting in accord with a rule, or exhibiting distinctive reactions to things of a certain kind is not sufficient for genuine objectivity, whereas following a rule, or classifying objects as of a specific kind is. This is because only the latter sort of case allows for genuine error. (This point often comes also under the rubric of the normativity of thought.2) In addition, Davidson thinks that a subject capable of genuine error must herself be aware of the possibility of error. In other words, objectivity requires an awareness or grasp of objectivity. Not only, then, is thought objective, but «this is a fact of which a thinker must be aware; one cannot believe something, or doubt it, without knowing that what one believes or doubts may be either true or false and that one may be wrong» (Davidson 2001a: 130). This requirement of reflective awareness is not one that Davidson always pauses to defend. When he does, the defense typically takes the form of giving reasons for his well-known claim that to have a belief one must have the concept of belief.3 Notably, then, triangulation enters the picture not to support the claim that objectivity requires reflective awareness of objectivity, but rather to explain what could make it possible for a creature to be aware of or grasp objectivity, by showing what must be in place for the concept of error (or belief) to arise. 2.1. Triangulation as Necessary but not Sufficient for Thought Davidson’s triangulation obtains between (at least) two subjects (S1 and S2) and some object (O) in the world around them. These subjects interact with one another as well as with the world; there are various causal connections between S1 and O, S2 and O, and S1 and S2. In the pure (nonlinguistic) case of triangulation, each subject is somewhat sophisticated in the sense that they can “classify” and “generalize”. That is, «each of the creatures is carrying out its habitual inductions whether learned or inherited. This means that for each creature there are stimuli which can be classed together by virtue of the similarity of the responses» (Davidson 2

Davidson also frequently references Wittgenstein in support of this point (Davidson 2001a: 130, 2001c: 2-5). 3 See Davidson (2001a: 103-105, 2001b: 170-171).

125

2001c: 5). Of course, this “classification”, “generalization” and “induction” are not here the genuine articles, as they do not constitute judgments that deploy concepts; they amount to nothing more than similar responses to similar objects, which responses are, in turn, grounded in nothing more than bare discriminatory dispositions, even if acquired and modifiable. The final component completing the pure triangular setup is the subjects’ coming to associate each other’s responses to O with O in the following way. S1 may be said to associate S2’s reaction to O with O when S1 responds to S2’s O-reaction as S1 responds to O (and vice-versa for S2 associating S1’s O-reactions with O). Davidson’s claim is that this picture, and this picture alone, is capable of providing a conceptual foundation for the objectivity of thought (assuming commitment to the requirements of normativity and reflective awareness mentioned earlier). Once each subject has associated the other subject’s O-behavior with O, that is, once a subject has come to form expectations about O, or about the other subject’s behavior, on the basis of perceiving the other subject’s behavior, or O, a certain discrepancy becomes possible. When the other subject’s behavior and the presence of O fail to match, expectations go unfulfilled. And where expectations can be established and go unfulfilled, “space is created” (as Davidson puts it) for the concept of error to develop (see Davidson 2001d: 12). Now, whether Davidson is correct that only a triangular situation could open this space is a matter that we wish to set aside.4 What we want to emphasize here is that, even if it is correct that triangulation is necessary for objective thought, by Davidson’s lights, it is not sufficient. As Davidson sees things, triangulation as described so far allows us to see the general structure of situations in which the concepts of error, belief, truth, etc. could come to have application. Earlier we saw that for the concepts actually to apply, the subjects must have reflective grasp of those very concepts. Yet nothing in the intersubjective relations described so far, it seems, supports the attribution of such reflective grasp. From each subject’s point of view, the other subject’s behavior is simply part of the surrounding “passing show”—something that can be correlated (or not) with items in the world (objects, events, state of affairs), as smoke 4

For a critical assessment of this claim, see Pagin (2001) and Gluer (2006); for a defense, see Verheggen (2007).

126

is correlated with fire, or deer tracks with the nearby presence of deer. Nothing so far supports the idea, crucial to objectivity (as Davidson sees it) of the two subjects treating each other as subjects who have takes on the world, which take can fit or fail to fit with the way things are.5 O S1’s O-responses

S2’s O-responses

S1

S2 mediated O-responses  purely behavioral disagreement

Figure 1: Intersubjective object-directed relations in “pure” triangulation

But if triangulation as so far described is insufficient to account for objectivity, one may wonder, as Davidson himself does, what would suffice. Davidson’s answer is that nothing short of the presence of linguistic communication between the two subjects in the triangular setup will do. (Also see, Davidson 2001a: 105-106 and 131, 2001d: 13, 2004: 140-141.) As Davidson explains: [U]nless the base line of the triangle…is strengthened to the point where it can implement communication of propositional contents, there is no way the agents can make use of the triangular situation to form judgments about the world. Only when language is in place can creatures appreciate the concept of objective truth (Davidson 2001a: 131).

As Davidson is aware, to invoke linguistic communication in an account of the objectivity of thought is unhelpful, since linguistic communication is itself what paradigmatically exemplifies objectivity (Davidson 2001d: 13). Yet he thinks that any enhancement of the basic triangular situation short of linguistic communication will fail to suffice for objective thought. His ground for that claim, however, is not a dualist view about the nature of the mental. (Davidson, for example, does not doubt that a thinking machine 5

See Davidson (2001c: 11-13, 2001a: 121), and Eilan (2005: 9) for a helpful interpretation of this point.

127

could be constructed.6) Rather, his resistance is rooted in his commitment to the irreducibility of intentionality: The triangular arrangement is a necessary, but not sufficient condition, of thought. What must be added to make the conditions sufficient? If we could answer this question we would have an analysis of thought. It is hard to think what would satisfy us which did not amount to a reduction of the intentional to the extensional, and this, in my opinion, is not to be expected (Davidson 2001d: 13; emphases added).

Thus Davidson thinks that any addition to pure triangulation that would carry us closer to genuinely objective thought, short of intersubjective linguistic communication, would commit us, implausibly, to the possibility of a reduction of the intentionality of thought to extensional—i.e., purely causal—relations between thought and its objects.7 What is it that language adds so as to make the interrelations between the subjects and the worldly object sufficient for objectivity? When S1 and S2 are language speakers, capable of responding to objects with meaningful, true or false utterances, we can straightforwardly fund a difference between what is believed and what is the case. For Davidson, meaning is but one node of a(nother) triangle that has beliefs and holdings-true of sentences as its other nodes. (See Davidson 2001b: 142-155 and 161-166.) If you fix which sentences the subject holds-true and fix a subject’s beliefs, you will have thereby determined the meanings of sentences for that subject; and the same relationship holds for each node with respect to the other two. Intersubjective disagreements can then manifest themselves as differ-

6

«We are just machines that are complex in ways flies are not, so the problem isn’t one of transcending mere physical devices. I do not doubt that an artificer could, at least in principle, manufacture a thinking machine. The problem, for philosophy anyway, is what to aim for; what would show that the artificer had succeeded? [...] I think in this respect, Turing had the right idea, though his test was not conclusive for a variety of reasons» (Davidson 2004: 136). 7 Though this goes beyond our scope here, the implausibility of such a reduction in Davidson’s mind can no doubt be traced to—perhaps just is—his anomalous monism. See Davidson (2005: 183-220).

128

ences in the sentences held-true by the two subjects.8 On a given occasion, S1 may produce a sentence (say, “There’s an O nearby”) which S2 presumes S1 to hold-true, and yet which he (S2) takes to be false. Genuine objectivity is straightforwardly provided for via the possibility of each subject recognizing a potential gap between what is held to be true (=believed) and what is the case.9 O S1’s O-sentences

S2’s O-sentences

S1

S2 mutual interpretation  genuine intersubjective disagreement

Figure 2: Intersubjective object-directed relations in “reflective” triangulation.

2.2. Davidson’s Non-Reductive Naturalism and “Continuity Skepticism” Davidson’s approach to thought as characterized so far is both naturalistic and non-reductive.10 It is naturalistic in that it locates the foundations of thought in the natural dispositions of creatures to discriminate and respond to parts the world, including other subjects’ responses to common objects. It is non-reductive in that it insists that the intentional is irreducible to the non-intentional—there can be no specification of conceptually necessary and sufficient conditions of thought in non-intentional terms. As we saw, 8

Note that the attitude of holding a sentence true is publicly observable, according to Davidson, and can be determined independently of knowing a subject’s beliefs and the sentence’s meaning for the subject. (See “Belief and the Basis of Meaning”, Davidson 2001b.) 9 For more on the role language plays, see especially, “The Second Person”, Davidson (2005: 111ff. and 119ff). 10 The themes of naturalism and non-reductionism in Davidson’s understanding of triangulation are explored in Sinclair (2002, 2005) and Ramberg (2001).

129

Davidson thinks that the only way to “upgrade” the basic, pure triangle from a mere necessary to a sufficient condition is to enhance the connection between subjects so it includes intersubjective linguistic communication. Absent mutual linguistic interpretation, even when (as in the pure scenario described above) one subject reacts to another’s reactions to some object, or anticipates the other’s reaction upon seeing the object, this can amount to nothing more than mere discriminative behavior (and thus fails to meet the normativity requirement). To suppose otherwise, Davidson thinks, is to commit to the reducibility of rule-following linguistic interactions to other forms of non-linguistic interactions. In Davidson’s hands, the irreducibility claim is transposed into a deep skepticism concerning not only the mindedness of non-linguistic creatures but also the very possibility of any sort of continuity between the minded and the non-minded parts of the world. By insisting that the only way to move significantly beyond the pure triangular scenarios found among nonlinguistic animals is to get linguistic communication into the picture, Davidson is clearly giving up on the possibility of an illuminating explanation of how objective thought could emerge. If Davidson is right, there is a deep, unbridgeable divide between the reflective/linguistic and the pure/non-linguistic triangles, which is rooted in a deep, unbridgeable divide between linguistic communication and other forms of interaction between subjects. We can locate in nature creatures with discriminatory responses and we can locate agents with the full package of interdependent thought and language. But we can find nothing in between through which the emergence of objective thought could be understood. Or, putting it in a “formal” mode (in terms of constraints on our vocabulary and concepts): «We have many vocabularies for describing nature when we regard it as mindless, and we have a mentalistic vocabulary for describing thought and intentional action; what we lack is a way of describing what is in between» (Davidson 2001a: 129). Indeed, the gap between mental and non-mental vocabularies «is particularly evident when we speak of the ‘intentions’ and ‘desires’ of simple animals. We have no better way to explain what they do. It is not that we have a clear idea what sort of language we could use to describe half-formed minds; there may be a very deep conceptual difficulty or impossibility involved» (ibid.).

130

It is important to recognize that this continuity skepticism represents a more radical position than the view that, as a conceptual matter, animals do not have our kind of mind, since they lack propositionally contentful states that are caught up in sufficiently intricate inferential networks and properly governed by norms of semantic correctness. It is one thing to maintain that non-linguistic animals are incapable of reflective intersubjectivity, and that (therefore) their interactions cannot support genuine objectivity. It is quite another to maintain that there is no philosophically intelligible, significant middle-ground between merely responsive-discriminatory behavior and fully conceptualized thought and speech, and thus between pure and reflective intersubjectivity. We may suppose (with Davidson) that no sense can be made of languageless animals having any propositional attitudes. Still, it may be possible to identify specific forms of non-linguistic behavior that are legitimate natural precursors of the sorts of behavior that (on Davidson’s view) do implicate objective thought. By “legitimate natural precursors” we here mean behavioral interactions that: a. can be found in the natural world, b. are described both by commonsense and by our best theories using familiar, intentional vocabulary, c. go beyond the merely discriminative behaviors of pure triangulation d. do not require us to presuppose the presence of propositional thought or language, and e. are genuinely intersubjective,11 f. support (or implicate) at least proto-objectivity Intersubjective interactions that have these characteristics will be richer than those found in pure triangulation, on the one hand, but will not yet amount to the linguistic interactions found in reflective triangulation, on the other hand. If such a middle-ground can be found, then even if no extant animal interactions (already) exhibit the sort of intersubjectivity Da11

“Genuine intersubjectivity”, as we’ll see below, has to do with the way subjects treat each other. While it’s true that we sometimes talk about mechanical devices using intentional vocabulary, we’re not similarly tempted to think of the “behavior” of mechanical devices as betraying their treating each other as subjects with a “take” on the world.

131

vidson takes to be essential for objective thought, Davidson’s continuity skepticism can still be undermined. What we’ll be calling below intermediate triangulation is designed to fit this bill. As a preemptive measure, we want to acknowledge that a committed continuity skeptic could insist that the interactions to be described in our intermediate triangulation only merit purely behavioristic, non-intentional characterizations, just like its pure predecessor.12 The obvious response to this, we think, will be to point out that the same is true of reflective triangulation. There is nothing to prevent a committed skeptic from redescribing intersubjective linguistic interactions in terms that leave it open whether the subjects involved are really minded, or really treat each other as minded subjects. Simply to invoke the possibility of redescription is to land in yet another kind of skepticism (a version of other minds skepticism) which is also more radical than continuity skepticism. (It would require not only denying that the “light” of mindedness “gradually dawns over the whole”—to borrow Wittgenstein’s famous dictum— but also accepting that the lights are out everywhere, so to speak.)13 Insisting that the animal case is entirely different from the linguistic case would either be question-begging, or would require a separate, substantial defense.

12

Davidson sometimes suggests that there is merely a veil of ignorance preventing us from explaining the behavior of non-linguistic creatures in non-intentional vocabulary, implying that the full story would not require anything more: «it is only necessary to reflect that someone might easily have no better or alternative way of explaining the movements of a heat-seeking missile than to suppose the missile wanted to destroy an airplane and believed it could by moving in the way it was observed to move» (Davidson 2001a: 102). 13 Other minds skepticism would, in any event, seem anathema to Davidsonian philosophy of mind, with its rejection of Cartesianism and externalism. These themes are especially evident in a number of essays under the headings of “subjective” and “intersubjective” in Davidson (2001a). See also Ramberg (2001) for an interpretation of Davidson as anti-skeptical regarding the external world and other minds.

132

3. Expressive Behavior We concede that genuine, object-directed intersubjectivity of the sort envisaged in Davidson’s triangles is essential for objective thought and linguistic communication. We also agree that pure triangulation fails to put animals on their way to full-blown objective thought. But we believe that certain expressive behaviors that we readily attribute to many non-human animals exhibit crucial features that Davidson claims to be absent from triangulation in its pure state—features that are relevant to the emergence of objective thought. Expressive behavior, we will suggest, can provide a linking bridge between merely passively discriminative animal responses apt only for avoiding extinction in an uncooperative world, on the one hand, and the reflective interactions among human agents that are subject to rational norms of correctness, on the other hand. But before explaining the relevance of expressive behavior to Davidsonian objectivity, we examine briefly the commonsense understanding of the notion of expression and expressive behavior.14 3.1. Expressive Behavior: “Showing vs. Telling” The class of behaviors we ordinarily describe as “expressive” spans a wide range. Though there may be no sharp lines to be drawn, at one end of the spectrum, we have so-called natural expressions (e.g., yelps, growls, grimaces, gasps, frowns and smiles), where both the behavior and its connection to the expressed states are supposed to be inculcated by nature. There are also mimicked or acquired facial expressions or gestures that become “second nature” (e.g., shrugging, tutting, but also a “shake” paw in dogs). Then we have conventional nonverbal expressions (e.g., tipping one’s hat, sticking out one’s tongue). Still in the conventional realm, we have expressive verbal utterances such as “Ouch!”, “Darn it!”, “Sorry!” and so on. We also find in the verbal domain sincere utterances of e.g. “There’s a crow on 14

Our discussion below draws on Bar-On (2004), especially ch.s VII and VIII, and on Green (2007). Bar-On and Green (2010), and (in progress), suggest that expressive behavior could figure in a natural reconstruction of the emergence of linguistic communication, where a “natural reconstruction” aims to show how each stage in a continuum develops from an earlier one in some intelligible way, given what we know about the natural abilities and history of the relevant creatures.

133

the telephone pole” expressing the speaker’s present belief, or “Let it rain” expressing the speaker’s wish for rain. Finally, at the extreme end of the conventional side of the spectrum, we have speech acts, such as assertion, or promising, which are often claimed to have the expression of certain mental states as part of their felicity conditions. In all these cases, what is said to be expressed (the “relata” of the relevant expression relation) are states of minds. Intuitively speaking, the expression relation concerned here—what we may call mental-state expression, or m-expression, for short—is different from what Sellars (1969) refers to as “expression in the semantic sense” (s-expression, for short).15 S-expression is a semantic, representational relation between linguistic strings or other representations and what they stand for (e.g., between the word “liberty”, as well as its translations, and the concept of liberty, or a tokening—verbal or mental— of “It’s raining” and the proposition that it’s raining at time t near the speaker). By contrast, m-expression is a psychological relation. In the realm of m-expression, we speak both of bits of behavior as expressing the states of mind that caused them (e.g., your trembling voice expressing— “betraying” as we might put it—nervousness), and of agents expressing their states of mind through intentional acts (e.g., a child expressing her joy at seeing you by giving you a hug, or your saying “It’s so great to see you”). A given state of mind can be expressed in any number of ways. On various occasions, upon being presented with a new fluffy teddy bear, little Jenny’s face may light up; or she may let out an excited gasp, reaching for the toy; or she may emit a distinctive sound (“Uh!”), or call out: “Teddy!” as she reaches for the toy; she may exclaim: “I want Teddy!” perhaps with no reaching. Jenny’s facial expression and her sigh may be purely “natural expressions” of her desire for Teddy; her reaching and subsequent utterances are things she voluntarily or perhaps intentionally does as she mexpresses her desire. On the other hand, Jenny’s utterances include English sentences, which s-express propositions or concepts. What renders all these instances of expressive behavior, it seems, has to do with similarities 15

Bar-On (2004, passim) makes use of Sellars’ three-fold distinction distinguishes the semantic from the causal and the action senses. For present purposes, the distinction between m- and s- expression will suffice.

134

among the performances or acts (whether voluntary or not). These similarities obtain despite significant differences that can be discerned among the different expressive vehicles used (or the various “products” of the acts).16 One can m-express joy at seeing a friend through a beaming facial expression or a spontaneous hug (neither of which stands in a semantic representational relation to the joy), or by uttering a sentence such as “It’s so great to see you!” (which s-expresses the proposition that it’s great to see the expresser’s addressee).17 Similar expressive performances, different vehicles of expression. Since our aim is to identify a middle-ground between non-linguistic and linguistic behavior, we have a special interest in a specific subcategory of expressive behaviors, namely natural expressions. Paradigm cases of natural expressions—yelps, growls, yawns, grimaces, teethbarings, tail-waggings, and so on—are often assumed by philosophers to be simply a subset of natural signs (such as clouds, deer tracks, tree trunk rings). They are often lumped together with physiological symptoms, such as red spots on the skin, sneezes, or galvanic skin response, and portrayed as simply reliable indicators of the internal states that regularly cause them.18 But a purely causal construal would fail to capture our ordinary notion of a natural expression. That notion seems designed to capture a different range of behaviors than the notion of indicative or symptomatic behavior. Following Hauser (1996: 9), we may distinguish natural signs from cues and signals. A natural sign is a symptom, a behavioral manifestation or a trace that provides information. A cue is a natural sign that has been biologically selected for its capacity to convey information (e.g., bright coloration in frogs, designed to warn away potential predators). And a signal is a cue that can be produced in response to changes in the environment 16

Bar-On (2004) distinguishes between an act of expressing and its product, on the one hand, and between the process and vehicle of expressing. 17 For a defense of a “neo-expressivist” construal, according to which the avowal: “I’m so glad to see you!” m-expresses the speaker’s joy at seeing the addressee, using a vehicle that s-expresses the self-ascriptive proposition (that the speaker is glad to see her addressee), see Bar-On (2004, especially ch.s VI-VIII). 18 See, for example, Grice (1957), Alston (1965), Bennett (1976).

135

(e.g., sexual swellings).19 In addition, there may be good reasons to separate off a sub-class of signals: communicative signals. These have the natural purpose of conveying information to (or producing certain reactions in) some designated audience (which is not to say that they are intentionally produced with that purpose by each expresser). A reasonable hypothesis concerning natural expressions is that they belong in the sub-category of communicative signals. If so, then this would already support the intuition that the notion of a natural expression has a much narrower application than that of a natural behavioral symptom or indicator.20 But natural expressions don’t just serve to communicate information or provide evidence about an individual’s states of mind; they also exhibit or display them. Correlatively, suitably attuned witnesses to expressive behavior don’t just infer or learn that the expresser is in the expressed state; they directly recognize it.21 So another attractive suggestion is that natural expressions are communicative signals that show the expressed states; they don’t just tell of them.22 Winces, growls, cowering demeanors, squeals of delight, gasps of fear, and so on, directly reveal the expressed state of mind. We often speak of seeing someone’s anger, hearing the nervousness in someone’s uneven voice, feeling the tension in someone’s body, and so on. Now one may hesitate to accept that naturally expressive behavior shows an individual’s state by literally making it perceptible. Still, it seems 19

For an alternative delineation of the relevant notions, offered by evolutionary game theory, see Maynard Smith and Harper (2003). 20 Relevantly, while Davidson denies that non-linguistic creatures can genuinely communicate, he does allow for signaling among such creatures. See Davidson (2005: 294, 2001c: 12). 21 So-called natural expressions also contrast with bee dances, which, unlike deer tracks and wounds, are presumably designed to convey information to designated consumers. 22 For the claim that natural expressions show or display expressed states, see, e.g. Ayer (1936) and Alston (1965). Bar-On (2004, especially ch.s VII, VIII and X) and Green (2007: especially ch.s 3-5) discuss the connection between expression and showing. Green (2007: ch. 3) includes showing-that (i.e., establishing the proposition that such & such) under expression. Bar-On (2010b) argues that Green’s “epistemic” understanding of expressing-as-showing threatens to blur the desired line between showing and telling.

136

reasonable to suggest that expressive behavior enables observers who are suitably attuned (by nature, habituation, or experience) directly to recognize through the behavior—as opposed to just inferring from it—the expressed state of mind.23 Naturally expressive behavior has a specific, social role to play in the lives of creatures capable of it. Arguably, such behavior is effective when it meets immediate and appropriate reactions on the part the designated audience, often toward the expresser or the environment; and these are likelier to ensue if the behavior enables direct recognition, as opposed to merely providing evidence about its hidden causes.24 3.2. The Complexity and Intentionality of Expressive Behavior The expressive performances of nonhuman animals do not employ expressive vehicles that have articulate semantic (or syntactic) structure; to use earlier terminology, they do not have powers of s-expression. Still, it seems that the capacity of nonhuman animals to engage in expressive behavior is a key factor behind the barely resistible practice we have of applying our mentalistic vocabulary to them (as well as to prelinguistic children). We think that this has to do with the fact that naturally expressive behavior possesses complexity and character that go well beyond those of mere natural signs or symptoms. It also has to do with the fact that creatures capable of such behavior exhibit more than discriminative responses to their environment (including other creaturely bits of it). An expressive creature is one capable of engaging in certain reciprocal intersubjective interactions. Its behavior can display severe pain and its location, a moderate as opposed to intense agitation, rage as opposed to panic, great excitement as opposed to mild curiosity, and so on. In addition, many naturally expressive acts reveal the state’s object: we recognize an infant’s excitement 23

A theory of expressive behavior will need to spell out what direct recognition amounts to (if it isn’t perception), so as to support the contrast between showing and telling. For relevant discussion, see Bar-On (2004, forthcoming), Green (2010) and Martin (2010). 24 The need for a construal of intersubjective interactions, not only among animals and prelinguistic children but also among adult human beings, in perception-like terms (as opposed to the terms of so-called “theory of mind”) is increasingly acknowledged by psychologists. See, e.g. Bermudez (2005), Gomez (2005), Hobson (2005).

137

at the sight of something in particular, or a pet’s fear of some specific threat, and so on; and they are designed to enjoin suitably endowed observers to act in certain ways (toward the expresser or the object of her expressed state). Consider, for example, human facial expressions such as those associated with anger, fear and sadness. These typically show to observers not only the presence of the relevant affective state, but also the object of that state—its source or target (perhaps through an accompanying overt gaze direction; see Eilan, et al, 2005). This is supported by ethological studies of nonhuman animal expressive communication (Miklósi, et al, 1998). A vervet monkey’s alarm call shows to other monkeys the caller’s fear of, say an aerial predator, thereby enjoining other vervet monkeys to hide from the danger (Cheney and Seyfarth 2007: ch. 10). Even a dog’s cowering demeanor upon encountering another dog will show to a suitably endowed recipient not only the dog’s fear (kind of state), how afraid he is (quality/degree of state), but also of whom he’s afraid (the state’s intentional object), and how he is prepared to act—slink away from the threat, or hide from it behind his owner’s leg (the state’s dispositional “profile”). Expressive performances thus have a janus-faced character, pointing inwards and outward at the same time. In contrast with automatic physiological reactions, or symptoms, they can exhibit aboutness. However, in contrast with purely perceptual representational states, expressive behavior also embodies action by showing what the animal is disposed to do, thereby enjoining appropriate responses on the part of the “consumers” of the behavior. How should the aboutness of expressive behavior be understood? Discussing alarm calls produced by certain birds under the rubric of “emotional displays”, the ethologist Peter Marler remarks: With careful study, we find that communication by emotional displays can be very complex, especially when prevarication is involved… Furthermore, if a bird couples a call with some kind of indexing behavior, such as headpointing or gaze direction, a certain object or point in space or particular group member can be precisely specified: the combination adds significantly to the communicative potential of emotion-based signals (Marler 2004: 176).

138

The suggestion is that a bird’s alarm is not to be simply understood as an instinctive or reflexive bit of behavior—a mere behavioral signal transmitting information about the presence of a predator. Rather, when producing an alarm call a bird m-expresses—and its intended audience can recognize—a complex state of mind: an affective state of a certain degree and quality (mild/extreme agitation, distress, or fear) directed at a predator of a particular type. But the state’s complexity need not be construed along the lines of the complexity of propositional attitudes (viz., having a particular psychological attitude—e.g. being afraid—toward a propositional representation of a worldly state of affairs, e.g. that a bird from above is about to attack). Instead it could be understood as a non-propositional, yet still intentional affective state.25 Just as we recognize that some psychological states—fatigue, hunger, thirst, general malaise—have no intentional objects, so we may sensibly recognize that there are more purely affective states that do not have propositions as their intentional objects. Such states may, of course, have propositional analogues/relatives. One may be afraid that x is about to pounce, angry that y is coming too close, excited by the fact that z is about to give one a treat. But it doesn’t seem plausible to suppose that all affective states (even in creatures like us, who are capable of entertaining propositions) must au fond be propositional attitudes.26 (In some places Davidson suggests that having any intentional state requires conceptualization: to be afraid of the cat, for example, requires the creature harboring the fear to have the concept of a cat, which, in turn, presupposes having lots of beliefs about cats.27 This argument belongs with the argu25

To say that a state is complex is not to say that it has parts or components that correspond to the dimensions or aspects of complexity. Similarly, a bit of behavior—e.g., an alarm call—that expresses a complex affective state may also lack composite structure. For some discussion which draws on useful distinctions by Sellars, see Bar-On and Green (2010). And see fn 28 below. 26 The contrary supposition (apparently endorsed by Davidson) hardly seems sustainable. Even in the case of linguistic creatures, can we accept that, e.g., John’s fear of spiders must be understood in terms of John’s fear that … ? The supposition also goes against recent findings in emotion research. See, e.g., Griffiths (1998). 27 Davidson says: “To have the concept of a cat, you must have the concept of an animal, or at least of a continuing physical object, the concept of an object that moves in certain ways, something that can move freely in its environment, something that has

139

ments designed directly to discredit the idea that non-linguistic animals can have thought, which we mentioned earlier and proposed to set aside, as they do not bear directly on continuity skepticism. Suffice it to say, however, that the argument from conceptualization rests on a rather specific view of concepts—viz., as devices of classification that must be backed up by general beliefs—which is not without more plausible alternatives, even in the case of linguistic beings.28) An alarm call, we’ve said, can be seen as an act that m-expresses an animal’s affective state, using a vehicle that does not s-express an articulate proposition (such as “There’s a hawk above”). But alarm calls (and other vocalizations) can be dissociated from the affective states that are mexpressed in acts of producing the calls,29 as well as from the presence of the intentional object they are designed to point to. And this, as we’ll see in the next section, may open up space for a variety of mismatches between expectations and action that seem to prefigure objectivity as Davidson conceives of it. At the same time, however, it is notable that the production and uptake of expressive behavior place much weaker demands on the cognitive capacities of both producers and consumers than does linguistic communication. On the expresser’s side there’s no need for the sort of sophisticated communicative intentions required even for (what Grice has sensations. There is no fixed list of things you have to know about, or associate with, felinity; but unless you have a lot of beliefs about what a cat is, you don’t have the concept of a cat” (2001: 124) And compare Brandom (2000) passim. 28 For a sample of alternative views of concepts, see Margolis and Laurence (eds.) (1999). Millikan (2000) offers a critique of the “concepts-as-classifiers” view. Due to limitations of space, we have had to gloss over a number of important issues concerning the content of non-propositional states/attitudes, such as the possibility that affective states only possess “nonconceptual content” (see Gunther, ed. 2003) and the question whether they obey the Generality Constraint (see Carruthers, Camp, and McAninch et. al. in Lurz 2009). For a very congenial “intermediate” view of animal intentional action, see Hurley (2003). Furthermore, note that the problem according to Davidson is not one of specifying an object, but rather one of the creature’s response to the object being a thinking, classificatory response. Even pure triangulation can specify an object. See especially Davidson (2001d: 12-13). 29 Indeed, the vocal patterns can be reproduced in the absence of the original expresser. Relevant here is Cheney and Syfarth discussion in (2007) of playback experiments with vervet monkeys.

140

called) speaker meaning. So less demand is put on the communicative capacities of the expresser. And on the observer’s side there’s no need for complicated “theory of mind”, or metarepresentational inferences. So less demand is put on the interpretive “mindreading” capacities of the observer.30 4. From Expressive Behavior to Linguistic Communication? In the previous section, we situated expressive behavior between the mere signaling behaviors of the animals that figure in Davidson’s pure, nonreflective triangulation, on the one hand, and the full-on linguistic communication that he takes to be required for interpretive, or fully reflective, triangulation, on the other hand. In this final section, we’d like to explain how expressive behavior can figure in intermediate—or “affective”— triangulation. Among the expressive behaviors mentioned so far, many are egocentric: they show an animal’s affective state that lacks a sharable cause, target, or object, or else they show it without pointing to a worldly object, event, or state of affairs that can provide a common focus for the expresser and his audience. By contrast, “world-centric” expressive behaviors do show the expressed state’s object, target, or source, as well as foreshadowing the expresser’s impending behavior. Thus, whereas a distress call may not reveal the source of the distress, an alarm call may show both the caller’s fear or agitation and what type of threat it’s afraid of. A dog’s whimper may show its pain, whereas a growl accompanied by a tilt of the head may reveal both the growler’s angry state of mind and at whom the anger is directed. And whereas a chick’s hunger peep only draws attention to the chick and its state, a hen’s food call draws attention to food that’s around. Now, we have accepted (at least for present purposes) Davidson’s claim that world-directed intersubjectivity is required for objectivity. What we wish to reject is the idea that there are no legitimate natural precursors of objective thought—no “proto-objectivity” prior to linguistic communication. What we oppose is the idea that the subjects occupying the base 30

See fn. 24 above for some references.

141

nodes in the triangles must already interact linguistically, before they could even be on their path to objective thought. For it is this idea that fuels Davidson’s continuity skepticism as we’ve articulated it. Using the above distinction, our anti-skeptical proposal is that world-centric m-expressive behavior can put creatures on the way to objective thought, even before they possess s-expressive vehicles. It’s important to emphasize the modesty of the claim we want to defend by appeal to (world-centric) expressive behavior, and distinguish it from several more ambitious anti-Davidsonian claims defended in recent literature. For example, we are not committed to the claim (defended by e.g. Carruthers 2004, 2009) that non-human animals (including insects!) have propositional thought and a range of other propositional attitudes. We are also not committed to the claim that non-human animals already possess the central abilities underlying language and linguistic communication (which is sometimes defended by appeal to experiments designed to teach language to animals). More relevantly, we do not wish to claim that animal non-linguistic communication already shares key essential features of linguistic communication. In particular, we are not claiming (as do McAninch et. al. 2009) that, e.g., episodes of animal alarm calls not only m-express the caller’s affective state, but also have products (or use vehicles) that already s-express propositions. Instead, we want to sketch an intermediate triangular scenario that exploits distinctive characteristics of the worldcentric expressive behavior found among non-linguistic animals, with the aim of making sense of a legitimate natural precursor of objectivity as Davidson understands it. 4.1. Expressive Behavior and Intermediate Triangulation For Davidson, the coordinated movements exhibited by schools of fish, the information transmission achieved by bee-dances, and the communication effected through the food calls of hens and the alarm calls of vervet monkeys, etc., all belong to a single, undifferentiated type of behavior when it comes to triangulation and objectivity.31 However, expressive behavior as we have portrayed it forms a significant sub-category within the sorts of behavior that Davidson lumps together under the umbrella of pure triangu31

Davidson (2001a: 129).

142

lation. The intermediate triangulation we describe below takes advantage of the special “affective intersubjectivity” characteristic of creatures engaged in the production and uptake of expressive behavior. And this may allow us to break through the barrier put up by Davidson’s continuity skepticism. Pure triangulation scenarios are marked by the fact that they involve interactions between two (or more) subjects and an object, such that each subject reacts to the object by way of reacting to the other subject. Intermediate triangulation scenarios, we think, are richer in a specific, objectivityrelevant way. As noted earlier, world-centric expressive behavior is Janusfaced: it show the expresser’s state of mind to a suitably endowed audience, foreshadowing the expresser’s impending behavior; it also directs attention towards the object/state of affairs to which the expressive behavior is responsive or at which it is directed, and enjoins appropriate behavior on the observer’s part. Thus, suppose that S1 engages in expressive behavior E. In so doing, S1 is showing to S2 (a suitably endowed audience) how he, S1, is about to behave—that is, do something A (pounce, slink away, hide) that is appropriate to the presence of O. Having observed E, S2 is in a position to respond to E in some way B that is not merely responsive to the presence of O (as indicated by E) but is also anticipatory of S1 doing A. Now we can consider whether S2’s doing B is itself appropriate to the presence of O. Suppose it is not. Suppose, for example, that E is an alarm call, that A is fleeing, and that O is a predator.32 Instead of also fleeing, for example, S2 may, upon hearing the alarm call and spying no predator, respond to S1’s alarm call by, say, moving toward S1 to consume S1’s soonto-be-abandoned meal. When S1’s O-behavior betrays her impending flight, the possibility opens for S2’s response to the behavior to agree or disagree with it, depending on whether or not S2’s response is itself a bit of behavior appropriate to the presence of O. But in a case as just described where S2’s behavior (B) departs both from S1’s anticipated behavior (A) and from the behavior enjoined by E, S2’s behavior can be said to embody 32

In saying “O is a predator” we’re not implying that either S1 or S2 themselves have the concept of a predator. All we’re assuming is that they have whatever discriminatory dispositions vis-à-vis O that Davidson allows subjects to have in pure triangulation. See again, e.g., Davidson (2001c: 12-13), (2001a: 117ff.).

143

O-related disagreement with S1’s behavior. Here it looks as though S2 is treating S1’s O-related behavior as separable from the imminent presence of O, as assessed by S2. S2 is keeping two distinct but simultaneous tabs, as it were, on the world and on S1’s reaction to it. The right space seems to be open for crediting S2 with recognizing S1 as having his own take on the situation. For O is no longer merely an external cause serving as a point of intersection of S1 and S2’s discriminatory responses; nor is S1’s behavior treated merely as a natural O-indicator by S2. Instead, S2’s responsive behavior is one that takes account of what amounts to S1’s getting things wrong (from S2’s perspective). O

S1’s O-centric expressions

S2’s O-centric expressions

S1

S2 production/uptake of O-centric expressive behavior  “embodied” affective disagreement

Figure 3: Intersubjective object-directed relations in “intermediate” triangulation

This sort of “affective interlocking” is poised between the “thin” case, in which S2 simply fails to respond to S1’s O-behavior with her own Obehavior, on the one hand, and the “thick” case in which S2 judges that S1’s O-behavior is incorrect (specifically, that S1 has uttered a false sentence, betraying a false belief), on the other hand. The thin case is certainly possible within Davidson’s pre-linguistic triangulation. It is possible, for example, that, despite the associative connection set up between S1 and S2, such that each responds to the other’s O-behavior with their own Obehavior, and despite perceiving S1’s O-behavior, S2 fails on some particular occasion to respond to S1’s O-behavior as if to O. Indeed, either subject may on any particular occasion fail to respond to O with their own O-behavior. As Davidson emphasizes, however, the possibility of such failures in regularities does not support any objectivity (even if such regularities do provide a necessary condition for objectivity by providing a

144

background against which the grasp of the concept could develop).33 What is missing is something that carries us beyond purely behavioral discord to genuine intersubjective disagreement. As explained earlier, once language is on the scene, such genuine disagreement becomes straightforwardly possible. But, of course, that takes us all the way to the thick, reflective case. What we’re calling “intermediate triangulation”, with its affective interactions, allows for a kind of intersubjective disagreement that does not presuppose the reflective grasp of objectivity or intentional concepts that require language (on Davidson’s view, at least). So it’s clearly different from the thick, reflective case. But it’s also different from the thin, or pure case. To see the crucial difference, recall that, on the expresser’s side, expressive behavior shows the affective state the expresser is in by revealing how he, S1, is prepared to react in light of O’s presence. Faced with S1’s expressive performance, the observer, S2, can anticipate not only O’s presence, but also the behavior on S1’s part that is foreshadowed by that performance. Moreover, since expressive behavior also has the function of enjoining appropriate behavior on the part of suitably attuned observers, proper uptake of S1’s expressive behavior requires a certain O-related reaction on S2’s part. By foreshadowing impending O-related behavior on S1’s part, and enjoining appropriate O-related behavior on S2’s part, the O-directed affective interactions in which the expresser and her audience are interlocked make room for a broader set of intersubjective mismatches/disagreement than Davidson allows in the pure case. Having been enjoined to some behavior, S2’s responses to S1 are no longer limited to her own O-related responses (appropriate or not). S2’s responses can differ not only from S1’s own O-behavior, but also from the behavior enjoined by S1’s expressive behavior. Moreover, faced with S1’s O-centric expressive behavior, S2 may anticipate not only O’s presence (treating S1’s behavior simply as a natural indicator of O), but also S1’s impending (Orelated) behavior, which may or may not match her own. S2’s responsive behavior can thus be said to be addressed to S1’s O-centric expressive behavior itself (say, the alarm call), as separable from the presence of O (say, the threat); it embodies disagreement with it. In this relatively noncommittal sense, S2 can be seen as treating S1 as an entity with a subjective take 33

Davidson (2004: 141-143), (2001c: 5-7).

145

on the world, rather than merely as a bit of the world that simply naturally indicates the presence of another bit (that is, something whose behavior is a natural sign of O’s presence). 4.2. Some Concluding Remarks In “What Thought Requires”, Davidson says: «Triangulation also creates the space needed for error, not by deciding what is true in any particular case, but by making objectivity dependent on intersubjectivity» (Davidson 2004: 144). And in “The Emergence of Thought”, he says «Where do we get the idea that we may be mistaken, that things may not be as we think they are? Wittgenstein has suggested, or at least I take him to have suggested, that we would not have the concept of getting things wrong or right if it were not for our interactions with other people» (Davidson 2001a: 130). It is clear that, for Davidson, genuine objectivity is founded on (the right sort of world-directed) intersubjectivity. We think that intermediate triangulation with its world-centric affective intersubjectivity opens up space for (at least) proto-objectivity. Though we do not yet have truth/falsity in the picture, since there are no vehicles that s-express truthevaluable propositions, we do have a set of interlocking appropriate reactions and actions on the part of the two subjects at the base of our intermediate triangle. And, although we may accept that the subjects in intermediate triangulation lack the full reflective awareness that’s involved in applying the concept of error, we have the right set up for the idea of a distance between a “subjective take” and the way things are to emerge. Adding expressive behavior to the pure triangular scenario does not threaten to reduce intentionality to mere causal relations between subjects and objects. Even if expressive behavior, like everything else in the universe, can be “described in the physical vocabulary”,34 we have rejected the idea that it can be fully understood in terms of purely causal regularities. Moreover, we have deliberately left it open that the O-behavior of each subject as well as O itself are inter-defined. Thus, S1’s O-behavior may be, in part, defined in terms of its impact on S2 (i.e., enjoining Oappropriate behavior on S2’s part). Qua communicative signal, then, expressive behavior may well be the behavior that it is partly by virtue of its 34

See Davidson (2001a: 128).

146

(naturally intended) effect on its (naturally designated) audience. Protoobjectivity is thus not bought at the price of reducibility. At the same time, the intermediate triangulation we have described is significantly richer than Davidson’s pure triangulation. As we saw, Davidson holds that the concept of error is required and sufficient for objectivity but that pure triangulation merely provides a situation in which the concept of error could come to— but does not yet—have application, since it does not make room for intersubjective agreement or disagreement. However, we have tried to show that interactions involving expressive behavior do. Even though the expressers and their audience do not (by hypothesis) possess the concept of error, and have no reflective awareness of objectivity, their interlocking world-directed acts and responses embody a contrast between subjective and objective, which seems to prefigure objectivity as Davidson conceives of it. Thus, we think that, by assimilating animals’ expressive behaviors to other sorts of non-linguistic merely discriminatory behaviors, Davidson misses the significance of expressive behavior for establishing continuity between the non-linguistic and the linguistic. The expressive behavior scenario we have envisaged is clearly at some distance from ones involving linguistic subjects. But a Davidsonian opponent might reject the triangular scenario we offered on the grounds that it is still in some indirect way “linguistically-infected”.35 The opponent may again call on Davidson’s claim that intentional states of mind of the sort we have supposed to be m-expressible by animals in our intermediate triangulation would require powers of conceptualization and propositional thought of a sort that only language users can be credited with. Our discussion of the intentionality of expressive behavior (see 3.3 above) has gone some way toward addressing this objection. Moreover, as explained at the outset, our aim has not been to rebut those arguments of Davidson’s that try to establish directly the dependence of all thought and conceptualization on language, or to support skepticism about animal mentality. Nor, as suggested earlier, do we take ourselves to have rebutted a skeptic who insisted that the affective interactions we describe could be redescribed in 35

This is a charge that can be legitimately raised, we think, against the pre-reflective triangulation envisaged by Eilan (2005), which is inspired by the phenomenon of joint attention between prelinguistic infants and caregivers.

147

non-intentional terms. For that, we take it, is true of the linguistic interactions that feature in Davidson’s reflective triangulation. Rather, our goal has been to break through the specific barrier put up by Davidson’s continuity skepticism, by showing that we can make sense of an intermediate case of triangulation in nature that can carry non-linguistic creatures beyond the merely passive, bare discriminatory behavior of pure triangulation, and bring them closer to reflective triangulation. We’ve accepted Davidson’s anti-reductionist naturalist view of intentionality, as well as the significance of triangulation for our understanding of it. But we have rejected Davidson’s continuity skepticism, with its attendant claim that there is an unbridgeable chasm between linguistic humans and nonlinguistic beasts. We appealed to a commonsense understanding of expressive behavior, as a type of behavior we share with nonlinguistic animals, as well as prelinguistic children. Commonsense descriptions of expressive behaviors (and the theoretical descriptions inspired by them) are well suited to characterizing “what is in between” descriptions of “nature when we regard it as mindless” and full-blown mentalistic descriptions of propositional thought and intentional action. And the category of expressive behavior may well form a significant “joint” at which to carve a natural world that contains both us and the beasts. At the very least, we hope to have shown that expressive behavior should command the attention of anyone interested in a “triangulation-friendly” yet non-skeptical answer to the question: What, other than language itself, could put nonlinguistic creatures on their way to objective thought?36 References Alston, W. 1965, “Expressing”, in M. Black (ed.), Philosophy in America. Ithaca, NY: Cornell University Press, 15-34. Ayer, A.J. 1936/1946, Language, Truth and Logic. London: V. Gollancz ltd. Bar-On, D. 2004, Speaking My Mind: Expression and Self-Knowledge. Oxford: Clarendon Press. 36

For comments on an earlier draft of this paper, we’d like to thank Mitchell Green, Alan Nelson, Dean Pettit, and Keith Simmons.

148

Bar-On, D. 2010a, “Avowals: Expression, Security, and Knowledge: Reply to Matthew Boyle, David Rosenthal, and Maura Tumulty”, Acta Analytica, 25 (1), 47-63. Bar-On, D. 2010b, “Expressing as ‘Showing What’s Within’”, Philosophical Books, 51 (4), 212-227. Bar-On, D. and Green, M. 2010, “Lionspeak: Communication, Expression, and Meaning”, in J. O’Seah and E. Rubenstein (eds.), Self, Language, and World: Problems from Kant, Sellars, and Rosenberg; in Memory of Jay F. Rosenberg. Atascadero, CA: Ridgeview Publishing Company. Bennett, J. 1976, Linguistic Behavior. Cambridge: Cambridge University Press. Bermudez, J.L. 2009, “Mindreading in the animal kingdom”, in R. Lurz (ed.), The Philosophy of Animal Minds. Cambridge: Cambridge University Press, 145-164. Brandom, R. 2000, Articulating Reasons: An Introduction to Inferentialism. Cambridge: Harvard University Press. Bruner, J. 1977, “Early social interaction and language acquisition”, in H.R. Schaffer (ed.), Studies in Mother-Infant Interaction. New York: Academic Press. Buchen, L. 2009, “Culture May Be Encoded in DNA”, Wired Science, . Camp, E. 2009, “A language of baboon thought”, in R. Lurz (ed.), The Philosophy of Animal Minds. Cambridge: Cambridge University Press, 108127. Carruthers, P. 2009, “Invertebrate concepts confront the generality constraint (and win)”, in R. Lurz (ed.), The Philosophy of Animal Minds. Cambridge: Cambridge University Press, 89-107. Carruthers, P. 2004, “On Being Simple Minded”, American Philosophical Quarterly, 41 (3), 205-220. Cheney, D. and Seyfarth, R. 2007, Baboon Metaphysics. Chicago: University of Chicago Press. Davidson, D. 1999, “Reply to Evnine”, in L. Hahn (ed.), The Philosophy of Donald Davidson. Chicago and La Salle, IL: Open Court, 305-310.

149

Davidson, D. 2001a, Subjective, Intersubjective, Objective, 2nd ed. Oxford: Clarendon Press. Davidson, D. 2001b, Inquiries Into Truth and Interpretation, 2nd ed. Oxford: Clarendon Press. Davidson, D. 2001c, “Externalisms”, in P. Kotatko et al. (eds.), Interpreting Davidson: Selected Papers from the 1996 Karlovy Vary Symposium on Analytic Philosophy. Palo Alto: CSLI, 1-16. Davidson, D. 2001d, “Comments on Karlovy Vary papers” in P. Kotatko et al. (eds.), Interpreting Davidson Selected Papers from the 1996 Karlovy Vary Symposium on Analytic Philosophy. Palo Alto: CSLI, 285-308. Davidson, D. 2004, Problems of Rationality. Oxford: Clarendon Press. Davidson, D. 2005. Truth, Language and History. Oxford: Clarendon Press. Eilan, N., Hoerl, C., McCormack, T. and Roessler, J. (eds.) 2005. Joint Attention: Communication and Other Minds. Oxford: Clarendon Press. Eilan, N. 2005, “Joint attention, communication and mind”, in Eilan et al. (eds.), Joint Attention: Communication and Other Minds. Oxford: Clarendon Press, 1-33. Gluer, K. 2006, “Triangulation”, in E. LePore and B. Smith (eds.), Oxford Handbook of the Philosophy of Language. Oxford: Oxford University Press, 1006-1019. Green, M. 2007, Self-Expression. Oxford: Clarendon Press. Green, M. 2010, “Replies to Eriksson, Martin, and Moore”, Acta Analytica, 25 (1), 105-117. Grice, H.P. 1957, “Meaning”, Philosophical Review, 66, 377-388. Griffiths, P.E. 1998, “Emotions”, in W. Bechtel and G Graham (eds.), A Companion to Cognitive Science. Oxford: Blackwell Publishing Ltd., 197203. Gomez, J. 2005, “Joint Attention and the Notion of Subject: Insights from Apes, Normal Children, and Children with Autism”, in Eilan et al. (eds.), Joint Attention: Communication and Other Minds. Oxford: Clarendon Press, 6584.

150

Gunther, Y. 2003, Essays on Nonconceptual Content. Cambridge, Mass. and London: MIT Press. Hauser, M. 1996, The evolution of communication. Cambridge, Mass.: MIT Press. Heal, J. 2005, “Joint attention and understanding the mind”, in Eilan et al. (eds) Joint Attention: Communication and Other Minds. Oxford: Clarendon Press. Hobson, P. 2005, “What puts the jointness into joint attention?” in Eilan et al. (eds.), Joint Attention: Communication and Other Minds. Oxford: Clarendon Press, 185-204. Hurley, S. 2003, “Animal Action in the Space of Reasons”, Mind & Language, 18 (3), 231-256. Lurz, R. 2009, The Philosophy of Animal Minds. Cambridge: Cambridge University Press. Margolis, E. and Laurence S. 1999, Concepts: Core Readings. Cambridge, Mass.: MIT Press. Marler, P. 2004, “Bird Calls: A Cornucopia for Communication”, in P. Marler and H. Slabbekoorn (eds.), Nature’s Music: The Science of Birdsong. London: Elsevier, 132-177. Martin, M. 2010, “Getting On Top of Oneself: Comments on Self-Expression”, Acta Analytica, 25 (1), 81-88. Maynard Smith, J. and Harper, D. 2004, Animal Signals. Oxford: Oxford University Press. McAninch, A., Goodrich, G. and Allen C. 2009, “Animal Communication and Neo-Expressivism”, in R. Lurz (ed.), The Philosophy of Animal Minds. Cambridge: Cambridge University Press, 128-144. McDowell, J. 1994, Mind and World. Cambridge, Mass.: Harvard University Press. Miklosi, A., Polgardi, R., Topal, J. and Csanyi, V. 1998, “Use of ExperimenterGiven Cues in Dogs”, Animal Cognition, 1, 113-121. Millikan, R. 2000, On Clear and Confused Ideas. Cambridge: Cambridge University Press.

151

Moore, C. and Dunham, P.J. (eds.) 1995, Joint Attention, Its Origins and Role in Development. Hillside, NJ: Erlbaum. Pagin, P. 2001, “Semantic Triangulation”, in P. Kotatko et al. (eds.), Interpreting Davidson: Selected Papers from the 1996 Karlovy Vary Symposium on Analytic Philosophy. Palo Alto: CSLI, 199-212. Ramberg, B. 2001, “What Davidson Said to the Sceptic”, in P. Kotatko et al. (eds.), Interpreting Davidson: Selected Papers from the 1996 Karlovy Vary Symposium on Analytic Philosophy. Palo Alto: CSLI, 213-236. Reddy, V. 2005, “Before the ‘third element’: understanding attention to self”, in Eilan et al. (eds.), Joint Attention: Communication and Other Minds. Oxford: Clarendon Press, 85-109. Roessler, J. 2005, “Joint attention and the problem of other minds”, in Eilan et al. (eds.), Joint Attention: Communication and Other Minds. Oxford: Clarendon Press, 230-259. Rosenberg, J. 2007, Wilfrid Sellars: Fusing the Images. Oxford: Oxford University Press. Rosenberg, J. 1985, The Thinking Self. Philadelphia: Temple University Press. Sellars, W. 1981, “Mental Events”, Philosophical Studies, 39, 325-45. Sellars, W. 1969, “Language as Thought and as Communication”, Philosophy and Phenomenological Research, 29, 506-527. Sellars, W. 1956, “Empiricism and the Philosophy of Mind”, in H. Feigl and M. Scriven (eds.), Minnesota Studies in the Philosophy of Science, vol. I. Minneapolis: University of Minnesota Press, 253-329. Sinclair, R. 2002, “What is Radical Interpretation? Davidson, Fodor and the Naturalization of Philosophy”, Inquiry, 45, 161-184. Sinclair, R. 2005, “The Philosophical Significance of Triangulation: Locating Davidson’s Non-Reductive Naturalism”, Metaphilosophy, 36 (5), 708-728. Verheggen, C. 2007, “Triangulating with Davidson”, The Philosophical Quarterly, 57 (226), 97-103. Werner, H. and Kaplan, B. 1963, Symbol Formation. Hillsdale, NJ: Erlbaum.

INTERPRETING ASSERTIONS Sanford C. GOLDBERG (Northwestern University, Evanston, USA)

1. In this paper I will be exploring the interpretation of assertions. My claims will be two: first, that assertion has an epistemic norm; and second, that this has implications for the nature of what Donald Davidson called “the task of interpretation” in general and “triangulation” in particular. Since this volume is focused on the epistemological dimension of triangulation, I want to start by connecting this topic to the theme I will be discussing. By “the epistemological dimension of triangulation” I understand the epistemological issues arising in the attempt by one subject to identify the objects, and subsequently the contents, of another subject’s speech acts and propositional attitudes, when these are directed at salient objects and properties in the shared environment. Here is Davidson’s characterization of the matter: It takes two points of view to give a location to the cause of a thought, and thus to define its content. We may think of it as a form of triangulation: each of two people is reacting differentially to sensory stimuli streaming in from a certain direction. Projecting the incoming lines outward, the common cause is at their intersection. If the two people now note each other’s reactions (in the case of language, verbal reactions), each can correlate these observed reactions with his or her stimuli from the world. A common cause has been determined. The triangle which gives content to thought and speech is complete. But it takes two to triangulate (Davidson 1991: 212-13).

For Davidson, the philosophical significance of triangulation (within the task of interpretation) is twofold. Most fundamentally, Davidson appealed to triangulation as part of an argument against “proximal” construals of the causes, and hence of the contents, of speech and thought (see Davidson 1992: 118-119). Here triangulation is part of Davidson’s account of the philosophical foundations of the intentionality of language and thought. But Davidson also put his account of triangulation to more applied ends: he offered it as a positive account of how a hearer identifies the “distal”

154

causes, and hence the contents, of speech.1 In this use, triangulation is a procedure by which hearers interpret speakers on those occasions on which there is some question as to the reference, and in particular the demonstrative reference, that the speaker is making. In connection with this latter use, discussions of the epistemology of triangulation can be seen as a part of a broader discussion of the epistemological dimension of the task of interpretation. Seen in this light, triangulation is not restricted to the context of what Davidson called “radical interpretation”. On the contrary, it can be and typically is involved in cases in which a hearer is trying to recover the demonstrative reference made by a speaker on a given occasion. And beyond this, a hearer might resort to a process of triangulation in cases in which her comprehension2 of the source’s speech act did not go smoothly, where the unclarity in the hearer’s comprehension reflects her uncertainty about one or more of the speaker’s intended references. In this paper I will be focusing on the epistemological dimension of the task of interpretation in cases involving sincere assertion. This takes us some distance away from contexts of radical interpretation: far from looking at one subject’s attempt to interpret an “alien language”, I will be focusing on situations in which two participants in a speech exchange are colinguals, and where the hearer recognizes the speech act as one of (sincere) assertion. Nevertheless, I aim to shed some light on the epistemological dimension of the task of interpretation generally, and on the epistemological dimension of triangulation in particular. My claim will be this: in a situation in which one subject, H, is warranted in regarding another subject, S, as having made a sincere assertion, H is entitled to assume that S regards herself as satisfying some substantial epistemic condition regarding the truth of the claim she (S) made. This entitlement, in its turn, can be seen as imposing a constraint on interpretation. More specifically, on the assumption that assertion has an epistemic norm (and that this is mutually 1

Actually, in speech and thought. I will continue, however, to focus on speech. I prefer “comprehension” to “interpretation” since the latter carries connotations—of the deliberateness with which one arrives at one’s construal of another’s speech— which are lacking in our ordinary verbal interactions with other speakers in our speech community.

2

155

familiar to all), we can justify a methodological injunction that requires the interpreter to exhibit a sort of “epistemological charity” in her rendering of the speech she is interpreting. I will not be claiming that this is the only, or the best, justification for the methodological principle I will be presenting; only that it is (part of) a justification for that principle. I regard this as significant if only because it suggests that a variant on a (now somewhat oldschool) theme in Davidsonian philosophy of language may get new life from the current work on the norm of assertion. 2. Davidson considered the phenomenon of triangulation, first and foremost, in terms of the role it played in the project of radical translation. As he conceived of it, the project of radical translation is the project of “solving for” the beliefs and meanings of another (would-be) speaker, without assuming knowledge of either. The radical interpreter solves for these by prompting the interpretee with sentences from her (the interpretee’s) language, and then correlating the interpretee’s assent and dissent behaviors with the salient environmental conditions prevalent at the time. Following Quine, Davidson held that there are certain ineliminable assumptions a hearer-interpreter would have to make if she is to “break into the beliefmeaning circle” in this way. Most salient among these assumptions was the Principle of Charity. Very roughly put, this principle maintains that interpretations of a subject’s language are to be preferred that maximize the number of truths (in speech and thought) that would be attributed to the speaker. We can appreciate the point of triangulation by understanding the role it plays in the interpreter’s attempt to satisfy the Principle of Charity. Charity demands truth-maximization; triangulation can be seen as the procedure by which the interpreter aims to satisfy this demand, when she is assigning objects to (what she provisionally takes to be) the referential expressions of the speaker’s language. It is easy to see how triangulation might play this role. Triangulation is a method for identifying the objective causes—or at least the objective eliciting conditions—of (some restricted set of) an-

156

other’s speech acts.3 Davidson proposes that these causes/eliciting conditions play two very important roles. On the one hand, they provide the contents of the relevant speech acts: roughly put, they are the objects and properties that the speaker is “speaking about”. On the other hand, in at least some cases these causes/eliciting conditions provide the speech act’s circumstances of evaluation, the circumstances in which the speech act in question is evaluated for truth. Of course, not all speech acts have contents whose truth-value can be determined in the very context of utterance: consider “On Saturday Sammy sang seven sultry songs in Singapore”, as uttered on Sunday, sans Sammy, in St. Louis. But while not all speech acts can be assessed for truth in the very context of utterance, some can. And it is in these cases that triangulation can be seen as serving the ends of Charity: in the simplest of these cases, the very objects and properties that determine the content, and so the truth conditions, of the statement are also the objects and properties that determine the truth-value of the statement, so interpreted. In the simplest of these cases, the speech acts (and the beliefs they express) come out true.4 Extrapolating from the simplest cases, Davidson himself was very clear about the demands of such a method of interpretation. He wrote, Torn between the need to make sense of a speaker’s words and the need to make sense of the pattern of his beliefs, the best we can do is choose a theory of translation that maximizes agreement (1969: 101). This method [of radical interpretation] is intended to solve the problem of the interdependence of belief and meaning by holding belief constant as far as possible while solving for meaning. This is accomplished by assigning truth conditions to alien sentences that make native speakers right when 3

I will continue to speak of triangulation in connection with the interpretation of speech acts, but similar points could be made in terms of the interpretation of (the sentences involved in) a subject’s assent and dissent behaviors, as well as the interpretation of the subject’s singular (especially demonstrative) thoughts. 4 In Davidson’s own words: «[T]he stimuli that cause our most basic verbal responses also determine what those verbal responses mean, and the content of the beliefs that accompany them. The nature of interpretation guarantees both that a large number of our simplest perceptual beliefs are true, and that the nature of these beliefs is known to others» (1991: 213).

157

plausibly possible, according, of course, to our own view of what is right (1973: 137). The general policy […] is to choose truth conditions that do as well as possible in making speakers hold sentences true when (according to the theory and the theory builder’s view of the facts) those sentences are true (1974a: 152).

But what is the justification for this interpretative procedure? The familiar Davidsonian response is that it is only by following something like this procedure that we can render the other speaker intelligible: «What justifies the procedure is the fact that disagreement and agreement alike are intelligible only against a background of massive agreement» (Davidson 1973: 137). Interpretability, then, is the crux on which the method’s justification turns. The task of radical interpretation is meant to highlight the nature of the evidence available to an interpreter in the case in which the interpreter knows nothing of the language or beliefs of the interpretee. In this sense Davidson’s early discussions of radical interpretation5 can be seen as a contribution to what we might call the “epistemology of interpretation”. But if we think of Davidson’s discussions in this way, we must be clear about how Davidson understood this: at the time Davidson was primarily interested in the epistemology of interpretation as this was seen from the perspective of the interpreter, with little or no attention to the epistemic perspective of the interpretee.6 It is not that Davidson did not characterize the interpreter’s perspective on the interpretee; of course he did. It is rather that Davidson’s characterizations of the interpreter’s perspective on the interpretee are primarily, if not entirely, in semantic terms. Absent is any ex5

By “early discussions” I mean those collected in Davidson 1984. This is not entirely true. After all, Davidson famously made an analogy between the problem of radical interpretation, in which one has to break into the belief-meaning circle of another speaker, with the problem of “radical decision-theory”, where one has to break into the belief-preference circle of another agent (Davidson 1974a, 1975). Still, so far as I can tell, in his early discussions he did not think systematically about how the task of interpretation might be affected by the fact that the interpretee is also an epistemic subject, one who aims to acquire (not just true belief but) knowledge.

6

158

plicit acknowledgement that the process of radical interpretation is one in which the interpreter must think of the interpretee as in the business of acquiring knowledge of the world. As Davidson described matters, it is as if the interpreter, H, simply regards the interpretee, S, as providing H with evidence for a theory H is constructing regarding S—as if S is nothing other than another piece of furniture in H’s world. True, this particular piece of furniture is a complicated one, in that at the very same time that S is in the world, S is also reacting to the world (i.e., by producing utterances in the face of worldly events and conditions). Still, S remains a piece of furniture for all that—or so Davidson’s early descriptions of the task of interpretation might encourage us to suppose. It is interesting in this regard to contrast Davidson’s early descriptions of the task of interpretation with his subsequent writings on the topic. These bring in a decidedly epistemological angle on the task, and one that appears to acknowledge the epistemic perspective of the interpretee as well. Thus in his 1991 paper “Three Varieties of Knowledge”, Davidson explicitly connects issues of triangulation with issues of knowledge. He writes that «the triangulation which is essential to thought requires that those in communication recognize that they occupy positions in a shared world» (Davidson 1991: 213). Since what holds for H when interpreting S also holds for S when interpreting H, and since this is something that both H and S can appreciate on reflecting on the process of interpretation itself, Davidson here has all of the pieces he needs to recognize that, in the course of interpretation, the interpreter must regard the interpretee as herself in the business of acquiring and communicating knowledge. I regard Davidson’s move to a more encompassing orientation—one that has in its view how things are with the interpretee both semantically and epistemically—as a step in the right direction. Still, I think Davidson himself did not move quite far enough in this direction. For while he speaks of the interpreter and interpretee as needing to «recognize that they occupy positions in a shared world», Davidson himself does not dwell on how this might affect their epistemological characterization of one another. He does speak the need, in the course of triangulation, for interpreters to regard interpretees as responding «knowingly and intentionally» to the relevant stimulation (Davidson 1992: 120-121). But he quickly moves on

159

from his “hasty remarks” without indicating how the point at issue affects our understanding of the task of interpretation itself. This is unfortunate. To see why, suppose that our conception of the task of interpretation acknowledges that an interpreter will have to think of her interpretee in distinctly epistemological terms: as an epistemic subject, one aiming to acquire knowledge of the world. In so doing we acknowledge that part of the explicit aim of interpretation is that of capturing of the interpretee’s distinctly epistemic perspective. It is not enough to see her as a creature with thoughts, who can give verbal expression to those thoughts. Rather, she is also to be seen as aiming (in her thought and speech) at knowledge. What emerges from this, I submit, is a deeper explanation of the value of intelligibility, and perhaps its role in the justification of thisor-that methodological principle of interpretation. Recall that Davidson’s justification of the method of radical interpretation was that we cannot render another speaker intelligible unless we follow the interpretative methods he described (Davidson 1984: 137, 2001: 121). Whether or not he was right on this point, we can still wonder: why all the fuss about rendering another speaker intelligible? It is in addressing this question that it pays to regard other speakers from a distinctly epistemological perspective. For one of the payoffs of rendering another subject intelligible is that such a subject can then serve as a potential source of worldly knowledge for the interpreter herself. So far as I can tell, Davidson himself never explicitly considered how this role others play, or can play, for us—a role in which they are sources of knowledge regarding a shared world—might bear on the methodological principles of interpretation themselves. But this raises an obvious question. How does a hearer H come to know, of a speaker S who H aims to interpret, that S is in the business of knowledge acquisition? Alternatively: how does S “signal” to H that she, S, is in that business? The answer, I think, is obvious: S makes assertions, and H apprehends this fact. 3. The main point of assertion, we might say, is to present-as-true a given content, in such a way as to implicate one’s own epistemic authority re-

160

garding the truth of the content so presented.7 It is thus by participating in the practice of assertion that a speaker manifests her role (for others) as a potential source of knowledge, or at least of warranted belief. And it is in correctly apprehending another’s assertion that a hearer recognizes that the speaker has so presented herself. Assertions are not always warranted, of course; nor are they always sincere. And hearers are not always entitled to accept the assertions they observe. But the central point remains: in making a sincere assertion a speaker has performed an act in which she manifests the role she plays as a potential knowledge source, and this is something that a hearer familiar with the practice of assertion is in a position to discern upon observing the assertoric act itself. For these reasons it seems that an attempt to derive methodological principles of interpretation from the role speakers play as potential sources of knowledge does well to start by focusing on cases involving the apprehension of sincere assertion. This is the focus of the remainder of this paper. Before proceeding further, however, a preliminary worry needs to be addressed. A thoroughgoing Davidsonian might doubt whether anything very deep can be learned about the methodology of interpretation from the sorts of case I will be examining, where a hearer recognizes that a colingual has made an assertion. As Davidson conceived of it, the «methodological problem of interpretation» is «to see how, given the sentences a man accepts as true under given circumstances, to work out what his beliefs are and what his words mean» (1975: 162). Given this understanding of the task, it might be doubted that the situation of the radical interpreter can be illuminated by reflecting on the very different situation in which a hearer observes what she regards as a sincere assertion by one of her colinguals. Doesn’t the fact that the hearer and speaker are colinguals in the latter case make this case very different from the situation of radical interpretation? This worry need not deter us. For one thing, I am less interested in the nature of radical interpretation than I am in the nature of the sort of interpretation that is involved in everyday speech exchanges. Davidson famously argued that «radical interpretation begins at home», but this point 7

Below I will characterize what this comes to; for now, I leave matters at this impressionistic level of description.

161

is far from obvious to me (see Goldberg 2004). On the contrary, given a group of individuals who «speak the same language» at least in the sense that verbal exchanges between them are typically smooth (and recourse for resolving any linguistic disputes are, broadly speaking, mutually acknowledged—see Goldberg 2007), it is no longer obvious that the best way to characterize the «task of interpretation» is in terms of the situation of the radical interpreter. It is this sort of everyday situation that I am describing here. But there is a second reason why the foregoing worry need not deter us: there is a good deal in common between the situation I am interested in, and the sort of situation Davidson described in connection with radical interpretation. Davidson assumed that, in contexts of radical interpretation, the interpreter H can discern that the speaker S has manifested an attitude of holding-true regarding a given sentence, even when H can’t identify the content towards which S manifests this attitude: [W]e can tell when a speaker holds a sentence to be true without knowing what he means by the sentence, or what beliefs he holds about its unknown subject matter, or what detailed intentions do or might prompt him to utter it (Davidson 1974a: 145).

Now it is at least arguable that a hearer’s discernment that an assertion has been made is independent of the hearer’s knowledge of what has been asserted, in precisely the way that a radical interpreter’s discernment that a speaker holds-true a given sentence is independent of the interpreter’s knowledge of what the speaker means by the sentence. Of course, assertion goes beyond holding-true: the former is a speech act, the latter is a sentence-directed attitude. But it would seem that in many, or perhaps even most,8 of the circumstances in which a speaker manifests her attitude of holding-true, she would (were she appropriately prompted) assert the con8

I say “most”, not “all”: it may be that there are cases where someone believes something, but doesn’t take herself to have the requisite epistemic authority to flat-out assert it. This possibility raises interesting questions concerning the relationship between what some have called the norm of belief (which specifies the conditions under which it is appropriate to believe something) and the norm of assertion. I cannot pursue these questions here.

162

tent in question.9 It would thus seem that our radical interpreter is in a position to discern assertions at roughly the same time that she is in a position to discern the sentences held-true by the interpretee. This said, I will focus exclusively on the case involving co-linguals: the points I wish to make about interpretation concern these cases. The question I seek to answer is this: given a case in which a hearer is warranted in thinking that a speaker has made a sincere assertion, how does this fact alone affect the interpretation of the speaker’s speech act?10 The thesis of the remainder of this paper can be put in terms of two claims. First, if H is warranted in regarding S as having made a sincere assertion, then H is warranted in regarding it as mutually familiar11 that S did something which was proper only if S satisfied some substantial epistemic condition vis-à-vis the asserted proposition. Call this claim the Epistemic Norm of Assertion hypothesis (or “ENA” for short). Second, ENA can be used to justify a methodological principle of interpretation that is an epistemicized version of the sort of principle often advanced in connection with radical interpretation (charity, humanity, what-have-you). After developing both of these claims in 3.1 and 3.2 respectively, I will go on, in sections 4 and 5, to draw out what I regard as the crucial implications in connection with the epistemological dimension of interpretation generally, and of triangulation in particular.

9

Of course, there are situations where a speaker who firmly believed something would refrain from asserting it, out of considerations of e.g. the desire for privacy, or politeness, etc.—but it would seem that in these circumstances she would also refrain from manifesting her attitude of holding-true. 10 I am primarily interested in cases involving the apprehension of sincere assertion; but a hearer’s judgment as to whether an observed assertion was sincere are not infallible. I will briefly discuss the complications that arise in this connection below. (The sincerity aspect of assertion does not differentiate the apprehension of assertion from the apprehension of the attitude of holding-true. A subject can be insincere in her expression of her attitudes, making as if she holds-true a given sentence when in fact she doesn’t.) 11 Mutual familiarity is in the same family as common knowledge, without requiring that the familiar content is known. I thank Gregory Ward for suggesting the use of this notion.

163

3.1. Let us focus first on the Epistemic Norm of Assertion hypothesis, ENA. This hypothesis involves two claims: first, that assertion has an epistemic norm; and second, that for this very reason hearers who discern that a sincere assertion has been made are entitled to make assumptions about what is mutually familiar in the speech situation. The first claim—the hypothesis that assertion has an epistemic norm— is familiar in the recent literature on assertion. Much of this literature assumes as much; attention is devoted instead to determining what the norm is. The question that theorists seek to answer here is: what epistemic credentials must a speaker have vis-à-vis the asserted proposition if her assertion is to be warranted? The two leading candidates are knowledge (Williamson 1996; Unger 1975; DeRose 1991, 1996, 2002; Hawthorne 2003; Stanley 2005; Adler 2002), and some variant on rational credibility or justification (Douven 2007; Lackey 2007; Kvanvig 2010). To be sure, not everyone endorses the hypothesis that assertion has an epistemic norm. Some agree that assertion has a norm, but deny that it is epistemic (see Weiner 2005, where it is argued that the relevant norm is truth). Others think that it is not helpful to think of assertion as having a norm that individuates it as a speech-act type in the first place (see MacFarlane 2011, for a description of several variants on this sort of position). Still, I think it is fair to say that an increasing number of people are coming around to the view that assertion has an epistemic norm. The claim that assertion has an epistemic norm requires that assertions are appropriately assessed by reference to whether the speaker satisfies the relevant epistemic standard vis-à-vis the content asserted. So, for example, defenders of the knowledge norm of assertion hold that S’s assertion that p (at time t) is appropriate, only if (at t) S knows that p. By “appropriate” here is meant “in keeping with the standard for speech acts of this sort”. Thus, according to defenders of the knowledge norm, S’s assertion that p is deficient qua assertion, and S is properly criticizable for performing the speech act she did, if S fails to know that p. This reflects what is perhaps the most controversial element in the discussion of the norm of assertion: those who maintain that assertion has a norm—epistemic or otherwise— maintain that this norm serves to individuate the speech act itself. This point is nicely introduced by Timothy Williamson:

164

What are the rules of assertion? An attractively simple suggestion is this. There is just one rule. Where C is a property of propositions, the rule says: (The C rule) One must: assert p only if p has C. […] The envisaged account takes the C rule to be individuating: necessarily, assertion is the unique speech act A whose unique rule is the C rule. […] All other norms for assertion are the joint outcome of the C rule and considerations not specific to assertion. If an assertion satisfies this rule, whatever derivative norms it violates, it is correct in a salient sense. Call this account the C account, and any account of this form simple (Williamson 2000: 241).

The hypothesis that assertion has an epistemic norm, then, is the result of the hypothesis that there is a simple C account of assertion, together with the claim that C is some epistemic relation that the speaker bears to the asserted proposition. Whereas ENA’s first claim is that assertion has an epistemic norm, its second claim, in effect, is that assertion’s having an epistemic norm is mutually familiar to everyone who participates in the practice (whether as a producer or a consumer of assertions). More specifically: if H is warranted in regarding S as having made an assertion, then H is warranted in regarding it as mutually familiar that S did something which was proper only if S satisfied some substantial epistemic condition vis-à-vis the asserted proposition. I submit that this claim can be established by reflecting on the status of the hypothesis that assertion has an epistemic norm. In particular, because this hypothesis purports to be a truth regarding the point and nature of the speech act of assertion, it is reasonable to think that this hypothesis itself (or something in its vicinity) is mutually familiar to speakers. There are two sorts of consideration that support this “mutual familiarity” contention. The first concerns common practice regarding assertion. It is common for those who question an assertion to challenge it by querying the speaker’s relevant epistemic authority. Thus it is common to challenge an assertion with the query, “How do you know that?”12 or “On what basis do you say that?” In both sorts of case, what is being queried or challenged is 12

Williamson (2000) uses this as one piece of evidence in support of the knowledge norm of assertion.

165

the propriety of the assertion. Speakers whose assertions are challenged in this way typically respond by citing their relevant epistemic authority (or else by retracting the assertion). What speakers don’t do is question the legitimacy of this sort of query. By contrast, consider how awkward it would be for a hearer to respond to an assertion that p with the retort, “But is it true that p?” A speaker whose assertion elicited this reaction might well protest, “Well, I just told you so, didn’t I?”—thereby in effect questioning the legitimacy of this sort of reply. The foregoing considerations—the naturalness of the responses querying the asserter’s epistemic position, and the unnaturalness of the response querying the truth of the asserted content—are naturally explained by the hypothesis that it is mutually familiar to speakers and hearers alike that assertion has an epistemic norm. To see this, suppose that this hypothesis is true. Then someone who asserts that p does something regarding which it is mutually familiar that what she did is appropriate only if she stands in some happy epistemic position vis-à-vis the truth of what she asserted. In that case, given what is mutually familiar, it is easy for all parties to appreciate that questions regarding the propriety of the assertion can take the form of questions regarding the speaker’s relevant epistemic credentials. And if all parties appreciate that such questions can be used to question the propriety of the assertion, then all parties understand that such questions are a legitimate way to reply to the making of an assertion. But now suppose that the “mutual familiarity” hypothesis is false. Then it would appear that we have several regularities in need of an explanation. Why it is that questions as to the propriety of an assertion typically take the form of questions regarding the speaker’s relevant epistemic authority? And why it is that speakers whose assertions are questioned in this way typically react to such queries, not by challenging the legitimacy of the query, but instead by accepting the burden placed on them (namely, they manifest their relevant epistemic authority, or else retract the assertion)? I have no doubt that candidate explanations can be found which do not make use of the mutual familiarity assumption; but I question whether these candidate explanations will hold up to scrutiny. The burden is on the proponent of such explanations: she must show that these explanations are as simple, and as wellmotivated, as the explanation provided by the mutual familiarity assumption.

166

The second consideration supporting the mutual familiarity assumption derives from speech act theory in general, and the theory of assertion in particular. It has been common in speech act theory to individuate the various speech-act types by reference to the “dimensions of variation” that differentiate speech-act types. (These were famously explored in Searle 1979.) What is more, it is typically, if only implicitly, assumed that the various features of a given speech act—the values of the various “dimensions” that serve to pick out acts of that type—are common knowledge (or at least mutually familiar to all mature speakers). This mutual familiarity assumption is very important to speech act theory. In particular, it is used to explain how arbitrary hearers discern the “speech act significance” of speech acts of the various types: not only can the hearer apprehend the speech act of the relevant type, but she can also discern relevant implicatures, and she can regard the speaker as recognizing that she (the hearer) will work out the relevant implicatures. Crucial to these tasks is the hearer’s reliance on her (perhaps implicit) representation of the factors that individuate the speech act types, and her (the hearer’s) assumption that these factors will be familiar to the speaker as well. Indeed, even those theorists who try to minimize the need for the hearer to do any “mind reading” in her recovery of implicatures (for which see Gauker 2001) appeal to such things as the hearer’s recognition of the force of the utterance, and the hearer’s (implicit) knowledge of what follows from the fact that the speaker performed a speech act with that force. But if we must assume that any arbitrary hearer would have this knowledge merely in virtue of being a competent speaker of the language, then any subject can reach the conclusion that this information is mutually familiar to the others in her language community, as she can reflect on her own role as hearer, and can recognize that she is no different from other hearers. It would appear, then, that it is a general feature of speech act theory that it assumes that the individuating features of the various speech acts are mutually familiar. 3.2. So far I have been defending ENA, a claim to the effect that if H is warranted in regarding S as having made an assertion, then H is warranted in regarding it as mutually familiar that S did something which was proper only if S satisfied some substantial epistemic condition vis-à-vis the asserted proposition. I now move on to the second of the claims I want to

167

make on this score: ENA can be used to justify an “epistemicized” methodological principle of interpretation. My basic reasoning on this score is this: by the lights of ENA, a hearer is entitled to make a certain assumption on apprehending another’s sincere assertion; and this assumption bears on the hearer’s attitude towards the interpretee’s epistemic position vis-à-vis the truth of the content asserted, in a way that can figure in the motivation of a constraint on interpretation. We can develop the point as follows. Confronting what he discerns to be an assertion by S, H is warranted (by the lights of ENA) in accepting the following: S did something which was warranted only if she (S) had the relevant epistemic authority. Of course, this does not entitle H to think that what S did was, in point of fact, warranted. What H is warranted in accepting (by ENA’s lights) is this: either S had the relevant epistemic authority, or else her assertion was unwarranted. But let us now suppose that hearers are in a position to tell when they are in the presence of a sincere assertion. (By “sincere” assertion I mean one in which the speaker does in fact aim to satisfy the norm peculiar to assertion—whether or not she succeeds in doing so.) Confronted with something that H is warranted in regarding as a sincere assertion by S, H would then be entitled to accept that the following is mutually familiar: in speaking as she did, S regarded herself as having the relevant epistemic authority. It goes without saying that S might not, in fact, have the relevant authority she regards herself as having. But the point remains: sincere assertion manifests a speaker’s selfregard as having the relevant authority, and a hearer who is warranted in his judgment that another speaker’s assertion was sincere is thereby warranted in thinking that the speaker so regarded herself. It is at this point, I submit, that we can motivate a thesis in the methodology of interpretation. To a first approximation, the relevant thesis is this (“ENAP” for the “Epistemic Norm of Assertion Principle”): ENAP: For any set of sincere assertions by another speaker, favor those interpretations which make the best sense of the speaker’s regarding herself as having been in an epistemically authoritative position vis-à-vis the propositions in question.

168

To get a sense of the strength of the constraint imposed by ENAP, we will have to address two questions. First: what is the epistemic norm of assertion? This will determine how we understand “epistemically authoritative” in ENAP; stronger methodological principles of interpretation result from more demanding epistemological properties. Second: do we want to acknowledge that a speaker can systematically fail in her attempts to satisfy the norm of assertion? This issue will determine whether ENAP merely requires us to make sense of the speaker’s self-conception as having the relevant epistemic authority, or requires instead that the speaker be seen as actually having this authority (for many or most occasions of sincere assertion). The significance of the first question can be illustrated by considering various proposals for the epistemic norm of assertion. If it is knowledge, then an interpreter who recognizes a speaker to have produced a sincere assertion must treat the speaker as regarding herself as having spoken from knowledge.13 If the norm of assertion is rational belief, then the interpreter must regard a sincere asserter as regarding herself as having rational backing for her assertion. By itself, however, neither result suffices to make clear how demanding ENAP itself is. For ENAP merely insists that the interpreter “make the best sense” of the speaker’s epistemic self-conception, and it is not yet clear what this amounts to. To clarify this, we must address the second question above. Suppose (for the sake of argument) that it’s possible for an arbitrary speaker to fail radically in her attempts to conform to the norm of assertion: it regularly or perhaps even always happens that she makes sincere assertions without having the relevant epistemic authority. Here the speaker’s own self-conception, as someone who possesses the relevant epistemic authority, is itself a delusion. In this case, ENAP’s demand—to “make the best sense of” the speaker’s self-conception as having the relevant authority—would not appear to yield a substantial basis for discrimi13

For an interesting argument in defense of a position in this vicinity, see Williamson (2000: 267, 2007: 264). It is worth noting, however, that Williamson’s argument does not derive from the norm of assertion, so much as it does on the conditions on reference: his claim is that reference-assignments are guided by the methodological principal enjoining us to maximize the speaker’s knowledge.

169

nating between rival interpretations. It is only if the speaker does regularly conform to the norm of assertion, and so (regularly) makes assertions only when she has the relevant authority, that ENAP’s constraint will have some bite. For in that case we would have to so interpret the speaker’s sincere assertions, that it turns out that she regularly has the relevant epistemic authority she takes herself to have regarding the asserted propositions. Davidson is famous for having argued that we have no choice but to see others as by and large speaking and believing truths—or at least what the interpreter herself regards as truths. Interestingly, given a very weak assumption, ENAP itself would appear to yield an argument for an epistemicized version of this conclusion. The weak assumption is one Davidson himself endorsed: interpreters have no choice but to approach the task of interpretation from their own perspective, that is, from the perspective of what they themselves take to be true. (See Davidson 1973: 137, 1974a: 152, 1974b: 196.) Suppose this assumption is true. Then in trying to make sense of another’s self-conception as someone who (on the various occasions of sincere assertion) has the relevant authority, the hearer/interpreter H has no choice but to put herself into the speaker S’s shoes, in order to discern what propositions are such that, so positioned, H would take himself to have the relevant epistemic authority regarding them. If all H has to go on here is his own background beliefs and the processes he uses to acquire new information, it would appear that H has no choice but to see S as in a position to know (rationally believe) the things H would take himself to be in a position to know (rationally believe), were he in S’s shoes. It looks, then, that insofar as H aims to interpret S’s assertions, he must regard S’s self-conception—that she is relevantly authoritative, on the occasion of her sincere assertions—as broadly correct. The point here is not that S’s self-conception must be correct; only that H himself has no choice but to assume this. My suggestion, then, is that when it comes to the interpretation of another’s assertoric utterances, we have no choice but to approach these by remaining sensitive to the speaker’s own epistemic perspective. In particular, the task of interpreting another’s sincere assertions requires the interpreter to engage in a decidedly epistemic activity: the interpreter must assess the speaker S’s performances from the perspective of S’s own self-

170

conception as relevantly epistemically authoritative. In doing so the interpreter regards the interpretee as involved in an activity that cannot be understood in wholly non-epistemic terms. Where Davidson speaks of the need to «[assign] truth conditions to alien sentences that make native speakers right when plausibly possible, according […] to our own view of what is right» (1973: 137), I would suggest that we speak instead of the need to assign truth conditions to sentence so as to make speakers relevantly (epistemically) authoritative whenever plausible. In saying this I am assuming that a speaker’s sincere assertions cannot massively fail to comply with the norm of assertion; if so, we must see another’s sincere assertions not merely as manifesting the speaker’s self-conception as relevantly authoritative, but as by and large manifesting that very authoritativeness itself. 4. In discussing the philosophical significance of “the second person” (see Davidson 1992) in the triangulation-constituting triangle, Davidson appears to acknowledge the mutuality of the recognition of the other’s status as epistemic subject. I quote Davidson at some length: [I]f someone is the speaker of a language, there must be another sentient being whose innate similarity responses are sufficiently like his own to provide an answer to the question, what is the stimulus to which the speaker is responding? […] [I]f the speaker’s responses are linguistic, they must be knowingly and intentionally responses to specific stimuli. The speaker must have the concept of the stimulus—of the bell, or of tables. Since the bell or a table is identified only by the intersection of two (or more) sets of similarity responses (lines of thought, we might almost say), to have the concept of a table or bell is to recognize the existence of a triangle, one apex of which is oneself, the second apex another creature similar to oneself, and the third an object (table or bell) located in a space thus made common. The only way of knowing that the second apex of the triangle—the second creature or person—is reacting to the same object as oneself is to know that the other person has the same object in mind. But then the second person must also know that the first person constitutes an apex of the same triangle another apex of which the second person occupies. For two people to know of each other that they are so related, that their thoughts are so related, re-

171

quires that they be in communication. Each of them must speak to the other and be understood by the other (Davidson 1992: 120-21).

In light of these very suggestive comments, we can ask whether they express Davidson’s awareness of the relevance to interpretation of the role others play as sources of knowledge. I am not sure about this. For Davidson, does knowing that “the second person” has the same object in mind require one to know that the second person is a subject who aims to acquire knowledge of the object in question? Or does it suffice that the second person has this epistemic aim—i.e., that she responds “knowingly and intentionally” to the stimulus—whether or not one oneself represents this fact about the second person? The matter is unclear from the quote above. It might be helpful to take a situation constituting the sort of triangle Davidson is describing here. In virtue of what does one subject, S1, regard the other subject, S2, as “having the same object in mind” as S1 herself does? The picture encouraged by the quote above is this: S1 need only trace out the relevant “lines of thought”—one from S1’s own line of sight to the world, the other from S2’s line of sight to the world—and see where these lines overlap. But why should S1 suppose that S2’s gaze identifies what it is she (S2) is thinking about? Even if S1 knows that S2’s “similarity space” is similar to S1’s own, this needn’t settle the matter: S2 need not be attending to the object(s) at which her (S2’s) gaze is fixed. Of course it would be a strange hypothesis indeed to suppose that a given creature whose gaze is fixed in the direction of a salient object is not, in fact, attending to this object. (This might happen on occasion; but typically it does not.) But what makes this hypothesis strange? Is it merely the fact that people with fixed gazes typically attend to the salient object in their “line of sight”? I submit that this is not the deepest explanation we can offer for why it is proper to suppose that one’s attention is on the salient objects in one’s line of sight. The deepest explanation appeals to one’s status as a seeker of knowledge. On the twin assumptions, first, that S2 is a seeker of knowledge, and second, that perception is the source of a good deal of our empirical knowledge, it makes good sense for S1 to regard S2 as fixing on the object determined by her gaze: for this maximizes the chance that S2 is (seen by S1 as) relevantly knowledgeable. The reason it is strange to suppose, as one’s initial assumption, that a speaker is not attending to the ob-

172

ject(s) or property/ies on which her gaze is fixed, is simply that such a supposition fails to square with the idea that the subject under interpretation is an epistemic subject, one whose interactions with the world are largely shaped by her epistemic aim to acquire knowledge of that world. It is this, I submit, that is underemphasized by Davidson’s reflections on triangulation. It may well be that Davidson had this sort of picture in mind all along. Nevertheless, the point is not as explicit in his thinking as one would have liked. Other examples can be found illustrating Davidson’s underemphasis of the importance to interpretation of the interpretee’s status as an epistemic subject. Return for example to Davidson’s official description of the procedure of triangulation, cited at the outset of this paper: each of two people is reacting differentially to sensory stimuli streaming in from a certain direction. Projecting the incoming lines outward, the common cause is at their intersection. If the two people now note each other’s reactions (in the case of language, verbal reactions), each can correlate these observed reactions with his or her stimuli from the world. A common cause has been determined. The triangle which gives content to thought and speech is complete. (2000: 212-213)

Davidson’s picture here seems to be this. Each participant occupies a position in which she can correlate (i) another’s speech with (ii) the stimulation conditions that prompted that speech. On the basis of these correlations, each participant can confirm the similarity of their respective reactions, and from there move to hypothesize a “common cause” by tracing the relevant “lines of thought”. In this way the interpreter regards the (distal) stimulation conditions as the cause of the other’s “verbal reactions”, just as it is the cause of his (the interpreter’s) own perceptual experience when he turns his own gaze in that direction. But this picture, it seems to me, fails to mention a crucial element in the task of interpretation. In interpreting another’s speech, one recognizes that the speaker’s “verbal reactions”—or at least that subset of them that are sincere assertions—are not just any old worldly event, to be correlated with other worldly events; they are rather a linguistic manifestation of the other speaker’s would-be knowledge of the very world that the interpreter uses in the course of triangulation. Triangulation, I submit, is not a merely causal affair; it is a causal-epistemic affair.

173

The “routes” that are traced from subject to subject, and from subject to world, are themselves routes to knowledge (or at least epistemically authoritative belief). The hearer must see the speaker not merely as having been causally affected by the “common cause”, but rather as occupying a position, in virtue of that causal connection, from which to acquire epistemically authoritative belief regarding the “common cause”. I doubt very much whether any of this would have been news to Davidson. After all, in his later discussions Davidson appears to appreciate that each subject recognizes the other as a subject at least some of whose reactions to the world are manifestations of her knowledge of it. My present claim is merely that at times Davidson’s “official description” of the process of triangulation, suffused as it was by the language of (mere) causation and correlation, obscures this insight. 5. I want to conclude by making clear how this insight affects our understanding of the methodology of interpretation. To do so I want to address a question not often addressed: what is “charitable” about this-or-that methodological principle of interpretation? Whereas Davidson appears to have been satisfied to regard the charitableness of his Principle of Charity to consist in the idea that it imposes a requirement to maximize others’ truths (in belief and in speech),14 the present proposal can explain the charitableness of ENAP as resting on two more fundamental points. First, the charitableness of ENAP derives from the fact that the favored interpretation vindicates the speaker’s own self-conception as a subject who succeeds at acquiring and communicating epistemically authoritative beliefs. (Or, at the very least, my proposal requires that the hearer endorse this aspect of the speaker’s self-conception.) Second, the charitableness of the interpretations sanctioned by ENAP can be traced to the epistemic perspective of the subject under interpretation. Thus, where Davidson was satisfied to regard the charitableness of his Principle of Charity to consist in the idea that it imposes a requirement to maximize others’ truths (in belief 14

In asking this question I am going to be bracketing skeptical scenarios in which the best that can be hoped for in interpretation is agreement between speaker and hearer, not the truth of the beliefs that constitute that agreement.

174

and in speech), the present proposal sees any truth-related implications of the method as deriving from its imposition of epistemic criteria—namely, those implicated in the epistemic norm of assertion. The point is most obvious on the assumption that knowledge is the norm of assertion. On this assumption, an interpretative method which would have us regard sincere asserters as often satisfying the norm of assertion is an interpretative method which would have us regard sincere asserters as often speaking knowledgeably—and so, since knowledge implies truth, as speaking truly. But the point can also be made even if the epistemic norm of assertion is something that falls short of knowledge: justification, say. So long as we understand the relevant epistemic term in truth-related ways—so long as a belief’s (or an assertion’s) being justified is understood in terms that make it more likely that the belief/assertion is true15—we can make the same point. In short, the present proposal can explain why the interpretations sanctioned by ENAP will be interpretations on which a good number of the speaker’s assertions come out true. It is not merely that this is the only outcome compatible with the intelligibility of the speaker; it is also the only outcome compatible with preserving the speaker’s self-conception as relevantly epistemically authoritative—a self-conception the speaker herself manifests in the course of making a sincere assertion. References Adler, J. 2002, Belief’s Own Ethics. Cambridge: MIT Press. Coady, C. 1992, Testimony: A Philosophical Study. Oxford: Clarendon Press. Davidson, D. 1969, “On Saying That”, repr. in Davidson 1984. Davidson, D. 1973, “Radical Interpretation”, repr. in Davidson 1984.

15

This is certainly true on any externalist understanding of epistemic justification, but it is also true on most (though perhaps not all) internalist understandings as well. (Admittedly, the connection between internalist justification and truth is looser than is the connection between externalist justification and truth—but a connection remains, at least on most internalist conceptions.)

175

Davidson, D. 1974a, “Belief and the Basis of Meaning”, repr. in Davidson 1984. Davidson, D. 1974b, “On the Very Idea of a Conceptual Scheme”, repr. in Davidson 1984. Davidson, D. 1975, “Thought and Talk”, repr. in Davidson 1984. Davidson, D. 1984, Inquiries into Truth and Interpretation. Oxford: Oxford University Press. Davidson, D. 1991, “Three Varieties of Knowledge”, repr. in Davidson 2001. Davidson, D. 1992, “The Second Person”, repr. in Davidson 2001. Davidson, D. 1999, “The Emergence of Thought”, repr. in Davidson 2001. Davidson, D. 2001, Subjective, Intersubjective, Objective. Oxford: Oxford University Press. DeRose, K. 1991, “Epistemic Possibilities”, Philosophical Review, 100, 581605. DeRose, K. 1996, “Knowledge, Assertion, and Lotteries”, Australasian Journal of Philosophy, 74, 568-580. DeRose, K. 2002, “Assertion, Knowledge, and Context”, Philosophical Review, 111, 167-203. Douven, I. 2006, “Assertion, Knowledge, and Rational Credibility”, Philosophical Review, 115, 449-485. Goldberg, S.C. 2004, “Radical Interpretation, Understanding, and Testimonial Transmission”, Synthese, 138 (3), 387-416. Goldberg, S.C. 2007, Anti-Individualism: Mind and Language, Knowledge and Justification. Cambridge: Cambridge University Press. Gauker, C. 2001, “Situated Inference versus Conversational Implicature”, Noûs, 35, 163-189. Graham, P. 2000, “The Reliability of Testimony”, Philosophy and Phenomenological Research, 61: 695-709. Hawthorne, J. 2003, Knowledge and Lotteries. Oxford: Oxford University Press.

176

Kvanvig, J. 2010, “Assertion, Knowledge and Lotteries”, in D. Pritchard and P. Greenough (eds.), Williamson on Knowledge. Oxford: Oxford University Press. Lackey, J. 2007, “Norms of Assertion”. Nous, 41, 594-626. MacFarlane, J. 2011, “What is Assertion?”, in J. Brown and H. Cappelen (eds.), Assertion: New Philosophical Essays. Oxford: Oxford University Press, pp. 79-96. Searle, J. 1979, “A Taxonomy of Illucutionary Acts”, in his Expression and Meaning. Cambridge: Cambridge University Press. Stanley, J. 2005, Knowledge and Practical Interests. Oxford: Oxford University Press. Unger, P. 1975, Ignorance: A Case for Skepticism. Oxford: Clarendon Press. Weiner, M. 2005, “Must We Know What We Say?”, Philosophical Review, 114, 227-251. Williamson, T. 1996, “Knowing and Asserting”, Philosophical Review, 105, 489-523. Williamson, T. 2000: Knowledge and Its Limits. Oxford: Oxford University Press. Williamson, T. 2005, “Contextualism, Subject-Sensitive Invariantism, and Knowledge of Knowledge”, Philosophical Quarterly, 55, 213-35. Williamson, T. 2007, The Philosophy of Philosophy. London: Blackwell.

PART III PHILOSOPHICAL GEOGRAPHY

THE SHORT HAPPY LIFE OF THE SWAMPMAN: INTERPRETATION AND SOCIAL EXTERNALISM IN DAVIDSON Mario DE CARO (Roma Tre University, Italy)

What makes interpretation possible […] is the fact that we can dismiss a priori the chance of massive error. (Davidson 1975: 169).

In Donald Davidson: Meaning, Truth, Language, and Reality (2005), Ernest Lepore and Kirk Ludwig have offered one of the most complete and thoughtful discussions of Davidson’s views about truth-theoretical semantics, radical interpretation, indeterminacy, extensionalism, and truth. In my opinion, however, Lepore and Ludwig’s presentation of Davidson’s externalism is partially inadequate, particularly as to the claim that Davidson defended a pure form of physical externalism about mental content. In this paper it will be argued that at the beginning Davidson advocated a mixed form of externalism that, besides a strong physical variety of that view, also encompassed a peculiar form of social externalism, according to which interpretability offers necessary and sufficient conditions for the attribution of contents to the interpreted speakers. As it will be noted, however, the equilibrium of these two strands of Davidson’s externalism was unavoidably unstable—and this explains why in his last years he gave up the strong version of physical externalism he had incorporated in his view at the beginning. 1. Varieties of Semantic Externalism Semantic externalism (or content externalism) is based on the idea that the relation between the thinkers and the environment they inhabit is necessary for constituting the content of at least some of their thoughts. Davidson presented his own version of externalism, which he calls “triangular externalism”, with the following words,

180

It takes two points of view to give a location to the cause of a thought, and thus to define its content [...]. If the two people [...] note each other’s reactions (in the case of language, verbal reactions), each can correlate these observed reactions with his or her stimuli from the world. The common cause can now determine the content of an utterance and a thought. The triangle which gives contents to thought and speech is complete (Davidson 1991: 212-213).

In order to understand and evaluate Davidson’s proposal, it is useful to see how it locates in a taxonomy of the versions of semantic externalism. First of all, the various versions of this view can be distinguished depending on the pervasiveness they attribute to the phenomenon of the externalist determination of content. This phenomenon can be seen either as limited to a few classes of expressions (proper names, natural kind words, and/or indexicals) or as an-across-the-board semantic feature.1 Davidson vigorously defended the latter version. The dependence of meaning on factors outside the head [...] is ubiquitous, since it is inseparable from the social character of language. [...] [I]t is a perfectly general fact about the nature of thought and speech (1988: 47-48).

Another important distinction is that between physical and social externalism. According to the first option, mental content is essentially determined by the physical relations between the subject, on the one hand, and the non-social objects and events in the external world, on the other hand.2 According to the second view, social factors play a crucial role in the determination of mental contents. In their presentation of Davidson’s philosophy of language, epistemology and philosophy of mind, Ernest Lepore and Kirk Ludwig have presented Davidson’s externalism as a form of physical externalism. In their opinion, such a view is actually incompatible with social externalism since, they write, «Davidson’s route to his externalist position would, if correct, […] undermine social externalism» (2005: 337). In my view, however, Lepore and Ludwig’s view is only partially correct. More specifically, it is true that: a) at the beginning Davidson de1 2

For this distinction, see Bilgrami (1992), passim. The loci classici for this view are Kripke (1972), Putnam (1975).

181

fended a form of physical externalism, and b) that form of externalism was actually at odds with social externalism. However, it has to be added that Davidson’s view also incorporated a peculiar form of social externalism— which we could call “interpretive social externalism”—that was not fully consistent with the physical strand of triangular externalism. As we will see, Davidson became more and more aware of the internal tension of his hybrid externalist view, as it shown by his unstable treatment of the famous thought experiment centered on the “Swampman”. The final step of that process was that in his very last years Davidson ended up privileging the interpretive social strand of his externalism over its physical component. Finally, externalist theories can be either synchronic or diachronic in character. That is, an externalist may claim that thought contents depend either on the historical or on the current relationship between a speaker and the environment she lives in. As we will see, in the first period Davidson advocated the diachronic view; however, when he realized that that view was in tension with interpretive social externalism, he switched to the synchronic view. 2. Davidson and the Classic Forms of Social Externalism Davidson rejected all classic form of social externalism—that is, those defended by Burge, Wittgenstein (or Kripkestein?), and Putnam—for two main reasons. First, he refused the idea that what a person means and thinks may depend on a social norm of which the person can be completely unaware. Second, he thought (correctly) that those views were at odds with the kind of physical externalism he endorsed. In the next paragraph, we will discuss Davidson’s peculiar version of social externalism, “interpretative social externalism”. As we will see, however, whereas that view was not touched by the first of the above mentioned problems, it was affected by the second. In general, Davidson did not think that in interpreting somebody—that is, in trying to understand what a speaker says and to attribute thoughts to him or her—the interpreter has to refer to socially shared norms of which that speaker may be unaware. In this light, let’s consider Davidson’s criti-

182

cisms of traditional social externalist views more in detail, beginning with his discussion of Tyler Burge’s views. Davidson discusses two forms of externalism that he attributes to Burge. The first is called “perceptual externalism” (Burge 1986a, 1986c, 1998), and claims that «the contents of utterances and thoughts depend on the causal history of the individual, particularly in connection with perception» (Davidson 1990a: 198). The second form of externalism defended by Burge (1990, 1982b, 1986c) is a classic version of social externalism, which Davidson describes in this way: «The meanings of a person’s words, and the contents of that person’s thoughts, depend in part on the linguistic practices of the person’s community, even in cases where the individual is mistaken about the relevant practices» (1990a: 198). While Davidson (1990a: 201) endorsed perceptual externalism, as advocated by Burge, he strongly rebutted the latter: I have rejected social externalism as championed by Burge. Do I therefore think social factors play no role in the externalism of the mental? Not at all. But I would introduce the social factor in a way that connects it directly with perceptual externalism, thus locating the role of society within the causal nexus that includes the interplay between persons and the rest of nature (Davidson 1990a: 200).

Burge has defended his version of social externalism by imagining a variation of Putnam’s celebrated Twin Earth thought experiment. In Burge’s scenario what varies is only the social, not the physical scenario. Let’s imagine that a subject, let’s call her Jane, has some true beliefs concerning arthritis (she thinks that its incidence depends on the age; that it may have a rheumatoid origin, etc.). However, she also falsely believes that arthritis may be developed in the thigh and that she actually has that. Let now imagine a Twin Earth in which, besides referring to articular pain, the word “arthritis” is also normally used to refer to thigh pain (in that word there is no word for what we call “arthritis”). As Burge puts it: The counterfactuality in the supposition touches only the patient’s social environment […] In our imagined case, physicians, lexicographers, and informed laymen apply “arthritis” not only to arthritis but to various other rheumatoid ailments. The standard use of the term is to be conceived to encompass the patient’s actual misuse (1979: 105).

183

In the counterfactual scenario, Jane’s counterpart—who has Jane’s same causal history and all the same non-intentional mental phenomena—thinks she has arthritis to her thigh. However, the ascriptions of content to the patients in the actual and in the counterfactual world have different extensions—as it is proven by the fact that in the actual case Jane’s belief “I have arthritis” (referring to thigh pain) is false, while Jane counterpart’s belief “I have arthritis” is true. The crucial point, Burge argues, is that the differences in the mental contents and in the meanings of the words in the actual and in the counterfactual scenario cannot depend on the physical environment or on the non-intentional mental states of the subjects in the actual and in the counterfactual scenario (since those are identical by definition). So, according to Burge, the differences in the mental contents of the two speakers have to depend on the diversity of the respective sociolinguistic environments. Davidson (1990a) rejects Burge’s social externalism for three reasons. The first is Davidson’s well known claim that in our communicative practices the idiolect has epistemic primacy on the common, socially-shared language. What really matters when we interpret a speaker is interpreting her in the way she wants to be interpreted—whatever deviant uses of words and concepts may be involved in that speaker’s speeches and thoughts. The second reason for which Davidson rejects Burge’s social externalism is that that view—by insisting that communal standards of correctness are prior to individual ones—is obviously in conflict with first person authority, as long as it «ties a speaker’s meaning to an élite usage he may be unaware of» (1990a: 199). Indeed, it was one of Davidson’s most well-known opinions that a correct epistemology and a correct theory of communication have to presuppose that the interpreted speakers know what they mean with their own words. Unless there is a presumption that the speaker knows what she means, i.e. is getting her own language right, there would be nothing for an interpreter to interpret. To put the matter in another way, nothing could count as someone regularly misapplying her own words (Davidson 1987: 38).

The third reason for which Davidson refuses Burge’s social externalism, if less discussed, is very interesting for the purposes of this article. Davidson claimed that he had «a general distrust of thought experiments that pretend

184

to reveal what we would say under conditions that in fact never arise. My version of externalism depend on what I to be our actual practice» (1990a: 199). And of course Burge’s social externalism was based on the mentioned variation of Putnam’s Twin Earth thought experiment; so Davidson’s objection is, at the very least, not obviously out of target. What is surprising, however, is that until his very last years of life, also Davidson’s externalism based a good part of its appeal on two very peculiar thought experiments (the Swampman and the Omniscient Interpreter) that undoubtedly were very remote from “our actual practice”. On this more, however, in the next section. In order to understand the way in which, according to Davidson, social phenomena (particularly, those related to interpretation) do contribute to the determination of mental contents and meanings, it is interesting to contrast Davidson’s proposal with Wittgenstein’s views or, more precisely, with the views that Saul Kripke (1982) attributed to Wittgenstein.3 First of all, while in the Wittgensteinian tradition the term “social” referred to a structured community of speakers who are bound by a common language and share a “form of life”, in Davidson’s use the term “social” peculiarly referred to a very basic intersubjective context, in which a “triangulation” can take place because two, or more, subjects can causally interact with each other, and with a shared environment. Then, as to the similarities, we can notice that both Davidson and Wittgenstein resolutely refute Cartesianism and advocate externalist views of thought and meaning; that both defend holistic conceptions of mental content; that both Wittgenstein and Davidson—notwithstanding their respective unequivocal criticisms of internalism—recognize that there is an important asymmetry between the way in which we know our own thoughts, and the other forms of our knowledge. However, Davidson refuses the main tenet of Wittgenstein’s later philosophy of language and mind—that is, the fundamental normative function of social rules and shared languages practices in the determination 3

On Kripke’s skeptical interpretation of the Wittgensteinian notion of “following a rule”, see Davidson 1992. Since it is controversial if Wittgenstein really endorsed those views and Kripke himself did not endorse explicitly those views, they are sometimes attributed to the fictitious philosopher Kripkestein. With this caveat, in the following I will attribute those views to Wittgenstein.

185

of mental content. According to Davidson, all we need to explain the supposedly puzzling Wittgensteinian notion of “following a rule” is simply the idea of the “joint expectations” of the speaker and the interpreter: «the hearer expects the speaker to go on as he did before; the speaker expects the hearer to go on as before» (Davidson 1994a: 233). Hilary Putnam is the last philosopher whose social externalist conception will be considered in this section. Beyond offering, with the Twin Earth scenario, the Gedankenexperiment that started the “externalist revolution” (to paraphrase John Heil),4 Hilary Putnam also advocated social externalism, in particular with the thesis of the “linguistic division of labor”. In “The meaning of ‘meaning’” (1975), Putnam wrote that: We could hardly use words such as ‘elm’ and ‘aluminum’ if no one possessed a way of recognizing elm trees; but not everyone to whom the distinction is important has to be able to make the distinction […] The features that are generally thought to be present in connection with a general name—necessary and sufficient conditions for membership in the extensions, ways of recognizing if something is in the extension (‘criteria’), etc.—are all present in the linguistic community considered as a collective body; but that collective body divides the ‘labor’ of knowing and employing these various parts of the ‘meaning’ of [the general name] (Putnam 1975: 227-228).

On this basis, Putnam formulated his “Hypothesis of the universality of the division of linguistic labor”, according to which in all natural languages there are terms whose criteria of correct application are known only by lit-

4

«The externalist conception of mind is, I believe, nothing short of revolutionary. It is, arguably, the philosophical contribution of the latter half of the twentieth century» (Heil 1992: 24). It is worth noticing that in a footnote Heil mentions an interesting claim by Tyler Burge according to whom semantic externalism was already implied in Hegel’s philosophy, because of its «preoccupation with the role of social institutions in shaping the individual and the content of his thought» (1979: 100). What is referred to in this context is of course social externalism. At any rate, in referring to debate on skepticism, Davidson wrote that «the fallout [from the raise of externalism] […] is revolutionary» (1988: 45).

186

tle subcommunities of experts on whose judgment the wider community in that regard depends. In Putnam’s words, Every linguistic community […] possesses at least some terms whose associated ‘criteria’ are known only to a subset of the speakers who acquire the terms, and whose use by the other speakers depends upon a structured cooperation between them and the speakers in the relevant subset (1975: 229).

The externalist nature of the phenomenon the “linguistic division of labor” is obvious. No internal mental states of an individual speaker may fix the extension of a general name like “elm” or “gold”, since this can only be done by the “sociolinguistic state of the collective linguistic body” of the linguistic community (Putnam 1975: 230). Davidson (1994b) offered two replies to the argument that supports Putnam’s version of social externalism. First, he claimed that that argument may at most prove that, with regard to some general names, a speaker has to believe that there are experts that possess the ability to determine the reference of some general names, such as “elm” and “gold”. This reply, however, is not entirely convincing. Can we really imagine that no expert exist that can really distinguish between elms and beeches, or between gold and pinchbeck? Wouldn’t this scenario smell as a form of skepticism that Davidson himself would certainly refuse? But Davidson develops also a second objection to Putnam. He claims that the social division of labor can be in effect only after that some other much more general ways of determining the reference of the words are in place—that is, once a successful communication patterns has been established and a linguistic community has been formed.5 This claim is extremely plausible, and probably even Putnam would approve it. In fact, Davidson does not intend to prove that the “linguistic division of labor” is not a real phenomenon; he only wants to emphasize the conceptual and epistemical priorities in the relationship between language and thought, on one side, and the world, on the other side. From Davidson’s point of view, what really matters is the triangular process that involves two speakers that 5

For the linguistic division of labor to appear, a linguistic community has of course already to be in place in which people can communicate with each other and, among other things, can determine who the experts in some fields are.

187

interpret each other (and intend to be correctly interpreted) and a shared environment. Compared with this constitutive condition, the conceptual relevance of a phenomenon such as the linguistic division of labor cannot be very high, and according to Davidson it certainly is subordinate to the constitutive condition. 3. Externalism and Interpretation It cannot be doubted that Davidson sympathized with physical externalism. In Davidson (1988: 44), for example, he wrote that, «The correct interpretation of what a speaker means is not determined solely by what is in his head; it depends also on the natural history of what is in the head»—that is, it depends on the history of the causal interactions of the subject with the physical world. In advocating physical externalism, Davidson offered two main arguments based on thoughts experiments concerning two famous Davidsonian philosophical fictions: the Omniscient Interpreter (an interpreter that knows all potential evidence about what an interpreted speaker thinks and means) and the Swampman (a perfect physical duplicate of Davidson himself, at the age of seventy or so, instantaneously created ex nihilo, while the real Davidson disappears). Lepore and Ludwig (2005: 323-332) have offered an excellent summary of the Omniscient Interpreter Argument (originally in Davidson 1983: 150-151). They present it as follows: (1) Necessarily, for us to interpret correctly other speakers, we have to find those speakers largely in agreement with ourselves; (2) It is possible that those speakers be interpreted correctly (other things being equal) by an omniscient interpreter. (3) Therefore, those speakers must have largely true beliefs, i.e. they cannot be in massive error about the world. If this conclusion is correct, the externalist thesis is proved—that is, at least some of our thoughts are shown to be correctly connected to the world (and this proves that Cartesian skepticism is in error). The story of the Swampman is more complicated. We are told that this perfect physical duplicate of Davidson behaves exactly like Davidson, so

188

that the Swampman seems to fit perfectly our standards of interpretability; and as a consequence, people mistake him for the real Davidson. But the Swampman was born instantaneously and ex nihilo: so by definition he had no past adequate causal interaction with the external environment.6 Whereas from a physical point of view he perfectly resembled Davidson, the causal history of his relations with the environment made his mind very different from Davidson’s mind: more exactly—his creator claimed— because of his total lack of relations with the external environment, the Swampman’s words and thoughts did not have any content—even if, of course, he could acquire them in the future, by properly interacting with the environment and the human beings. In Davidson’s words, My replica can’t recognize my friends; it can’t re-cognize anything, since it never cognized anything in the first place. It can’t know my friends’ names (though of course it seems to), it can’t remember my house. It can’t mean what I do by the word “house”, for example, since the sound “house” it makes was not learned in a context that would give it the right meaning—or any meaning at all. Indeed, I don’t see how my replica can be said to mean anything by the sounds it makes, nor to have any thought (1987: 19).

Some years later, Davidson added: In one word, the Swampman’s trouble was that his brain stored nothing that could connect his sounds with the world, nothing to give his thoughts and words a semantics, a content (1999: 35).

Therefore, when Davidson introduced the Swampman in the philosophical arena, he thought that, even if all the normal interpreters would have attributed to him contents and meaning, he would have indeed been thoughtless and meaningless; and so he had to be interpreted, if the interpretations had to be correct. However, interpreting correctly the Swampman was only possible for someone like the Omniscient Interpreter—or at least for someone who could know things that the normal interpreters could not 6

For reasons mysterious to me, in referring to the Swampman, Davidson used the neutral pronoun “it”. However, even granting that the Swampman had no thoughts, it is unclear why the creature should have no sex – considering that from a physical point of view he was identical to Davidson ex hypothesi.

189

know, since as the story is told, nobody was around when the Swampman came to light. The moral of this scenario is clear. In determining whether and what a speaker thinks, the “natural history” of that speaker’s causal connections with the physical environment has absolute priority on our actual interpretive practices. To state it differently: in case of conflict over the attributions of content to a speaker (as it happens with the Swampman), the fact that the normal interpreters can interpret that speaker in a way that seems largely correct is irrelevant. Indeed, an interpretation can tell us whether a speaker think and what she thinks only if that interpretation fits the history of that speaker’s causal interactions with the physical environment. And this is true also when all normal interpreters would irretrievably be unaware of the evidence that is necessary for the correct interpretation of a speaker—since that evidence would only be accessible to an omniscient interpreter or someone comparably endowed (one who knows all the relevant physical facts of the past). Now, if this were the end of the story, Lepore and Ludwig would have been correct in claiming that Davidson endorsed a version of physical externalism. But that was not the end of the story. Some very important components of Davidson’s philosophical system, in fact, pulled in a very different direction. In “Inscrutability of Reference”, for example, he wrote that The semantic features of language are public feature. What no one can, in the nature of the case, figure out from the totality of the relevant evidence cannot be part of meaning (1979: 235).

And in “The Structure and Content of Truth”, Davidson presented this view again: Meaning is entirely determined by observable behavior, even readily observable behavior. That meanings are decipherable is not a matter of luck; public availability is a constitutive aspect of language (1990b: 314).

What Davidson claims here is that no alleged part of meaning can be in principle opaque to interpretation. In this way, he was accepting the “Nothing is hidden” view, as Normam Malcolm famously summarized it, an idea that came from Wittgenstein. In this light, “public availability” and “ob-

190

servable behavior” entirely determine semantic content. But it should be noted that, in this context, availability and observability refer to normal interpreters, not to philosophical oddities such as the Omniscient Interpreter. So at the same time the normal interpreters should interpret the Swampman as if he had no thoughts; on the other side, this interpretation would be unreachable for the normal interpreters. It is important to notice that Davidson was trying to find the sufficient conditions of a theory of interpretations for normal interpreters like us.7 That this is what Davidson had in mind becomes clear, for example, if we consider one of his most famous arguments, that concerning the so-called “dualism of scheme and content”—i.e., the alleged dichotomy between an unconceptualized “given” (the empirical content) and a conceptual scheme by which the given is conceptualized and organized (see Davidson 1974). Let’s therefore consider this argument. Its premises are:8 1) The notions of conceptual scheme and empirical content are complementary—they stand or fall together; 2) One conceptual scheme is possible only if a plurality of alternative (incommensurable) conceptual schemes is possible; 3) Conceptual schemes are necessarily associated with languages (a consequence of the Davidsonian thesis that thought requires language); 4) If two schemes are reciprocally incommensurable, then the languages with which they are respectively associated will be mutually untranslatable. From these premises, it follows that in order to refute scheme-content dualism, it is sufficient to prove that there cannot be mutually untranslatable languages. Mutual untranslatability—if there is such a thing—could be partial or total. But the latter, according to Davidson, is inconceivable: in fact, if the language spoken by an alien creature were totally incomprehensible for us, how could we know that this creature speaks a language at all? And then how could we attribute thoughts to it? Vice versa, how could we know that a translation we offer of the utterances of a particular speaker is 7

See De Caro (1999b: 3-4). In this paragraph I follow of the reconstruction of the argument I gave in De Caro (1999: 6-7), with which Davidson himself was happy (as he said in a personal email).

8

191

radically wrong, if no evidence could ever show that to us? So the idea of complete untranslability does not make sense. The alternative hypothesis of a partially untranslatable language is on the contrary legitimate, but it does not imply that incommensurable schemes are possible (this part of the argument is not relevant here). So, neither partial nor total untranslability can be regarded as adequate foundation for the notion of incommensurable schemes; therefore, this notion vanishes, and so does the whole conceptual system connected with scheme-content dualism. This reasoning is easily adaptable to the case of the Swampman. Claiming that his thoughts have no content and his utterances no meaning even if no normal interpreter could ever get to know that, amounts to say the correct interpretation of the Swampman would be impossible for the normal interpreters, since he would have no conceptual scheme at all. So the normal interpretations of the Swampman would be based on an extreme case of complete untranslatability: whereas the normal interpreters would translate his utterances with apparently no difficulty, in reality there would be nothing at all to translate—and the normal interpreters could never know that, since behavioristically the Swampan would be completely indistinguishable from Davidson (that is, from a being that we could interpret correctly). But the possibility of complete untranslability is one that Davidson did not accept. If we look at Davidson’s theory of interpretation we can find another reason for the idea that we should interpret the Swampman as a thinking creature. This is because otherwise we would be violating the principle of charity. Since can have absolutely no evidence that the Swampman is thoughtless, what reasons could we have for not being charitable with him as we are with all other speakers? (It is true that sometimes we find out that someone to whom we were attributing thoughts is completely out of his mind; so in that case we stop applying to the person the principle of charity and we withdraw our previous interpretation and claim that the person is not really thinking. But this process is based on the availability of new relevant evidence that in the case of the Swampman is impossible in principle). Summarizing, for Davidson’s physical externalist view, the normal interpreters should interpret the Swampman as if he had no thoughts. But according to the Davidsonian theory of interpretation, the Swampman would

192

be perfectly interpretable as he was thinking and speaking. This shows that in Davidson’s early version of externalism there was a remarkable tension between physical externalism and the his own theory of interpretation, according to which the Swampman fulfills all the requirements for being interpreted, so we should attribute thoughts to him and meanings to his utterances. A possible way out from this predicament could be that of appealing to the distinction between narrow and broad content. More specifically, one could claim that, at the moment of birth, whereas the narrow content of the Swampan’s intentional states was identical to Davidson’s narrow content, those states had no broad content. Davidson, however, never accepted the idea of narrow content (see for example his 1987). But this means that the conceptual conflict concerning the possibility of attributing mental contents to the Swampman was unavoidable, since on the one side there was a huge difference in Davidson’s and the Swampman’s broad content (the latter did not have any), and on the other side they would be interpreted exactly in the same way by the normal interpreters, so that the same contents would be attributed to them. This shows that the Omniscient Interpreter and the Swampman were the symptoms of a conceptual tension in Davidson’s philosophical system; and this is why they did not live long. The Omniscient Interpreter, after having being introduced in Davidson (1977), softly disappeared from the scene. The Swampman, on the contrary, departed in a more spectacular way. In (1999), Davidson wrote that the Swampman was “embarrassing” for him, and therefore had to be dispensed: «Science fiction stories that imagine things that never happen provide a poor testing ground for our intuitions concerning concepts like the concept of a person, or what constitutes thought» (1999: 35). The right question philosophers should ask themselves—Davidson wrote—was, «how do we tell when a creature is thinking, when we use our ordinary intuitions and knowhow?» (1999: 35). According to him, the answer to that question was easy, and no bizarre philosophical fiction—such as the Swampman or the Omniscient Interpreter—should stop us from answering the query correctly: «If we can communicate with the creature or robot on a range of topics in our natural environment, it is conscious and it is thinking» (1999: 35). So whatever bizarre stories of causal interactions with the physical environment a speaker may have—perhaps it only is a robot that blindly exe-

193

cutes a program and has no previous history of causal connections with the environment—, it is only her interpretability by the normal interpreters that determines if that speaker thinks and speaks, and what the semantic contents of her thoughts and words are. As Davidson wrote: «It is understanding that gives life to meaning, not the other way around» (1994b: 121). If this is not a form of social externalism, nothing is. References Bilgrami, A. 1992, Belief and Meaning. Oxford: Blackwell. Burge, T. 1979, “Individualism and the Mental”, repr. in Burge (2007), pp. 100150. Burge, T. 1982a, “Other Bodies”, repr. in Burge (2007), pp. 65-81. Burge, T. 1982b., “Two Thought Experiments Reviewed”, Notre Dame Journal of Formal Logic, 23, pp. 284-294. Burge, T. 1986a, “Cartesian Error and the Objectivity of Perception”, repr. in Burge (2007), pp. 192-207. Burge, T. 1986b, “Individualism and Psychology”, repr. in Burge (2007), pp. 221-253. Burge, T. 1986c, “Intellectual Norms and Foundations of Mind”, repr. in Burge (2007), pp. 254-274. Burge, T. 1988, “Individualism and Self-Knowledge”, Journal of Philosophy, 85, pp. 649-663. Davidson, D. 1975, “Thought and Talk”, repr. in Davidson (1984), pp. 155-171. Davidson, D. 1979, “The Inscrutability of Reference”, repr. in Davidson (1984), pp. 227-241. Davidson, D. 1984, Inquiries into Truth and Interpretation. Oxford: Oxford University Press. Davidson, D. 1983, “A Coherence Theory of Truth and Knowledge”, repr. in Davidson (2001), pp.137-153.

194

Davidson, D. 1987, “Knowing One’s Own Mind”, repr. in Davidson (2001), pp. 15-38. Davidson, D. 1988, “The Myth of the Subjective”, repr. in Davidson (2001), pp. 39-52. Davidson, D. 1990a, “Epistemology Externalized”, repr. in Davidson (2001), pp. 193-204. Davidson, D. 1990b, “The Structure and Content of Truth”, Journal of Philosophy, 87, pp. 279-328. Davidson, D. 1991, “Three Varieties of Knowledge”, repr. in Davidson (2001), pp. 205-220. Davidson, D. 1992, “The Second Person”, repr. in Davidson (2001), pp. 107121. Davidson, D. 1994a, “Self-Portrait” in S. Guttenplan (ed.), A Companion to the Philosophy of Mind. Oxford: Blackwell, pp. 221-236. Davidson, D. 1994b, “The Social Aspect of Language”, repr. in Davidson (2005), pp.109-125. Davidson, D. 1999, “Interpretation: Hard in Theory, Easy in Practice”, in De Caro (ed.) (1999b), pp. 31-44. Davidson, D. 2001, Subjective, Objective, Intersubjective. Oxford: Oxford University Press. Davidson, D. 2005, Truth, Language, and History, Oxford: Oxford University Press. De Caro M. (ed.) 1999a, “Davidson in Focus”, introduction to De Caro (ed.) (1999b), pp. 1-29. De Caro M. (ed.) 1999b, Interpretations and Causes. New Perspectives on Donald Davidson’s Philosophy. Dordrecht: Kluwer. Heil, J. 1992, The Nature of True Minds. Cambridge: Cambridge University Press. Kripke, S. 1972, Naming and Necessity, Cambridge, Mass.: Harvard University Press.

195

Kripke, S. 1982, Wittgenstein on Rules and Private Language. Oxford: Blackwell. Lepore, E. and Ludwig, K. 2005, Donald Davidson. Meaning, Truth, Language, and Reality. Oxford: Oxford University Press. Putnam, H. 1975, “The Meaning of ‘Meaning’”, repr. in Mind, Language, and Reality (Philosophical Papers, vol. 2). Cambridge: Cambridge University Press, pp. 215-271.

THE EXTERNALISM OF TRIANGULATION Gerhard PREYER (Goethe University, Frankfurt am Main, Germany)

The theory that language and meaning follow from radical interpretation (RI) is a continuation of Quine’s post-empirical theory of meaning, which claims to identify a cyclical relationship between belief and meaning into which an radical interpreter must break in. In Davidson’s theory, the essential feature of language is that the speaker himself is an interpreter of another speaker. The nature of language demands that every speaker be radically interpretable. Therefore, the third person perspective is global. Only the radical interpreter has the task of interpretation. The third person perspective and the constraints of RI create the epistemic position for understanding meaning and mental concepts. The problem with Davidson’s philosophy is that it makes the global third person point of view of an interpreter the foundation for understanding linguistic behavior. Davidson’s philosophy contends that the speaker’s disposition determines the meaning of words. This conclusion is drawn from the belief that the radical interpreter’s position toward a speaker forms the foundation of the theory of interpretation. This belief also suggests that facts about the speaker’s actual prior or subsequent behavior and facts about his participation in conventional word use do not form the basis of understanding linguistic behavior or other types of behavior. Davidson’s theory of language and thought is a version of the theory-theory approach because the concepts of the mental state are to understand by our theory of interpretation and its specification to causal relations which trigger the observed behavior in the environment of the interpreter and the speaker. Firstly, I will show that the truth-centered theory of RI leads to a nonreductive externalism of the individuation of thought content. However, the principle of charity cannot accomplish the task of interpretation. The constraints of making behavior intelligible are not provided by charity. In fact, RI is impossible. Secondly, I argue that the view of triangulation as a link between the mental (attitudes), the linguistic, the communicative (the social) and other aspects of nature is mistaken. This conclusion leads me

198

back to the epistemic restrictions of making behavior intelligible. I conclude that both RI and triangulation are impossible. 1. Language-Grounded Externalism Radical Interpretation The global third person position of interpretation constrains intelligible redescription by RI. The causal structure of the world in which the speaker is embedded is insufficient to select correlations between true attitudes and conditions in a speaker’s environment. The interpreter must choose between sets of correlations, and incompatible interpretation theories do not enable the interpreter to exclude certain attitudes or correlations. The notion of RI focuses the constraints of intelligible redescription to accomplish its task on a priori knowledge of the principle of charity (rationality) and to select successful interpretive markers. This knowledge is not sufficient for determining a unique correlation. With this approach, the theory of interpretation cannot explain this under-determination. Davidson’s view is that every speaker is radically interpretable in a particular environment (the ambitious version of RI).1 A radical interpreter is not, however, a mind reader.2 Therefore, Davidson has concluded that RI is possible and is successful, regardless of what is true about a speaker as such. However, additional constraints arise, and the assumptions about the subject of interpretation are not conditioned simply by the speaker. Further assumptions must be made about the speaker (as thinker and agent). These factors include the speaker’s psychology, culture and social status. The notion of RI is a confirmation (justification) of an interpretive truth theory for a single speaker. The constraints are 1. the assumption of a rational agent, that is, the radical interpreter applies the Principle of Coherence and Correspondence (charity, basis of ra1

On the ambitious and the modest project of radical interpretation, see Lepore and Ludwig (2005: 166-173). The first is that radical interpretation is possible from the third person position a priori, that is, it concludes that the speaker is right in most matters; the second is that radical interpretation is only possible a posteriori. The question is whether charity is to justify a priori as a transcendental guarantee. 2 This is the difference in principle to the proposal of Goldman (2006).

199

tionality, normativity/rationality of mental features) as a bridge principle (constraints of interpretation3) and 2. complete information about physical interactions with the environment (triangulation)—the theory assumes an omniscient interpreter with perfect knowledge of the world but not of the attitudes of speakers and the meaning of the uttered sentences under study.4 From this point of view, the interpreter does not use semantical concepts to fix lingual objects to non-lingual objects (the external world) thereby sentences of a language are informative (informational content), and we state (judge) something about the assumed universe of discourse of these objects. A semantical concept of truth reflects a relationship between a sentence and its content. According to this concept and not the theory of truth (fulfillment), semantical expressions have an intension.5 We describe contents, judge something and communicate information. From my point of view, semantical interpretation of expression requires the inference from an all-sentence to the specific case (example). But the concept of interpretation does not align with the expressions of this language (object language). Therefore, we do not possess this concept in the object language. This concept is, therefore, a part of a meta-language. We classify in the meta-language the sentences as true and false for a given object language. We utter (use) the sentences of the meta-language and mention the sen3

Davidson and Quine attribute to the speaker the modicum of classical logic, which creates the basic situation of radical translation (interpretation). I think that the power of classical logic should not be forsaken. However, it does not justify charity and RI. 4 Davidson’s argument is problematic. The omniscient interpreter attributes attitudes to the speaker on the basis of his own beliefs, and this is the usual praxis of all speakers. Massive error about perceptual beliefs is, therefore, impossible and unintelligible. If we assumed that, we would view an interpreter as an omniscient one who would correctly interpret speakers as soon as they make extensive mistakes. This is impossible from Davidson’s point of view; see Davidson (1977: 201). It is doubtful whether this argument works because there is no transcendental guarantee that most of our beliefs are true. This is simply presupposed from the stance of an omniscient interpreter (see Lepore and Ludwig (2005: 323-326)). An omniscient interpreter cannot have false beliefs. The question is: why is charity required for interpretation? The two are not compatible; see Fodor and Lepore (1992: 159-161). 5 On the Fulfillment-Conditional Approach, see Lepore and Ludwig (2007: 276-282).

200

tences of the object language. These sentences could also be an object of a true-false statement, but on a higher (meta) level. Therefore, I conclude that there is no theory of truth. These levels are unlimited and open; there is no ultimate statement of truth. I do not restrict the principle of abstraction, but I also do not assume abstract entities. The alternative is to level classes of entities and to suppose that properties and classes of entities cannot be expressed on the same level. But the strategy which is powerful for fixing ontological commitments cannot be proven. 6 Davidson’s semantics does not consider this option.7 3. Davidson adds the first person authority to the procedure of RI because the asymmetry of RI works in our communication of our own mental states and what others can say about us. The mental is determinate by the internal, but it is not determinate by itself. 4. Holism, rationality of mental features and externalism work together. None can be eliminated without changing the subject. Since the middle of the last century philosophy, the philosophy of language, and the theory of science have been dominated by different versions of holism, such as epistemic holism in epistemology (Hempel, Quine), semantic holism (the conceptual role/inferential role semantics of Wilfrid Sellars) and the holism of radical interpretation. Quine discusses the holistic picture in semantics. Jerry Fodor and Ernest Lepore have characterized holism in their critique on semantic holism: «Holistic properties are prop6

See Essler (1988). Therefore, the distinction between concept of truth, criterion (evidence) of truth and theory of truth is fruitful. In the theory of truth would give a truth definition. The question is: For what—for languages with a unique ontology and for all cases or for an interpreted language with its ontology which we can modify? 7 On the modifications of Davidson’s concept of truth and the ontology of RI in the history of his work, see Preyer (2011: 126-33). See also an overview of Davidson’s theory of truth in Lepore and Ludwig (2007: 315-24). They conclude that Davidson’s theory of truth is a correspondence theory because the true predicate is to interpret semantically by the satisfaction of conditions (reference) and not of the intension of truth. But the issue is that every meta-language states something about a language of a lower level. There are no maximal universes of the discourse of objects. Therefore, let us make the assumption that the concept of the universe of discourse is dependent on a supposed language. The universe of discourse is fixed to syntactical (logical) expressions. We always speak about that fixation in a meta-language.

201

erties such that, if anything has them, then lots of other things must have them, too. 2. This is a metaphysical characterization of holism (as generic properties). Some authors speak on a logical dependency only» (Fodor and Lepore 1992: x). But in the last case, we do not characterize the properties as holistic. The logical implications between sentences are not holistic according to the characterized meanings, which is theoretically trivial and does not commit us to semantic holism. Fodor and Lepore have emphasized that holism «is a doctrine that only the whole language or whole theories or whole belief systems really have meaning, so that the meaning of smaller units [...] are merely derivative» (1992: x). Quine gives this statement an epistemological turn because, according to his theory, we do not confirm a single belief (confirmation/epistemological holism). This notion is connected in the Quine-Davidson tradition with a belief in holism. For them semantic holism is valid: a conceptual content and meaning holism (semantic holism). This is the common reading of Quine’s “Two Dogmas of Empiricism” (1951), his critique of the Wiener Kreis’s verification theory of meaning and logical empiricism.8 An alternate reading of Quine suggests that the indetermination of translation (incompatible translations) does not contradict his naturalism (scientific realism) because differences in translation are empirically equivalent, that is, there are no physical differences between either. Quine ascribes this power to the stimulus synonymy, which is tested empirically (statistically). The naturalizing of epistemology excludes that different translations express different intentional states. If this were the case, the museums-myth would reemerge. Coherence applies a degree of the principle of consistence and ascribes to the speaker a modicum of logic. Therefore, the interpreter assumes that the speaker is logically consistent in his thoughts. The guarantee of the empirical turn of formal semantics and decision theory is the application of the basis of rationality. This rational basis describes the mentality of the speaker (thinker, agent) and what he thinks, wants or means by a degree of logical coherence when we ascribe a content to attitudes, thus linking meaning with linguistic utterances. 8

Quine’s article was one of the most discussed articles in philosophy after its publication. For a critique see, for example, Fodor and Lepore (1992: 37-58).

202

Correspondence assumes that the speaker is disposed to respond to a feature of the world in the same way the interpreter would. The interpreter redescribes the speaker he is interpreting in a way that he finds true to himself. Davidson has argued that the indeterminacy of translation does not arise because of questions regarding the truth of the speaker’s sentences. Applying the theory of truth in the style of Tarski on linguistic behavior in the procedure of intelligible redescription by an interpreter, for example, we see that the statement “Peter said snow is white,” at a particular time point with a reference to a speaker (the Davidson truth-convention9) does not have as output a qualified knowledge. Truth is a central external component of Davidson’s non-relativist theory of linguistic behavior.10 But Davidson considers that we are not born with the knowledge of how words and the world are connected; we must learn the connection.11 The shared response individuates the content of thought, forming a triangle between speaker, interpreter and the external world by the intersection of causal chains, which trigger the same (similar) response. The omniscient interpreter is introduced because the true belief of the external world and the individuation of content (most of our beliefs are true) work together.12 Da9

See, on the Davidson truth-convention, Ludwig (1999: 30-31). On the paradigmatic switch away from traditional semantics—from Davidson (1965) and the initial project (the compositional theory of meaning: the meaning of sentences depends upon the meanings of words) to Davidson (1967), which is continued as the extended project (1. extensional truth theory for natural language given by the Tarski-style truth theory; 2. confirmation of the interpretative truth theory by RI) or a re-interpretation of the compositional meaning theory by the replacement theory (the compositional meaning theory is exchanged by the truth theory; the replacement of the theory with the theory of reference)—see Lepore and Ludwig (2005: 19-37). In Davidson (1967), the step to a replacement theory is not considered. But in continuation, Davidson switches to a replacement proposal as a critique of a building block theory. See, for example, as textual evidence, Davidson (1988a: 180, 1990: 299-300). 11 The question is whether language is learned, and we learn a language by correlations with caused behavior. Language cannot be learned, as behaviorism—trivial or not—has argued. 12 Davidson is not a pragmatist like Richard Rorty; for example, among other responses to Rorty, see «Correspondence, while it is empty as a definition, does capture the thought that truth depends on how the world, and this should be enough to discredit most epistemic and pragmatic theories» (Davidson 1999: 114). 10

203

vidson calls this knowledge the basis rationality that we (all rational beings) have (a priori). It is, therefore, not merely a policy we can choose.13 The theory of thought, meaning and decision is unified because the primitive term in the theory is the attitude of preferring one sentence as true rather than another sentence. No entities are required for the description of this relationship. Meanings are not required (not sufficient) for a compositional semantics of natural language. In contrast, translation/interpretation is possible only when we assume that speaker and interpreter believe that the same statement holds true. First Person Authority Saying something about first person authority is relevant epistemologically and for the analysis of psychological concepts. The problem is whether we agree with the relational account of mental states. Do first person authority and externalism work together? Davidson explains the first person authority by RI, that is, we must assume that the speaker knows the meanings of the uttered words, while this is not the case for the interpreter. Therefore, the speaker is in a particular epistemic position. The epistemic state of the thinker is determined by behavioral issues regarding the relationships among the speaker, the sentences the interpreter directly receives, and the role of the first person authority in the procedure of RI because this knowledge is authoritative in contrast to knowledge acquired by evidence. Lepore and Ludwig (2005) have shown that this argument is not successful. Firstly, this is not an explanation of the speaker’s knowledge, and, secondly, it does not explain the asymmetry of first person knowledge. The assumption that the speaker is in a particular epistemic position does not necessarily mean that he can single out his consciousness and does not consider the primary role of our knowledge of his mental states from the interpreter’s position. Therefore, Davidson’s concept of first person authority cannot prove that the speaker’s knowledge of himself is not inferential.14 In addition to this critique, if an interpreter assumes that a 13

This notion is emphasized in Davidson (1985: 196-197). See Lepore and Ludwig (2005: 343-372). Regarding the master-argument (of first person authority), see Ludwig (1994).

14

204

speaker is interpretable only if he knows the meanings of his words and the content of propositional attitudes, we have an explanation of the speakers’ knowledge, which provides an answer to the challenge of non-inferential knowledge. The first person stance is not, though, the foundational methodological stance of interpretation.15 Holism about content of thought is a feature of the whole system of attitudes, which the radical interpreter ascribes to the speaker. He does not ascribe single attitudes, but he assumes that there are endless related propositional attitudes. Attitudes are related by their content. However, the related attitudes do not dictate the particular correlation of content to particular beliefs, which is called the two holistic constraints of attitude sets (Lepore and Ludwig 2005: 21216). These constraints are of particular significance because the ascription of attitudes within a logical (abstract) structure does not fix the empirical belief and the concepts of the individual speaker. Davidson’s view is that attitudes always come to mind in patterns. The question is whether the mental attitudes and the mental language are expressed holistically in a whole system of beliefs. It is significant to mention that this question leads us to the problem of communication from the semantical point of view. When we assume that the attitudes of one person are structured by inference roles in the person’s own belief system, then there are no two persons who have the same conceptual content because there are no beliefs that 15

Lepore and Ludwig (2005: 368) have argued similarly. They distinguish a rigid holism: there is a relationship between particular attitudes with particular contents, which are related to specific attitudes and their specific content that the speaker has, from an extreme holism: particular attitudes with their particular content are dependent on all of the other attitudes that the speaker has and, therefore, a change in one attitude changes, at the same time, the content of every other attitude. Both holisms are not aligned with Davidson’s proposal (Lepore and Ludwig 2005: 211-213). They claim that Davidson refuses to commit to meaning holism (2005: 213). He has argued contrarily: «only in the context of the language does a sentence (and therefore a word) have meaning» (Davidson 1967: 22). He has corrected this view: «I am not an unbuttoned holist in that I do not say the meaning of a sentence depends on the meaning of all sentences» (Davidson 1994: 124). But the problem is a deeper one because, when the principle of compositionality is applied, we are not committed to holism in semantics. The problem is not that the sentence’s meaning depends on the structure that we derive from it. 16

205

have the same inferential relations in the whole system of beliefs of two persons. This is not a critique on semantic holism, but, under these conditions, communication would not be possible. Recalling the constraints of RI from the third person stance, we must consider the question of the possibility of RI. RI is obviously not possible because the conformation of the truth-centered theory of RI borders on the epistemic restrictions of interpretation in general. This notion is a variation of Lepore’s and Ludwig’s (and Fodor’s and Lepore’s) critique17—with a slightly different intent—which concludes the impossibility of RI. If constraints were realized a priori, every speaker would be radically interpretable, and interpretation from the third person position of a radical interpreter would be possible (ambitious project, sufficient evidence of interpretation). But the concern is that holding true sentences supported by behavioral evidence caused by related conditions in the surroundings cannot qualify the attitudes of the speaker globally. Lepore and Ludwig argue that the so-called possibility of RI is a posteriori. Some speakers are radically interpretable (modest project). From their point of view, ceteris paribus, the success of interpretation is true. This belief qualifies the speaker’s knowledge, which is considered in the procedure of interpretation. Therefore, it is not the external component of truth (distal stimulus, meaning, common cause) which is the guarantee of the individuation of the content of thought in the procedure of RI. If this is true, it also affects the concept of the first person position and the speaker’s epistemic position. I conclude, then, that language-grounded externalism of triangulation—the linguistic, truth-centered theory of RI and the external individuation of the content of thought by common causes—is an epistemological condition for RI. That is, language is necessary for thought. This version of externalism assumes, in continuation of Davidson’s philosophical work, that meaning is determined by observable behavior; it is not defined by, or reduced to, this behavior. It is a synchronic (not diachronic) and nonreductive (physical) version of externalism. Davidson agreed with Quine, 17

See Lepore and Ludwig (2005: 174-197). They introduce a stronger principle, Grace (2005: 194-197), which I do not discuss here. I agree with them that charity is not sufficient. It is not disputed that, when we accept semantic representations as semantic properties, we give up attitude (semantic) holism.

206

though, that language is a set of dispositions that we harmonize with utterances within observable circumstances. This concept of meaning followed from the step Davidson has taken from the compositional theory of meaning to a truth-centered theory of RI that is constructed from the third person position. Therefore, language is public in a strong sense. Epistemically, there are no intermediaries between us and the world; also, language is not a medium (on that in particular matter, see Davidson 1997a). Semantic internalism (Cartesian intuition) is a contrary view.18 2. Triangulation From Quine’s point of view, meaning, reference, and representation do not have an objective status. For naturalized epistemology, language is a set of dispositions which are fitted by a translator to evidential circumstances. All ascription of a representative content to a speaker is indeterminate. Quine’s critique on the logical empiricists is that belief and meaning are connected in principle. Therefore, there is no pure linguistic meaning. Stimulus meaning is fixed as a class of sentences, which have no linguistic or theoretical meaning. The empirical net-meaning of observed stimuli breaks into the circle of belief and meaning. Quine’s concept of meaning is restricted by the observational circumstances of a radical translator. Learning a language requires a direct association with a concurrent stimulation. The concern is that the analysis of observation sentences—in this case, their meaning is obvious—tells us nothing about the expressions of particular types of sentences by the specification of the conditioned stimuli and responses in particular situations, which are observed by a translator. Davidson’s triangulation of the content of thought is a critique on the role of stimulus meaning. His central critique of Quine’s ontogenesis of reference is the rejection of proximal stimulation as an explanation of meaning and reference.

18

I characterize the Cartesian intuition that the mental is determined by itself as an epistemic version of internalism. From my point of view, we must redraw the line between the mental (the internal, the subjective) and the physical (the external, the objective) (Preyer 2011).

207

The triangulation model of RI is the foundation for Davidson’s theory of meaning, truth, action and evaluation and their unification.19 This theory claims that triangulation explains how we conceptually link the mental (attitudes), the linguistic, the communicative (the social) and other aspects of nature. We should consider that triangulation, as the individuation of the content of thought, intends to explain how two speakers with different perspectives can view the same reality and how what they say about the world is intelligible among two or more speakers. Therefore, triangulation justifies a realistic conception of the world (of objective truth) in which we are thinkers (speakers, agents). The (direct) objects of awareness are, thus, not interpretations, as Nelson Goodman has argued. Perspectives of Triangulation Triangulation is a pre-linguistic, precognitive situation that is a necessary condition for language and thought.20 It occurs in the ontogeny and phylogeny of attitudes. Building the triangle is based on the third person perspective. The interpreters of Davidson’s epistemology of triangulation agree that the lines of triangulation emerge in perspectives between the interpreter, the speaker and the external world by the intersection of causal chains. Triangulation is, at the same time, a model of learning. The claim of triangulation is that the common cause individuates the content of thought of the speaker when he is in communication about the object of thought with others. Without the public object, communication is not possible. The baseline of the triangle, which requires communication, determines the knowledge of the world and, at the same time—through the observable behavior of others—the knowledge of other minds.

19

I do not analyze the unification of the theory of interpretation and decision itself but only the connection between his theory of meaning and the individuation of thought (propositional attitudes). On the unification, see Lepore and Ludwig (2005: 248-260). There is a problem in Davidson’s unification, in principle, because he provides no evidence that propositional attitudes are countable, that is, he does not pair numbers to these attitudes. On decision and practical thought, see Preyer (2011). 20 First introduced in Davidson (1982b: 105, 1992: 117-21, 1997b: 128). See on triangulation, for example, Lepore and Ludwig (2005: 404-12); Glüer (2006).

208

This notion agrees with Quine’s behaviorism in linguistics as a basis theory. Critics have said that the thesis (of indeterminacy of translation) is a consequence of my behaviourism. Some have said that it is a reduction ad absurdum of my behaviourism. I disagree with this second point, but I agree with the first. I hold further that the behaviourist approach is mandatory. In psychology one may or may not be a behaviourist, but in linguistics one has no choice. Each of us learns his language by observing other people’s verbal behaviour and having his own faltering verbal behaviour observed and reinforced or corrected by others. We depend strictly on overt behaviour in observable situations […] There is nothing in linguistic meaning beyond what is to be gleaned from overt behaviour in observable circumstances (Quine 1990: 37-8). A child learns his first words and sentences by hearing and using them in the presence of appropriate stimuli. […] What I have said of infant learning applies equally to the linguist’s learning a new languages for which there are previously accepted translation practices, then obviously he has no data but the concomitance of native utterance and observable stimulus situation (Quine 1969a: 81).

In this case, there is no difference between the epistemic stance and the individuation of content of thoughts by the proximal stimuli meaning and the distal meaning caused by the public objects. The interpreter extracts from observable behavior the theory of thought and language of the speaker, that is, the third person position is the interpersonal stance, as the epistemic foundation, which gives thoughts their content. Therefore, knowledge of the world, of one’s own mind and of other minds are essentially interconnected. This belief revises Cartesian epistemology. The problem is whether the argument from triangulation support the global third person position in epistemology and gives thoughts, by the public object, their content as the language content of speech.

209

First perspective The speaker and interpreter are both aware at the same time of similar things in their surroundings. The relevant stimuli are objects and events that are correlated with the responses of the speaker, which the interpreter finds similar to his own responses. The line of triangulation runs from (a) the speaker in the direction of the objects and events—the direction is perceived by similar response (b) the interpreter in the direction of the objects and events—the direction is perceived by similar response (c) the lines run between both the speaker and the interpreter—the intersection of causal chains, which causes the common response, gives thoughts a content. The common cause is the public object.21 All perceptions are caused simply by what we see, hear, touch, taste and smell. The content of observation sentences (“perceptual sentences”) is directly connected to perception, which provides the content for the belief (Davidson 1997a: 137). These sentences are expressed and caused by particular circumstances. Therefore, having a concept is not distinct from having a thought.

21

Davidson argues that there is an “ambiguity of the concept of cause”: «Without other people with whom to share responses to a mutual environment there is no answer to the question what it is in the world to which we are responding. The reason has to do with the ambiguity of the concept of cause. It is essential to resolve these ambiguities, since it is, in the simplest cases, what causes a belief that gives it its content. In the present case, the cause is doubly indeterminate: with respect to width, and with respect to distance. The first ambiguity concerns how much of the total cause of a belief is relevant to content. The brief answer is that it is the part or aspect of the total cause that typically cause relevantly similar responses. What makes the responses relevantly similar in turn is the social sharing of reactions that makes the objectivity of content available. The second problem has to do with the ambiguity of the relevant stimulus, whether it is proximal (at the skin, say) or distal. What makes the distal stimulus the relevant determiner of content is again its social character: it is the cause that is shared. The stimulus is thus triangulated; it is where causes converge in the world» (1997b: 129-130).

210

The distal stimulus is located where the lines from the speaker and the interpreter to the object and event converge. Davidson argues that the object and event are a robust (common) cause in the triangulation that gives thoughts their contents because both are public entities. The response of the creature is evidence that both have these concepts of such particular objects and events. They are concepts as concepts of, that is, the speaker has some particular objects or events as object of his thought (Davidson 1992: 119). Yet it is obvious when the object is the cause of the content of thought—whatever this means—the question is of “which link” causes the content (Fodor 2008). The significant question is whether we know the content of our thoughts without an observation of behavior. Thoughts require second, third and fourth perspectives. Second perspective The speaker who is interpreted also interprets the interpreter, that is, there is a reciprocal interpretation. This, however, is an asymmetrical relationship. To be a linguistic being is to be a radical interpreter. The evidence of intelligible redescription is the third person position. The evidence from this stance is globally available for every interpreter of a speaker. Therefore, Davidson assumes RI is possible for all linguistic creatures. This theory leads to the problem of possibility of RI. When this perspective breaks down, triangulation does not work and does not give thoughts their content directly; the individuation of the content of thought by building the triangle is not an argument for the possibility of RI. Why is it not possible that we correlate our thoughts and our behaviors from the first person perspective and project the correlation to others? Davidson must argue that this correlation generally works. This conclusion leads us back to Cartesian intuition. Third perspective We must add that both the speaker and the interpreter observe others in communication and can communicate with people. «Communication begins where causes converge: your utterance means what mine does if beliefs in its truth is systematically caused by the same object and event» (Davidson 1983: 151). That is, the common (public) cause is shared by

211

both the speaker and the interpreter. Therefore, there is no communication among people without responding to selected content-determining causes. A justification of the attributing of attitudes is not required (Davidson 1989: 198). Thus, communication is only possible on objective grounds of the third person position. Two distinct concerns arise: 1. what is required by the concept of objective thoughts and 2. how do we identify a unique object of thought (Lepore and Ludwig 2005: 408)? The first question is a problem of the theory of truth; the second is one of the individuation of the content of attitudes. I do not discuss this point here because there is a deeper problem: communication requires the recognition of space-temporal things and events for both speaker and interpreter. From my point of view, communication requires that we recognize (identify) spatial and temporal concepts and events within a linguistic framework, and that we identify/reidentify these entities. They are not given directly. Fourth perspective Each participant in the triangulated communication is understood in his speaking by others when RI is possible without speaking the same language. This level closes the references to the world (common causes) and to other people as interpreters. Therefore, the third person position is, in triangulation, the basis of linguistic communication because the speaker and interpreter do not share a common language as the basis of communication. The work of the causal chain between the speaker, the interpreter and the world makes communication possible by common responses in the particular environment in which they interact.22 I have argued that triangulation requires perspectives between the interpreter, the speaker and the external world. But the perspectives are not level in triangulation. Triangulation always considers the same perspective: the third person position. The four perspectives lead us back to the constraints of interpretation because they do not eliminate issues of relativity 22

For Davidson, there is a historical component to thought content. I think it is correct that the historical content is not intrinsically motivated by the methodological stance of his externalism. The support offered by the Swampman thought experiment is not persuasive; I agree with Lepore and Ludwig (2005: 337-342).

212

and indetermination. Davidson’s point of view also uses triangulation to find the proper cause from the third person stance to give thought its content as an argument for RI. The external condition closes the triangle. This is Davidson’s version of triangulation as a language-grounded-externalism. This version of externalism agrees with externalism in general because for triangulation all content is wide. This creates a myth of subjective from Davidson’s point of view (1988b).23 Therefore he concludes there is no private language (thought). His critique of a private language is not reasoned, though, from the rule-following argument but from the argument that language is necessary for thought (first introduced in Davidson 1974: 170). The argument is that we have the concept of a belief only by the interpretation of language from the third person stance of RI. Thus, the theoretical description of interpretation from the third person stance as a stance of yours and mine— not only as a philosophical exercise—makes interpretation and understanding possible. The argument is reasoned from triangulation. However, the belief that the only evidence for the interpretation of linguistic behavior is the third person evidence, which is available for the radical interpreter globally, does not create a successful argument. Davidson’s language-grounded externalism assumes, in continuation of his philosophical work, that meaning is determined by observable behavior. He agrees with Quine that language is a set of dispositions that we harmonize with utterances within observable circumstances. This concept of meaning follows his movement from the compositional theory of meaning to RI as a theory of linguistic behavior from the third person stance. This theory leads us back to a reinterpretation of the relationship between language, semantics and ontology and a reconsideration of what we can learn about ontology itself from the semantics of language, which illuminates the ontology of language we speak.

23

This is a continuation of Quine’s myth of a museum, that is, «the exhibits are meanings and the words are labels» (1969b: 27).

213

3. The Epistemological Role of Publicity Externalists, in particular, proponents of the language-grounded externalism of triangulation, argue that the individuation of thought by distal stimuli and dispositional meanings are basically connected. This belief aligns with mainstream of philosophy in the last century (Ludwig Wittgenstein, Gilbert Ryle and their fellows). Ryle, for example, assumes that what a person intends with his behavior can be found in the epistemic public surroundings of his actions. It is often argued that this notion is the antecedent for the ascription of intentions and beliefs. However, an agent may have a mental disposition to do something but does not think of acting in this way. It may be that, in many cases, we cannot discover how attitudes work together. For Ryle (1949), some words that describe human behavior refer to dispositions and do not refer to episodes (occurrences), as, for example, “smoke a cigarette” does. Words like “knowing”, “intelligent”, “clever”, and “humorous” specify dispositions, tendencies, or abilities.24 Ryleian language is a public language. This is not so obvious, though, as it prima vista appears. The question is whether publicity plays an epistemic role for the externalistic individuation of the content of thought as conscious states in principle. This issue coincides with another: is meaning dispositional? For language-grounded externalism, the question arises of whether the publicity of observable behavior is a constitutive feature of language: «There are obvious relations between holding a sentence true and linguistic (and other) behaviour» (Davidson 1988a: 190. my emphasis): Perhaps someone (not Quine) will be tempted to say, ‘But at least the speaker knows what he is referring to.’ One should stand firm against this thought. The semantic features of language are public features. What no one can, in the nature of the case figure out from the totality of the relevant evidence cannot be part of meaning. And since every speaker must, in some dim sense at least, know this, he cannot even intend to use his words with a 24

We should mention in this context that this turn goes back to the dispute between Moritz Schlick and Otto Neurath and the switch from a phenomenological language to the universal assumption of thing-event language as a basic theory in epistemology. It was perhaps an idea of Wittgenstein’s (1929), as Merrill B. Hintikka and Jaakko Hintikka (1986) have argued. On the protocol-sentence debate and the problem of publicity, see also Davidson (1982a).

214

unique reference, for he knows that there is no way for his words to convey this reference to another (Davidson 1979: 235).

From Davidson’s point of view, there is a dilemma in the theory of knowledge: 1. Moritz Schlick’s self-certainty of observation statements or events are private, and it is not possible to write down these sentences without losing their certainty (they have no connection to public language). 2. Sentences (or beliefs expressed) within public language have no intelligible link to self-certification (Otto Neurath’s problem). Therefore, Davidson has concluded that the foundation of knowledge must be subjective and, at the same time, objective (Davidson 1982a: 168). The question is whether Davidson’s answer actually works. Davidson’s language-grounded externalism contends that every interpretation essentially refers to things and events that give the uttered words their meaning. He claims to explain communication by the convergence of common causes between the speaker and the interpreter. The distal meaning (reference) individuated by common causes assumes that the content of thought of observation sentences is, in most cases, determined by the significant features that the speaker and interpreter perceive; it is an unmediated correlation of our commonsense understanding of learning a language. Objectivity originates with the triangle of the speaker, the interpreter and the world that they share.25 The consequence of this origin is that language is public in a surprisingly strong sense. In Davidson’s view, the theory of truth gives us the conceptual resources for understanding the world and intelligible redescription of behavior, and truth is ultimately based on beliefs and on effective attitudes, that is, the thinker (speaker, agent)’s dispositions. Publicity and dispositional meaning primarily act together. However, this proposal is too strong. It does not include constraints on belief. Publicity is a domain of verification, but it is not a foundation of knowledge and of objective thoughts. Publicity is a principle of verifying knowledge in general, but not an exclusive epis25

On the argument of triangulation having thoughts, see Lepore and Ludwig (2005: 404-411).

215

temological account of language and the ascription of knowledge. Public circumstances have, as constraint, the notion that the thinker (speaker, agent) and the interpreter have the same ontology and that, within linguistic framework and its expressibility, they identify (re-identify) entities. The problem with this theory is that no ascribing person reaches another’s consciousness. In this sense, we are captives of a Leibnizian Monadology. We also cannot broaden our consciousness to the external world; we see our picture in a mirror, but we cannot observe ourselves from an external point of view. This notion is valid from the third person position. However, there is also, from the first person perspective, the transparency of my own consciousness. I do not know myself better as an observer but in another manner. My first person authority is not non-correctible and absolute. There are self-deceptions. These aspects represent one side of the first person authority, but the other side contends that I know myself immediately, as others cannot know me, or that I ascribe to myself something I cannot ascribe to others.26 Publicity is an Ersatz (surrogate) for an epistemic qualification of the procedure of RI (interpretation in general). For the truth-centered theory of RI and the individuation of the content of thought, there is no epistemic qualification of the speaker under study. I call this restriction the epistemic restriction (resources) of interpretation. This notion is also valid if the utterance is true from the standpoint of a (radical) interpreter. The notion of RI is not possible because epistemic restrictions are taken in every interpretation (understanding) of behavior, linguistic or not. The truth condition of lingual utterances does not depend on the causal history of the interpreter’s relation to objects and events as the proper causes in general. This conclusion is not reached in ignorance of Davidson’s basis problem of understanding one’s language because a framework is required for the analysis of the connections between words and world, which are established for the speaker and the interpreter in the selections of communicative intercourse. The critique of this basis theory is that the epistemic restrictions of interpretation cannot be found from the third person point of view of interpreta26

The concept of tabula rasa is a critique of the Cartesian ideae innatae. Also, John Locke did not assume that we are empty boxes within which we can put any contents. On the limit of first person authority, see Georg H. von Wright (1994: 157-163).

216

tion (translation) and behavioral evidence, which bring together inductive reasons that are confirmed by distal causes. The constraints of triangulation of the content of thoughts by distal stimuli answer the question how we draw the distinction between the inside and outside of thoughts and their contents. Similarly, is there a direct relationship between what we are aware of in our environment when we think about something and what we then communicate to others? The distinction between the inside and the outside is not between what is under the skin and what is outside of us in the world. How do we draw the distinction between the inside and outside of us, or how is this distinction made? This is the question of the non-conceptual content of thoughts and at the same time of the mental. Without the distinction between self-reference and reference to others, there is no inside-outside distinction. We make the distinction between self-reference and reference to others as an elementary operation of our consciousnesses. This distinction, and no causal relationship, makes communication possible. The epistemic point of view of a radical interpreter is insufficient to ascribe knowledge to a speaker we are interested in understanding, and the externalism of triangulation in epistemology cannot close the gap between the beliefs of the interpreter and the speaker by common causes. I do not, however, relinquish the claim that there are correct and incorrect interpretations of what a speaker means and intends by his words. The problem is that Quine and Davidson have not said enough about the justification of the evidential constraints of radical translation and interpretation. The reasoning of the constraints of interpretation from the third person position is described as an a priori argument (transcendental argument) and an a priori truth. Davidson communicates that he does not give a transcendental argument for the constraints of RI, but the possibility of it is shown by an “informal proof” on the evidence, which is available for RI (see Davidson 1993; on the transcendental argument of RI, see Fodor and Lepore 1992). It is obvious, though, that the principle of charity and correspondence is stronger because the radical interpreter cannot dispose of both for an interpretive theory to understand linguistic beings. Epistemic restrictions of the speaker or a group of speakers contradict the work of RI and the application of the triangulation model of RI. The a priori theory available to break into the circle between belief and meaning and the radically

217

interpretable nature of speakers present certain limits. The notion of RI does not have all relevant evidence for intelligible redescription at its disposal. I think there are actual relationships between thought, language and communication. These relationships are not founded, however, by behavioral evidence as the only “evidence” that forms the meaning of linguistic and mental concepts and expressions, nor from obvious behavioral evidence from the global third person position in the epistemology and philosophy of language. Therefore, we renew a version of the Third Dogma of Empiricism: the distinction between content and scheme is not eliminated by Davidson’s version of a naturalized epistemology, that is, by making behavior intelligible by deriving the content of thought by common causes from the third person perspective because an epistemic gap of translation and interpretation of linguistic behavior may emerge in the procedure of intelligible redescription. This theory represents a borderline of understanding linguistic beings and their responses to the social and cultural features of the world. I call this notion the paradox of understanding and communication: Understanding other people is possible by crossing context only, but epistemic restrictions (capacity) limit the success of re-interpretation and communication. We cannot exclude the possibility that there is an upper limit to an interpreter’s understanding of a speaker. The limit of understanding is the metaphysics of experience, whether the speaker and the interpreter share or not. I do not believe that people from the Western world, who adhere to standards of rationality (charity), have an understanding of a magic trick, such as rain magic. This gap leads us back to many old problems in epistemology and in understanding foreign social intercourse, which are difficulties in communication that have not disappeared in the time of globalization. References Davidson, D. 1965, “Theories of Meaning and Learnable Languages”. In Proceedings of the 1964 International Congress for Logic, Methodology and Philosophy of Science. Amsterdam: North Holland Publishing Co., 383-394.

218

Davidson, D. 1967, “Truth and Meaning”; repr. in Inquiries into Truth and Interpretation. Oxford: Oxford University Press, 1984. Davidson, D. 1974, “Thought and Talk”; repr. in Inquiries into Truth and Interpretation. Oxford: Oxford University Press, 1984. Davidson, D. 1977, “The Method of Truth in Metaphysics”; repr. in Inquiries into Truth and Interpretation. Oxford: Oxford University Press, 1984. Davidson, D. 1979, “The Inscrutability of Reference”; repr. in Inquiries into Truth and Interpretation. Oxford: Oxford University Press, 1984. Davidson, D. 1982a, “Empirical Content”; repr. in Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001. Davidson, D. 1982b, “Rational Animals”. Dialectica, 36, 317-327. Davidson, D. 1983, “A Coherence Theory of Truth and Knowledge”; repr. in Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001. Davidson, D. 1985, “Incoherence and Irrationality”. Dialectica, 39, 345-354. Davidson, D. 1988a, “Epistemology and Truth”; repr. in Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001. Davidson, D. 1988b, “The Myth of Subjective”; repr. in Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001, 39-52. Davidson, D. 1989, “The Conditions of Thought”. In J. Brand and W.L. Gombocz (eds.), The Mind of Donald Davidson. Amsterdam: Rodopi, 193200. Davidson, D. 1990, “The Structure and Content of Truth”. The Journal of Philosophy, 87 (6), 279-328. Davidson, D. 1992, “The Second Person”. In P. French, T. Uehling, et al. (eds.), The Wittgenstein Legacy, Midwest Studies in Philosophy, 17. Minneapolis: University of Minnesota Press, 255-267. Davidson, D. 1993, “Replies to Fodor and Lepore”. In R.R. Stoecker (ed.) Reflecting Davidson: Donald Davidson Responding to an International Forum of Philosophers. Berlin: De Gruyer, 77-84. Davidson, D. 1994, “Radical Interpretation Interpreted”. In J. Tomberlin (ed.) Philosophy of Language. Atascadero, Cal.: Ridgeview, 121-128.

219

Davidson, D. 1997a, “Seeing through Language”; repr. in Truth, Language, and History. Oxford: Oxford University Press, 2005, 127-141. Davidson, D. 1997b, “The Emergence of Thought”; repr. in Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001. Davidson, D. 1999, “The Centrality of Truth”. In J. Peregrin (ed.), Truth and its Nature (if any). Dordrecht: Kluwer, 105-15. Essler, W.K. 1988, “Open Philosophizing”. In Erkenntnis, 29, 149-67. Fodor, J. 2008, Lot2: The Language of Thought Rrevisited. Oxford: Oxford University Press. Fodor, J.A. and Lepore, E. 1992, Holism: A Shopper’s Guide. Oxford: Blackwell. Glüer, K. 2006, “Triangulation”. In E. Lepore and B. Smith (eds.), Oxford Handbook of Philosophy of Language. Oxford: Oxford University Press, 1006-1019. Goldman, A.I. 2006, Simulating Minds: The Philosophy, Psychology, and Neuroscience of Mindreading. Oxford: Oxford University Press. Hintikka, M.B. and Hintikka, J. 1986, Investigating Wittgenstein. Oxford: Oxford University Press. Lepore, E. and Ludwig, K. 2005, Donald Davidson. Meaning, Truth, Language, and Reality. Oxford: Clarendon Press. Lepore, E. and Ludwig, K. 2007, Donald Davidson’s Truth-Theoretic Semantics. Oxford: Clarendon Press. Ludwig, K. 1994, “First Person Knowledge and Authority”. In G. Preyer (ed.), Language, Mind, and Epistemology. Dordrecht: Kluwer. Ludwig, K. 1999, “Theories of Meaning, Truth, and Interpretation”. In U.M. Zeglen (ed.), Donald Davidson: Truth, Meaning, and Knowledge. London: Routledge, 27-45. Preyer, G. 2011, Donald Davidson’s Philosophy. From Radical Interpretation to Radical Contextualism, second edition. Frankfurt: Humanities Online. Preyer, G. 2011, Intention and Practical Thought. Frankfurt: Humanities Online.

220

Preyer, G. 2011, Back to Cartesian Intuition. Internalism, Externalism and the Mental (manuscript). Quine, W.V.O. 1969a, “Epistemology Naturalized”. In Ontological Relativity and Other Essays. New York: Columbia University Press. Quine, W.V.O. 1969b, “Ontological Relativity”. In Ontological Relativity and Other Essays. New York: Columbia University Press. Quine, W.V.O. 1990, Pursuit of Truth. Cambridge, Mass.: Harvard University Press. Ryle, G. 1949, The Concept of Mind. New York: Barnes and Noble. Wright, G. H. von 1994, Normen, Werte und Handlungen. Frankfurt a.M.: Suhrkamp Verlag.

TRIANGULATION IN ACTION: A RATIONALIZING PROPOSAL Ingvald FERGESTAD (University of Oslo, Norway) Bjørn RAMBERG (University of Oslo, Norway)

Davidson’s triangulation figure is notoriously difficult to explicate. At issue is not only what the triangulation thesis is and how purported arguments for it stand up to closer scrutiny. There is disagreement, also, about Davidson’s purpose, about what his philosophical aims are, and what position the triangulation figure is supposed to occupy in Davidson’s pursuit of a coherent grasp of our ability, as natural creatures, to think and to communicate. We briefly consider some revealing attempts to come to grips with Davidson’s use of the triangulation figure. These show quite persuasively, we conclude, that the triangulation relation, construed as a nexus of causal relations, will not mitigate worries about how to determine content, how to fix reference, or how to account for the difference between mere discrimination and genuinely intentional states. With this diagnosis in hand, we go on to suggest an alternative project in which the triangulation figure may serve. This project can be brought into view, we believe, by emphasizing aspects of Davidson’s views that resonate with Robert Brandom’s inferentialism. Thus, we hope to locate what we call a Radical Davidson (hereafter R-Davidson) by triangulating the more traditional, or conservative, rendering of Davidson with Brandom’s inferentialist position. R-Davidson is an idealization of a tendency that, while genuinely Davidsonian, exists in Davidson’s work in tension with the perhaps more pronounced, even dominant, naturalistic tendency, which seeks accounts of mind and meaning in naturalistically acceptable terms. R-Davidson, unlike the Quineannaturalistic version, is not at all concerned to explain how a creature might pass from discrimination to perception or from perception to propositional thought, nor with explaining how intentional relations might supervene on causal structures. R-Davidson is concerned to construe the conceptual as a certain kind of dynamic interaction. The significance of triangulation, for

222

R-Davidson, is to make a point about what kind of dynamic interaction this might be. 1. In a probing effort to get clear on what Davidson’s triangulation story is meant to establish, Peter Pagin takes the basic structure to be given by «a ‘triangle’ of two speakers and some third point, and relations of semantic import, which together help to determine the third point, also of semantic import» (Pagin 2001: 200, author’s italics). Pagin admits he is puzzled as to what kind of account the triangulation thesis is supposed to be. As a first pass, Pagin suggests triangulation may be «telling us something about what linguistic meaning, or thought content must be like […] [starting] out with a realistic picture of, say, the physical world […] and state general conditions on thought or meaning, in terms of relations between objects and pairs of thinkers» (2001: 200). However, this seems to leave us with a dilemma. Either an observer must have knowledge of the whole triangular situation in order to be able to single out particular objects or events as what she herself is reacting to, a requirement which means no ground at all is gained. Alternatively, if we accept that Davidson is not trying to reduce psychological or intentional concepts to causal concepts, the putative need for a second person to enable the first to single out a particular object or event as the object of her response is simply unmotivated. We cannot understand, says Pagin, why she is needed «except by attributing thoughts, awareness and knowledge to the two creatures to begin with» (2001: 205), i.e. by saying that in order to have thoughts we must have thoughts. There may well be a need for the second person, Pagin allows, but he cannot see why such triangulation is a condition for the application of intentional concepts when no reduction of the intentional is sought. Pagin suggests an alternative reading of triangulation, where we «start out with thoughts, and frame conditions of objectivity […] in terms of shared [epistemic] availability […] given that these subjects can know about the cognitive sharing itself» (2001: 201). For Pagin, this second construal expresses a form of idealism, suggesting that the objectivity of thought is secured by making interpersonal relations—sharing— constitutive of what it is to be an object or an event. He also rejects this way of understanding it, the core of which he takes as the claim that grasp-

223

ing the concept of truth requires communication with others. The upshot, for Pagin, is that the interesting difference between being on one’s own and being in a community cannot be stated without presupposing that community members possess the capacities that were considered problematic to begin with. What is actually determined by the intersecting lines in triangulation is not the object of thought, Pagin claims, but the concept of thought and reference, or the «aboutness relation between thoughts and objects […] [which] is that relation which, in triangular situations, best explains how creatures manage to identify what other creatures are thinking about» (2001: 211-212, author’s italics). Pagin treats the triangulation device as relating thinkers with objects and events in the world, either by rooting, putatively, the possibility of objective belief and thought in the world, or by rooting the possibility of objective truth in the communication between thinkers. A causal pattern of a particular sort would be the enabler in either case. If we take Davidson’s arguments as pertaining specifically to thinkers, Pagin’s criticism seems on the mark; it is difficult to see how adding a second person is of use, since what it is apparently intended to facilitate is already present. However, the triangulation figure may pertain to a more basic kind of situation. This possibility is underlined in Davidson’s slightly grumpy answer to Pagin. There are two basic situations in which triangulation may be discerned, Davidson maintains; a primitive situation that does not require intentional attitudes, and the learning situation, where the social environment provides some form of teaching to a creature not yet inculcated into thought and speech. He further claims that, «nothing in what I have described [as part of the primitive triangular situation] implies that interacting creatures are capable of describing what they are doing» (Davidson 2001a: 292). And regarding the second application, Davidson maintains that the purpose was not to give a reductive account of how the learning situation works, but to «indicate how the triangular arrangement makes the process possible» (2001a: 293). Indeed, it seems unlikely that the triangulation situation is an attempt to state conditions for how empirical thought content and objective truth is possible for creatures that are already thinkers, in the sense that one thinker would be required for another to have anything like a belief of the external world, or that one thinker be required for an object to be constituted as an object for the other. Of course inten-

224

tional concepts are fully applicable to isolated individuals, once we take those individuals to be thinkers and speakers. The point is not that I need another person present to confirm what stimuli I am then actually responding to and thus what object is at the end of our projection of the “stimulilines” outwards into the world. Having already learnt this, I can manage all by myself to identify the stimuli as being, say, a cow-sighting. But this ability of mine I owe to the triangulation structure of the learning situation. In our line of thinking about triangulation we will stick quite literally to Davidson’s two basic applications of it, and thus to the idea that triangulation is about how we are to think of what it is—for minded creatures—to learn to speak and think at all; what conditions must be met if linguistic performances are to be possible. The basic claim is that they are only possible or thinkable if they are from the very beginning, historically and individually, performances about something. This is a fundamental insight already in Kant, and one from which Brandom also starts off. According to Davidson, every person must have a causal history that traces back, directly or indirectly, to triangular experiences, which can be seen as, «aspects of our interactions with others and the world [that] are partially constitutive of what we mean and think» (Davidson 2001a: 294). Davidson does not think there can be proof of this claim: Its plausibility depends on a conviction which can seem either empirical or a priori; a conviction that this is a fact about what sorts of creatures we are. Empirical if you think it just happens to be true of us that this is how we came to be able to speak and think about the world; a priori if you think, as I tend to, that this is part of what we mean when we talk of thinking and speaking (2001a: 294).

In these passages Davidson is unambiguously clear at least on the point that the triangulation relation is to be taken constitutively. It must be admitted, though, that he is not so consistently clear as to the kinds of situations in which such a constitution relation is demonstrated. Thus, explaining the social nature of language, Davidson remarks: If you and I each correlate the other’s responses with the occurrence of a shared stimulus […] an entirely new element is introduced. Once the correlation is established it provides each of us with a ground for distinguishing the cases in which it fails. Failed natural inductions can now be taken as re-

225

vealing a difference between getting it right and getting it wrong, going on as before, or deviating, having a grasp of the concepts of truth and falsity. A grasp of the concept of truth […] depends on the norm that can be provided only by interpersonal communication, and, indeed, the possession of any propositional attitude, depends on a grasp of the concept of objective truth (Davidson 1994: 124).

This passage does seem to provide support for Pagin’s reading of triangulation. Davidson suggests that the second person is needed for the possibility of their being success-conditions for the application of words, and thus for sentences’ having truth conditions at all. While accepting the Wittgensteinian point Davidson is making, we believe that passages like this one also illustrate one source of difficulty in getting clear on the role of triangulation, namely, the difficulty in holding apart the perspective of the triangulating creatures and the perspective of the philosopher who reflects on triangulation-situations, and who, in doing so, sometimes observes and even illustratively takes part in them. It is not an individual thinker’s grasp of the concept of truth that is unthinkable without there being a second person present, once that individual person has learnt the possibility of error and thus objective truth. Rather, it is the philosopher’s grasp of the concept of truth that is not possible without a basic idea of interpersonal communication, of which triangulation is the fundamental form. This basic point leads us to restrict the constitutive role of triangulation to situations where the acquisition of thought and speech is at stake. Such situations may be framed both in a natural-historical perspective and in an individual learning perspective, as we will see. However, they are not designed to answer the question of how thought and speech, once acquired, can be about objects and events, and also be true about objects and events. We will return to this point later. The constitutive role of triangulation is highlighted in Jason Bridges’ critical but for our purposes also very illuminating reading of Davidson’s externalism (see Bridges 2006). The core externalist claim is the idea that causal relationships to things outside the body are metaphysically necessary conditions for thought; «any creature who can think at all must stand in causal relations to, say, water or people who talk about arthritis» (Bridges 2006: 291). The triangulation thesis is a specification of the form of the required causal relationship; thought requires «a causal nexus in-

226

volving two creatures and a third item in the shared physical environment» (2006: 291). Bridges thinks the conclusion questionable on various grounds, but his main concern is to undermine the premises of the thesis. The basic claim under scrutiny is one that Davidson repeatedly makes; «[if] we consider a single creature by itself, its responses, no matter how complex, cannot show that it is reacting to, or thinking about, events a certain distance away rather than, say, on its skin» (Davidson 1992: 263). Bridges then recasts Davidson’s argument in a more general form: 1. There is nothing in the scene of a creature inhabiting its environment, considered independently of our own judgments on the matter, that determines the creature’s behavior on a given occasion as a response to one of the events on the causal sequence producing it rather than another. 2. We, observing the creature’s behavior, judge it to be responding to the cause that we do—typically, an event involving a local, middle-sized object—simply because that’s the kind of event we find it natural to fix upon. 3. If the creature’s behavior is to count as a response to one of its causes rather than another, this fact must be partly constituted by another creature’s judgments about (or at least reactions to) that behavior (2006: 302-303).

Of course Davidson is right, Bridges comments, that it comes naturally to us to classify and group together certain external features; these dispositions are already programmed at birth. However, in a minimal sense natural dispositions condition all of our judgments, so there is nothing in this claim that could provide a basis for some distinctive thesis of our judgments about the object-directed responses of other creatures. The move from the fairly uncontroversial claim that our natural dispositions condition judgments about the object-directed responses of other creatures to the controversial claim that natural dispositions must bear the full burden of accounting for their being such objects is without warrant. It rests on the unfounded claim that when considered independently of a second creature’s triangulating response, nothing in a creature’s environment can settle which particular element in the causal stream behind its response behavior is the proper object of that response. Bridges cannot see Davidson providing any other kind of “evidence” for this claim other than that it will be evident to us once we take an honest look at the matter (2006: 307).

227

But the view on non-linguistic animals Davidson thus bestows upon us is insupportable, Bridges argues. The presumption that an animal’s responses are directed to something «out there» is «essential to our ordinary understanding of animal life» (2006: 307) and gives us all the reasons required, without any need of a triangular arrangement. The idea of a distal, object-directed response is built into our ordinary concepts of animal behavior; it is simply the way we think about animals, in the sense that it is part of what it is (to us) to be an animal. Bridges’ central line of argument is that our thinking about the objectdirected responses of any creature is an integral part of a conceptual pattern; «a larger conception that we constantly bring to bear in our observations and reflections» on animals and babies, and which involves a number of interrelated components, such as the idea of «perceptual awareness», of animals as «self movers», of «basic biological needs», and of «sensitivity to surrounding circumstances» (2006: 308). The conceptual, holistic pattern available to us for understanding the possibility of object-directed response means that we do not need to triangulate, as observers, with the creature whose responses we want to understand; nor do we need the notion of the presence of another creature to determine, through the notion of causally determined triangulation, what a creature responds to. The “larger conception of animal nature” is all the warrant we need for the factuality of judgements attributing object directedness. Davidson asks, «how minded creatures can come into their capacity to have responses directed toward objects in the world» (2006: 312). But, concludes Bridges, his «question cuts too deep. There may well be features of thought for which we can account only by considering the thinker’s relationship to others, but the fact that we thinkers can respond to things in the world is not one of them» (2006: 312-313). We propose to grant the fundamental point here, as we understand it. The task then becomes this; is there a question that cuts, in some pertinent sense, less deep, and to which triangulation may be (part of) the answer? Depth, in this context, seems linked to the nature of philosophical ambition. And it is possible to construe the triangulation figure such that it seems designed to support both an account of what object-directedness is and also an explanation of how creatures com by that ability—all in terms of interactive relations described in natural-causal terms. Our hypothesis is

228

that there is a sense in which Davidson does want to answer these questions. However, this cannot be in the sense of those questions as they are asked in the mouth of a naturalist wondering how it is that some causal relations come to be also intentional relations, and how some nodes in a causal network come to be objects of representation. The suspicion that Davidson’s triangulation figure may not be deployed to solve the sorts of problems faced by naturalist semantics may be further nourished through consideration of the objection that Dagfinn Føllesdal develops against Davidson. Føllesdal (1999) briefly considers what Davidson’s purposes in appealing to triangulation may be, and argues that the triangulation device is assigned the important job of introducing perception into Davidson’s picture of meaning and interpretation. On Føllesdal’s account, the motivation for appealing to the category of perception emerges in Davidson’s later writings. The early Davidson takes it as a fact that an interpreter observes the same objects and events as the speaker, and thus the interpreter’s access to suitable evidence against which to test her tentative truth-theorems is not questioned. The later Davidson, not willing to take the success of joint reference for granted, tries to mend the lacuna by introducing perception via the process of triangulation. Perception is accounted for in the form of a causal theory, where reference to the object or event perceived is secured by appeal to the idea of a last common cause in intersecting causal chains. Thus one could say that the later Davidson construes the triangulation thesis in order to unite the social and linguistic dimension of meaning constitution with the requirement that there be some account given of the empirical or real-world content of our concepts. However, Føllesdal argues, triangulation captured in terms of a distal causal theory simply isn’t up to the job. «The events with which I am familiar have a multitude of causes. One should rather talk about causal trees [than causal chains], branching out from the teacher and learner. One then sees immediately that the notion of the last common cause makes little sense. Which of the many junctions between our two causal trees is the last one?» (Føllesdal 1999: 725). Barry Stroud (1999: particularly 150ff) echoes this point. He, too, emphasizes that what an utterance means cannot be identified as the only cause of the utterance, not least because there are too many available causes along the causal chain.

229

Now clearly this critical point has bite—if our task is to define a particular causal pattern that will secure identification in naturalistic terms of a determinate object of reference. For in this case, a distinction between real perception of an object and differential response to stimuli, a distinction which anchors the very possibility of intentional attitudes, depends on the empirical tracing of causal patterns that lead to a definite point of intersection. However, it is possible to read the argument in the opposite direction. The idea of a well-defined point of intersection of causal chains is an idealization, one that abstracts away from the enormous complexity of the actual causal webs that tie organisms to other organisms and their common environment. If Davidson does in fact abstract away from this complexity, this can only mean that he is simply not interested in showing how it is that we may use naturalistically-framed knowledge of causal relations to determine the objects of our intentional states. If this is right, then Davidson would have no reason to disagree with Stroud when he emphasizes that it takes more than an utterance’s being caused for it to mean something, and more than that for it to mean what caused it. The point is that triangulation, considered as a causal nexus, cannot explain thought in the sense that the naturalist wants it explained—that is, placed as a phenomenon in a world structured by causal relations all describable without intentionality coming anywhere into view. In so far as the ability to form thoughts can be illuminated by asking what communication of thought requires, we will need richer resources in our description of the communication situation than a naturalistic view of triangulation affords. The way that creatures successfully communicate, and the way that they must rely on their surroundings in doing so, cannot be captured through the delineation in non-intentional terms of a causal pattern, no matter how subtle or complex. 2. This diagnosis provides a starting point for our attempt to redescribe the point of Davidson’s triangulation image. Its importance, we suggest, lies in the role it assigns to patterns of communication between creatures as activity or interaction, rather than in the role it assigns to the particular structure of the causal nexus made up of communicators and their objects of reference. Much of the criticism of Davidson’s use of the triangulation scenario is warranted, in so far as it is taken as a device for producing a naturalistic

230

conception of what intentional states are, for reference fixing, or for accounting for intentionality in causal terms—that is, for explaining, by appealing to a particular causal structure, how we as individuals come to be placed so as to form objective concepts of and entertain true propositions about objects and events in the world about us. If this were the aim of Davidson’s appeal to the triangulation figure, then criticisms like those we have rehearsed would certainly be well taken, and probably devastating to the project. An alternative possibility is that the triangulation figure is offered as something far more modest and, in a sense, more primitive than a naturalistically explanatory model or account. Although a figure pertaining to a basic condition for the possibility of acquiring thought and speech, triangulation does not illuminate any particular individual’s acquisition of thought and speech. Rather, it may be taken to indicate something about concept constitution. We propose to read Davidson as attempting to show that concept constitution in this sense must be conditioned by more primitive patterns of communication between creatures and between them and their environment; indeed, that we must be able to see a basic outline of concept constitution within such primitive forms of communication. To make this proposition viable, however, we must make more out of triangulation as active, social interplay and less out of triangulation regarded simply as a causal nexus. We think that the fundamental idea Davidson sets forth with the triangulation figure, when seen as a basic idea about concept constitution, may be sustained within this perspective, and that it actually makes better sense of his insistence on its a priori character. This idea or proposition may seem close to Bridges’ alternative to triangulation, viz. a conceptual holism of what it is to be an animal (or for that matter a learning child). The difference is that triangulation, as part of our “space of reasons”, is on a far more basic or primitive level than the holistic, rational conception of sentient animals that according to Bridges guides our ability to identify their object-directed responses. A way to frame the tension we are addressing is to say that Davidson seems to be attempting to fuse two quite different vocabularies. The toodeep question of Bridges’ diagnosis actually requires a simultaneous application of both—which is exactly why the question cuts too deep. The naturalistic Davidson ends up submitting a basic vocabulary of rational social

231

behavior, in which triangulation is described as communicative, responsive interaction between triangulating creatures, to the terms of a basic naturalistic vocabulary in which triangulation is described as a nexus of causal processes. Submitting the one to the terms of the other may facilitate the naturalist questions, but it also puts the answers out of reach. Within the naturalistic-causal vocabulary a Quinean proximal theory is the natural consequence, while Davidson presents the change to a distal perspective as one that can do the work of fusing the two vocabularies. But this change from a proximal to a distal perspective does not seem to be able to do the job, as long as triangulation is being viewed primarily as a purely causal process. Davidson’s line of reasoning then ends up either as an attempt at reductionism, which he himself clearly rejects, or as having to presuppose what should be demonstrated. We do not intend to object to the common criticism of what we have here called a conservative Davidsonian position on triangulation. As we see it, it hits its mark rather well. Our mission is instead to ask how the triangulation figure appears if we construe claims about the causal nexus as made from within the perspective of a pragmatic-rationalistic vocabulary (as we will call it, alluding to Brandom). Could we then maintain the claim that triangulation outlines a basic, a priori condition of concept acquisition—in the sense of concept constitution rather than individual concept acquisition—one without which we simply could not conceive of concept constitution at all? And if so, could we still claim that what we had found would be sufficiently reminiscent of Davidsonian triangulation to defend that label? A creature or person having already learnt to identify objects and events in the vicinity obviously can do this completely on its own. But could it do so from the beginning? Or more precisely, is it possible to see concept constitution as conditioned by the basic social processes or games within which we as individuals acquire the world as our own, as a horizon of claims and interests and practical concerns in the manner that is distinctive for thinking agents? We believe there is an essential question here that is not asking about the naturalistic conditions of the possibility for thought to arise, or under which representational relations may obtain. We hope to bring out the sense of this question about what is distinctive of a thinking relationship to the world through a triangulation exercise of our own; taking Conservative Davidson as one sight point and Robert Brandom’s posi-

232

tion on concept constitution as the other, we hope to locate a position that provides a convincing setting for the triangulation figure. Davidson’s claim that a sentient creature, or a child not yet inducted into the practice of thought and speech, could not know by itself what in the environment it was responding to, should not be taken as a claim pertaining to its relation to the causal stream of stimuli hitting its senses. Rather, we suggest, let us take it as a claim to the effect that the child would not pay particular attention to a table or a chair, as long as these objects do not play any particular role in the child’s world. This may not be so for a lion cub, since we cannot be sure of the extent to which its ability to pick out prey as salient is prewired and to what extent that ability needs to be developed as hunting behavior through interaction with the mother lion. But this difference is not significant as long as it actually is a condition of concept constitution we are looking for. For in this perspective the animal similes should rather be looked upon as purely illustrative, and not as revealing anything at all about what the animals themselves learn when triangulating. If we were to find that all creatures—lions, dogs, mice, and children alike—are prewired to pick out distal objects or events as salient, then, sure enough, the triangulation figure would have no epistemic significance as a reference-fixing gambit. But from the point of view of an interest in the constitution of the conceptual as such, such a discovery would have no bearing. 3. This last point requires elaboration. With that in mind we turn now to Robert Brandom’s account of the significance for the constitution of thought of reliable differential responsive dispositions to sensory impressions from the environment. Brandom, we believe, offers a useful contrast to the Consevative Davidson’s conception of the role of causally produced responses in triangulation relationships. Brandom certainly acknowledges that world-mind causal connections producing sensory impressions have significance for intentional agency, as do the causal chains by which mind produces action in the world. The incorporation into language of objects and events in our environment, or what he sometimes calls the “corporeal” or “lumpy” items of the world, is the result of causal processes by means of which impressions of these items are fed through our senses into our

233

linguistic inferential processes in the form of “non-inferential reports”. The contents of a discourse in a typical game of giving and asking for reasons «are not merely placeholders in abstract, purely formal, relational structures […] Discursive practice comprises non-inferential entries and exits as well» (Brandom 1994: 631-632). And: «Solid discursive practices incorporate non-linguistic things in them (are corporeal)» (1994: 715, n28, author’s italics).1 This is, according to Brandom, the reason why the conceptual contents of a community’s discursive practices sometimes may outrun the community’s ability to apply the concepts correctly, in the sense that the incorporation into language of corporeal items may be constrained by those items in ways that may sometimes conflict with the application of the concepts by which the community believes they incorporate them. This is also a basis for understanding Brandom’s approach to the problem Davidson set out to solve by means of the triangulation figure: that is, to account for a necessary externalist condition for the possibility of conceptual content being both objective and objectively true. Brandom endorses Davidson’s view that a creature can be recognized as having a belief only if it is in practice able to distinguish between true and false belief. And making that distinction requires that «the putative believer acknowledges in practice the objective representational dimension of its content—that it is being held is one thing, but its being correct is another, something to be settled with how it is with what it is about» (1994: 152, author’s italics). Settling this is only possible in a context of social, notably linguistic practices, involving interpretation of the speech acts of other persons, assigning propositional attitudes to them, and assessing the correctness of the conceptual contents of their assertions. Davidson thus in effect merges the conditions of objectivity of belief with the conditions of eliciting and assessing those attitudes in interpretation as a social, linguistic activity, in so fart as objectivity is dependent on the possibility of an assessment of what in the external world putative beliefs are about. This Davidsonian insight into the coeval conceptual status of intentional states and speech acts contributes to sparking off Brandom’s opus magnum, in so far 1

On an interpreter’s view a community’s application of concepts thus is «in part determined by the actual proprieties of and facts concerning the things that the linguistic practitioners are perceiving, acting on, and so talking about» (Brandom 1994: 632).

234

as the philosophical work undertaken in Making It Explicit «focuses on the development of an account of linguistic social practices within which states, attitudes, and performances have, and are acknowledged by the practitioners to have, pragmatic significances sufficient to confer on them objective representational propositional contents» (1994: 152) . At this same point, however, we may locate a significant difference between Brandom and Conservative Davidson. According to the latter, a basic, necessary condition for an externalist conception of the possibility of objective thought and belief is to be found in an account of the constitutive force of the causal relations obtaining between the world and the communicating creatures’ responses to the effect on their senses. This suggests that Davidson, as long as he is taken to understand the externalist requirement in terms of causal constitution, is offering no principled difference between sentient and sapient creatures with respect to the possibility of alertness to the difference between error and objective correctness. Brandom, by contrast, while maintaining an externalist perspective, immediately restricts the possibility of such awareness to sapient creatures, in so far as the main condition for the possibility of objective thought and belief is to be found in an account of the constitutive force of the response of a creature to environmental stimuli when that response is being stated and taken as a commitment—a commitment that can both serve as and stand in need of reasons, and not in an account of the reliability of “noninferential” responses to caused sensory impressions. This is so, at least, if the response is to be taken as true belief or as knowledge. The key point is that such commitments can only be stated, interpreted, and assessed in linguistic, inferential discursive practice. This is Brandom’s position in a nutshell, but we shall of course have to consider his view more closely in order to triangulate with Davidson. At the basis of the ability to undertake inferential commitments is what we must take as a fact; that we are equipped with reliable differential responsive dispositions to sensory impressions of the world. This fact is to Brandom a necessary but very basic condition for observational knowledge, and of course not at all sufficient (see 1994: 214). In that sense it is reminiscent of Davidson’s claim that triangulation is no more than a very basic necessary condition for the possibility of objective thought and belief. The difference is, as we have remarked, to be found in the different

235

roles the two assign to the responsiveness of creatures. Davidson discusses the role and significance of responsiveness in a naturalistic perspective, as a characteristic of any creature that is causally affected by its environment. Brandom restricts his discussion of the role and significance of responsiveness to a pragmatic-rationalistic perspective, as pertaining only to sapience—in short our capability to convert our reliable differential responses into inferential commitments in a “space of reasons”—while holding that sentience is merely an exclusively biological phenomenon construed as awareness. Though we may note that even sentience requires more than the most basic presupposition of causally-constituted content, that is, «the mere reliable differential responsiveness we sentients share with artifacts such as thermostats and landmines» (1994: 5). A critical distinction is drawn between the reliability of the kind of belief-forming mechanisms that Brandom calls non-inferential reports on the one hand and the authority such reports acquire once they are stated in an inferential game of giving and asking for reasons. Only when they are being so stated, and we can undertake commitments to their inferential conditions and consequences, are we entitled to assess them as candidates for true belief or knowledge. Brandom extends, though, the classical epistemological conception of knowledge as justified true belief—typically in an inferential game of giving and asking for reasons—to comprise also certain aspects or cases of reliabilist epistemology, i.e. incorporating cases where we might be justified in saying that there is true belief or knowledge without being able to cite explicit reasons. There are cases, he admits, although few and special, where we would say that someone responds differentially to some sort of stimuli in such a reliable, consistent way that it might qualify as true belief or even knowledge to an observer, regardless of whether the person forming the belief considered herself a reliable reporter or not. However, the authority of even these kinds of noninferential reports requires not only that they arise from the exercise of reliable differential responsive dispositions, but also that the response allows endorsement of a claim or commitment with a certain (conceptual) content.2 2

For Brandom’s account of the potential and limitation of reliabilist epistemology as an approach to the forming of belief and knowledge, see e.g. 1994: 206-213, 2000: 97122.

236

It is the presence of such authority that allows us to say that noninferential responses and reports become objective, imbued with conceptual content—in Brandom’s frequently used example; that physicists can say that it is really a mu-meson they observe (or muon, as this elementary particle is now called) and not just a hooked vapour trail. Brandom thus treats a «reliably differentially elicited response as conceptually classifying the stimulus to which it is keyed just in case that response occupies a position in an inferentially articulated space of claims that can be offered as, and stand in need of, reasons» (1994: 618, author’s italics). This is the background against which we must also understand Brandom’s own discussion of triangulation, which in Making It Explicit is part of a semantic discussion of singular terms and substitution (1994: 426432), and in Reason in Philosophy is part of a discussion of empiricist conceptions of concepts (2009: 188-196). Brandom sees triangulation strategies as arising from considerations of how we are able to discriminate stimuli to which some sort of response can be reliably keyed. He accepts, as we have seen, that proximal theories offer reliability in linking response to stimulus, but rejects them as “disastrous” because they only enable identification of states of the responding organism, while what we are after, of course, is responses that reliably and differentially classify objects and events in the outside world. For this Brandom, as Davidson, calls for a distal strategy, typically developed by means of the triangulation figure. However, he does not refer to Davidson, but to Fred Dretske (1981) when considering triangulation strategies concerned with the most basic forms of reliable differential responses, such as are instantiated in thermostats. His discussion of Dretske, the particularities of which we will not comment on here, concludes with the claim that Dretske’s triangulation gambit, drawing on intersecting pairs of causal chains from room to thermostat, does not ultimately succeed in fixing the stimulus distally rather than proximally. Brandom treats this failure as an instance of a general failure of the strategy; reliable differential responsive dispositions, harnessed to causal chains of covarying events that reliably culminates in a certain response, are in principle insufficient to identify or classify the object to which the response is keyed (as an object, we might add). The conclusion is the same as above: classifying something as an object is a conceptual classification and requires that the response is brought

237

into the space of reasons; it cannot be done on the basis of a pure, empiricist strategy. Until and unless it is brought under conceptual classification, a non-inferential elicited response, regardless of its reliability, will not qualify as identification or classification of whatever it is a response to, as an object: «What picks out one kind of thing as what is being reported out of all those that are being differentially responded to is a matter of the inferential commitments that response is involved in» (1994: 430). Causal triangulation, Brandom argues, which is associated with reliable differential responsive dispositions, must be supplemented; content requires a response to be placed in an inferentially committed discourse, in a pattern of what presupposes it and what it entails, that is, a pattern of chains of propositional antecedents and consequences. This in effect yields another form of triangulation; which may be called inferentially articulated triangulation. Brandom depicts it thus (excepting “purely theoretical” concepts, that is, those with no non-inferential application conditions): «all concepts in this way answer to two masters: their own non-inferential circumstances of application and their inferential links to the non-inferential circumstances of application of other concepts» (2009: 195). This very sketchy account of Brandom’s perspective on the kind of stimulus-response thesis that Davidson discusses by means of triangulation, suffices to demonstrate that neither causality nor triangulation as such pulls weight in Brandom’s account of the possibility of thought, in the sense of conceptual content, comparable to the load it must carry for Davidson. Appeal to mere causality contributes little to understanding a creature’s ability to identify or classify something as an object or event, regardless whether we take a distal or a proxy approach. In Brandom’s philosophy our reliable differential responsive dispositions are nothing more than what we might call entering and exit gates of causal stimuli, and as such keyed to responses that “translate” the stimuli into conceptual classifications carried out in discursive, inferential practice. Brandom moves our attention from the causal, which is not assigned any constitutive role, to the inferential practice of giving and asking for reasons in discourse, which is what enables us to classify and “report” a non-inferential elicited response as an object or an event. It requires of course that the response can be held to be reliable and differential, and thus it might seem appropriate to call this requirement a basic condition of the possibility of conceptual content.

238

But then it would be so more in the sense of being a basic fact, on which we have to rely without reason or argument, about how our senses and our brains work, rather than a basic constitutive aspect of thought and concept formation. Even if Brandom sometimes names our non-inferential responses “protoconcepts”, he actually does not in his arguments try to give them any kind of role as primitive forms of concepts. To Davidson our basic responses to sensory stimuli—once they are responses not only to stimuli from the external world, but also to stimuli emanating from others’ corresponding responses to the external world in triangulation—have the potential to make the triangulating creatures aware of the possibility of error, thus of the possibility of true and false belief, and thus of the possibility of objective conceptual content. To him such primitive responses in triangulation are conditional of thought in so far as they make up a structure or pattern that also applies to concept formation or constitution. To Brandom, by contrast, concept constitution solely and uniquely occurs in linguistic, inferential, discursive practice within the space of reasons. A response, say to some red thing, is not a candidate for thought-constituent or conceptual content until it is given the kind of normative authority that it is imbued with in inferential discourse, i.e. in the kind of practice in which we may undertake commitments as to the premises and consequences of stating that something is red, and in which we may thus be entitled to state it. 4. Our attempt to locate an alternative position for Davidsonian triangulation—R-Davidson—takes off from the difference we have just sketched. On the one hand, Brandom’s assignment of a constitutive role for social discursive practice with respect to the possibility of thought and concept formation, by virtue of the demand that the practitioners be committed to the proprieties of their practice, changes the significance that attaches to the causal nexus of triangulation. While we may well, on Brandom’s picture, retain triangulation as an identifying device, we are relieved of the need for assigning it the constitutive role for thought that Conservative Davidson confers on it. We ought to accept this relief, not only because of the problems attached to constructing a sustainable distal theory of stimulus and response, but also because of the problems that emerged from the

239

attempt to integrate the naturalist conception of a causal nexus with the normative perspective of assessing or judging something as true or false, correct or incorrect. In a more Davidsonian wording we might say, perhaps, that Brandom suggests the possibility of interpreting the triangulation nexus as a pattern of interaction between rational communicators— triangulation as a form of rational practice—in place of a reading that emphasizes the aspect of (mere) causal regularity. On the other hand, Davidson’s insistence that we see a basic, conditional pattern of thought constitution even in primitive processes of communication and learning, paradigmatically exemplified in processes where children learn to speak, expresses an acknowledgement of world dependence that we would like to preserve. Indeed, it may mitigate concern that the outside world—the “corporeal” or “lumpy” parts of discourse—is being left out of an account of the constitution of thought. Brandom does hold, as we have seen, that an interpreter must take into consideration that corporeal things determine what interlocutors are talking about and therefore may constrain conceptual content even to the point of correcting our application of a particular concept. But he is not concerned to specify or provide examples of how the integration of corporeal items in linguistic performances actually may affect concept constitution. Nor does he appear to think of concept constitution as involving any kind of direct interaction with the corporeal parts themselves; for Brandom, the determination of what those things are that we are talking about takes place within our discursive, inferentially governed practices. Corporeal items morph, so to speak, into linguistic entities, occupying positions in inferential space structurally defined by patterns of giving and asking for reasons. The position we are aiming to locate will play down the role of causal regularity, emphasizing instead the dynamics of the triangulating practice between the creatures involved. The critical question is what sort of significance the triangulation figure may then retain in our efforts to come to grips with the possibility of thought formation or thought acquisition, in the sense of initiation to the (conceptually structured) world. Brandom makes it clear from the beginning of his work that it is the very discursiveinferential practice as such that is constitutive of concept formation, and not, say, the rules or proprieties or regularities of the practice:

240

we are characterized not only by normative statuses, but by normative attitudes—which is to say not only that our performances are correct or incorrect according to various rules but also that we can in our practice treat them as correct or incorrect according to various rules … saying that Kant’s principle [that we act according to our conception of rules (IF/BR)] focuses demarcational interest on the normative attitudes exhibited by the activity of assessing, rather than just on the normative statuses being assessed […] assessing must be understood as something done; the normative attitude must be construed as somehow implicit in the practice of the assessor, rather than explicit as the endorsement of a proposition (1994: 33, author’s italics).

This is exactly the point that we could like to bring to bear; is it possible to see the triangulation figure in a similar perspective of practice, of triangulating activity rather than (causal) triangulation structure, and still keep Davidson’s insight that we must be able to see some form of condition for thought and formation of concepts in the basic relation between creatures and between them and their common environment? As indicated, the common ground is that we must take it as a kind of pre-conditional fact that the environment affects our senses causally and that we are equipped with reliable differential responsive dispositions to receive those caused stimuli, such as seeing a red house as a red house, yesterday, today, and tomorrow, or hearing the chiming of a bell as bellchiming in similar ways, etc. Unless we grant such persisting similarities, it is hard to see how communication about responses—up to the stage of propositional communication and interaction—would be possible. At any rate, neither Davidson nor Brandom is inclined to place in question the legitimacy of this ground. But this fact, as Brandom’s critique of naturalistic notions of content brings out, does not by itself play any further conditional or constitutive role in the formation of thought, belief, and concepts. It cannot make it clearer to us how concept constitution is possible, or what it is to acquire thought and belief. The issue at stake for us here is what kind of philosophical gain that externalism is supposed to secure. The significance that Davidson assigns to the triangular pattern of causal regularities is supposed to provide assurance that our thoughts and words are linked constitutively to the relevant features of the external world. Can this assurance be maintained? Perhaps, if we are able to see our way to an ex-

241

ternalist perspective that depicts reason and discourse as fundamentally integrated in linguistic practices that are also genuinely “corporeal” and involves non-linguistic items as part of the practices themselves.3 Their being so integrated is in a certain sense the original Zustand; any description of reason and thought processes as separate processes are the result of analysis in the sense of separation through abstraction of elements that naturally belong together as a unity. This claim is fully in line with Davidson’s point that thought and world are not relata standing over against each other in a problematic relation, but basically an integrated unity (e.g. 1975: 164). The question is then whether we are able to see this unity of the world and the possibility for concept constitution as coming to expression in triangulation construed as a communicative process. We claimed earlier that it seems natural to think that a child who is not yet speaking would not pay particular attention to a table or a chair by itself, as long as these objects do not play any particular role in the child’s world. We believe this simple point may yield a clue to an understanding of Davidson’s triangulation figure, when taken in a Brandomian spirit, as a figure outlining a basic communicative practice. What causes a child to associate the sound “table” with the sighting of a table is not merely, probably not even primarily, his mother’s uttering the sound while simply pointing to the object. It is, rather, the beginning of an introduction of the significance of tables in the child’s world; the significance of it as a thing on which food is served and food is spilled, as a place for gathering the family, a place to read newspapers, to talk or quarrel, as a thing under which the child can hide, or the dog can be found sleeping. The sound of the word “table” is gradually being endowed with the meaning of the word “table” through the way the child is introduced to and trained in using it— introduced to and trained in “table practices”. Or on a more Wittgensteinian note: the simple, ostensive conditioning process described by Davidson is only a tiny fragment of all possible moves in introducing the 3

An experienced skeptic would naturally find a host of arguments against the idea that causality or the “integration” of reason and discourse in corporeal practices could constitute anything like a “guarantee” for our thoughts and beliefs being about the external world. We do not intend to discuss possible consequences of our discussion for skepticism here, but maintain that the perspective on Davidson we are trying to explore may make it easier to back away from skeptical reasoning. See also Ramberg 2001.

242

child to the language game of tables. Davidson highlights the conditioning aspects of this development as a causally conditioned process. He does not, maybe just for that reason, pursue the practical aspects of how the meaning of the word “table” is gradually constituted for the child. We may call the first phase of an idealized introductory game a kind of conditioning game for the following reason: At the beginning we may look upon the learning of words as not principally different from learning simple games, e.g. how to fit wooden blocks of certain shapes into holes of similar shapes. Initially the child performs a certain action—utters a sound, directs a block at a hole—and the “device” of the action may or may not fit a particular aspect of the environment (use of table, use of hole). There are no meaning components in this situation. The child may fail to do what we know is right in so far as it has not yet “seen” the specific connection between device and object. In this case, however, it can hardly be said even to try to get it right or to be aware that it does not. It simply does not yet have an idea of “error”. It doesn’t “ponder” or “consider” or “doubt” whether the sound fits this or that object, or whether the block fits this or that hole. Its moves are not yet moves in a game; the massive background of purposes and expectations required for anyone to grasp that some response is wrong is not yet there. “Getting it right” would then be a mere coincidence and not problem-solving. A space has been made for the child to do things rightly or wrongly, but it has not yet filled the space. As long as the child is not in a position to doubt (and as long as no one purposefully interferes with the learning process in order to disturb it) we, as observers or even fellow triangulators, can say that it has no alternative but to gradually learn to fix the sound “table” to the presence of corporeal tables. It may go wrong a number of times, but still has no alternative; through encouragement and sanction it will learn to do it right. The child may still have no awareness of there being a norm embedded in what it is doing (which is therefore not yet an “action”). When does this awareness occur? When has it learnt to play the game? When does it occur to it that the sound is a word; that it can represent corporeal tables and therefore be used both rightly and wrongly? It has begun to acquire the skill of the block-fitting game not only when it is able to fit the right blocks in the right holes, but when it is also able to correct itself and change block or hole if erring. It has then acquired awareness of the norm of block fitting; it

243

“knows”—at least in a certain primitive and “reliabilist” sense of the word which does not imply that it knows that it knows—when it does it correctly and when not. There is then sufficient agreement—in the sense of sufficiently many instances of correspondence in action between the child and the trainer (but not in the sense of a willed agreement of course)—for the child to become aware that it, or another child, can err and use the blocks in the right or the wrong way. Our suggestion is to see the learning of the use of words by conditioning the use of sounds in a similar way. At this stage, however, we still cannot say that the child acts intentionally; that it has acquired the concept of tables, or block fitting, or whichever word it has been trained to apply. The idea is that we see the triangulation figure as depicting a basic conditional pattern of thought acquisition precisely at this level; enabling an understanding of how children are being initiated into speaking and thinking by learning “table-practices”, “washing practices”, “drawing practices”, “noting bird song practices”, etc, etc; practices that are required for the words “table”, “washing”, “drawing”, “bird song”, etc. to be endowed with meaning. Ostensive learning is just a very small fraction of those practices. If we look at sentient creatures, a similar picture seems to emerge. A lion mother kills a gazelle and brings it to her kids to eat, smell, and learn that gazelles are food and thus prey. Later she will take her cubs to train them in the game of hunting down gazelles. This can be described as a triangulating practice or game as well; the cubs are trained to triangulate with their mother—and gradually with other lions—in order to identify and locate gazelles and other prey. When trained, a lion can see from the responses of other lions that there is a gazelle in the neighbourhood, even without seeing or smelling it, and can thus use its triangulating experience and training to locate the prey; looking and smelling in the direction pointed out to it by the responsive behavior of the others. It goes without saying that the purpose of such a training and hunting game is to introduce gazelles as essential elements in the lion cubs’ world—as prey. Davidson’s animal examples are however different in a principled way from the example of the child learning words (or block-fitting) in the sense that what we may learn, as observers of lion cub behavior, is: (a) biological and behavioral knowledge of lion and animal behavior (we will never get

244

to know what acquisition of a lion’s world would amount to for a lion), thus developing our biological and behavioral vocabularies further, precisely in the sense described by Bridges as developing our holistic conceptions of animal behavior; and (b) the philosophical insight that the basic communicative pattern we find in how animals (by Davidson’s description) are being conditioned and trained to pick out what is salient to them in their world, is basically the same kind of pattern as we find in how children are being conditioned and trained to pick out, say, tables as salient to them. This last point therefore gives us as philosophers the suspicion that the triangular, communicative pattern that allows Davidson to describe the learning process as a process of acquiring a world—of saliences and significances—is not something we “find”, for instance through empirical observation of such processes. Rather, we may take it to be a rational pattern by means of which we are enabled to see the significance of such conditioning and training behavior for the understanding of the possibility of concept constitution, or of objective conceptual content. This amounts to nothing less than agreeing with Davidson that we should see triangulation as having an a priori status, and even to affirm the ambition that Bridges identifies; that the triangulation “thesis” is supposed to rest on a kind of transcendental argument—a “transcendental externalism”. The triangular pattern, in this perspective, is assigned a transcendental role in the philosophical argument that we find in Davidson’s triangulation, in the form of a condition for the possibility of grasping the basic processes or practices of social games through which we are enabled to acquire thought and speech. This is true both in so far as it pertains to the initiation of individuals into the world as located in a space of reasons, as well as to the manner in which these practices and games themselves contribute to maintaining the world as conceptually structured. The perspective so introduced enables a translation of the triangulation figure from Davidson’s fixed, causal perspective to the socio-dynamic perspective inspired by Brandom on the significance of triangulation as practice; something we do when learning to respond correctly to stimuli through conditioning processes, pick out the salient responses, and then gradually learn to assess these saliencies, state our assessments as propositions, and assign similar intentionality to others. Translating the triangulation figure in this way is, however, not solely the result of bringing in

245

Brandom’s socio-dynamic perspective as something totally new to Davidson’s account. As we have already seen indirectly, this perspective finds strong support in Davidson’s examples and descriptions of actual triangulation, if not so much in the more general parts of his account. When the child is introduced to tables as salient in its world, it is through what we may call an ostension “game”, and even if ostension is but a small part of the game in which tables are endowed with significance to the child, the interesting aspects of the example is the description of what the child and the mother do when playing the game, far more than in the casual roles of the table, the finger, the sounds, etc. The animal examples bear the same mark: the emphasis is on what the lions do more than on the causal nexus of their triangulation (the relevance of which is not weakened by the illustrative role of these examples). We therefore believe that there is a stronger resemblance between the socio-dynamic perspective we have brought in through Brandom’ inferentialist account of conceptual abilities and the socio-dynamic flair of Davidson’s actual triangulating stories or examples than is typically acknowledged by the commentators we have discusses. Nevertheless, we acknowledged above that we do want to preserve Davidson’s insight that the basic relation between creatures and between them and their common environment offers a condition for thought and formation of concepts. This is a stronger claim than Brandom’s basic fact, and one may even offer a somewhat clearer picture of the link between this “basic fact” and the inferentialist games of giving and asking for reasons than Brandom is concerned to offer. To meet this aim, we need to consider the sense in which the socio-dynamic learning or training practices of triangulation are what Brandom calls “corporeal”. To Conservative Davidson, a central point of triangulation is to make evident that the content of thoughts and beliefs are the actual corporeal items of the world, located where the outward lines of the triangulating creatures cross. Framed in terms of a pragmatic-rationalist perspective, this is to insist that the learning practices or games are not purely discursive, as they appear in Brandom’s construal, but include the introduction and integration of corporeal items such as tables and soaps as salient in a child’s world and therefore as non-eliminable parts of those practices or games. The gradual conditioning process into using a sound correctly, and thus beginning the transforming of it into a word, is integrated in activities in

246

which certain objects and certain events are being made salient to the child through the right (or wrong) use of those objects and events. The use of sounds is but one integrated part of the use of the objects or events. For this transformation to take place—for sounds to become words, concepts, and sentences that can be true or false, or for actions to become right or wrong—the essential element, the very core of Brandom’s conception, is still lacking. For the transition to happen, the child must learn or grasp the fundamental normative character of intentionality. We asked above when the awareness of the possibility of error occurs; when the child has learnt to play the game. And we said that it begins to acquire the skill of playing the game when it is able to correct itself and change sound or block or hole if erring. It has then entered a space of awareness of there being a norm, but we can still not say whether it just performs the movements reliably without having any belief, knowledge, or conception of what it does, or whether it has acquired the concept of tables or of block fitting. So when are we entitled to say that it has moved from conditioning to acquiring the concept? Davidson holds that we cannot retrace this transition from non-intentionality to intentionality, from just reacting appropriately to coming to be able to judge about the world. It requires, he says, that «the base line of the triangle […] is strengthened to the point where it can implement the communication of propositional contents» (1997: 130), or where it becomes «as thick as language» (1999: 731). But he admits that it is «probably beyond our powers to tell» (2001b: 15) how the room of awareness of error and truth conditions is filled in the learner. Now, from the Brandomian perspective we are bringing to bear, this is quite obviously true. Once inside the space of reasons it is impossible to retrace our steps backwards, to some origin outside this space, at least without taking a reductionist path: the access to our non-intentional origin as individuals is blocked far back in our educational history. Similarly, the non-sapient origin of humankind is long lost in our collective history of an ever-widening space of reasons. Brandom takes the consequence of this insight by keeping his account of the conceptual strictly within the space of reasons. The lesson to be learned for the Davidsonian is that this inability to backwardtrace should not be construed as a concession, as a failure to deliver on an explanatory ambition. Rather is should be taken as a reflection of the dis-

247

tinctiveness of the pragmatic-rationalist perspective, of the form of existence characteristic of thinking agents. Still, as we have indicated, it is possible to say something about what is particularly characteristic of the form of existence of thinking agents once the transition from conditioned to intentional behavior has taken place. This particular characteristic is simply that the child is capable of acting normatively, both in the sense of being committed to reason such as in the proprieties of discourse, and in the sense of being committed to acting morally right. From a sapient perspective, an animal doesn’t perceive the world in true or false ways, or behave rightly or wrongly, normatively speaking. A lion that errs in behaving as if a gazelle were grazing in the high grass or in hunting one down, just fails; it doesn’t perceive the world “falsely”, and it doesn’t do anything “wrong”. But a child, when being rewarded for using the word “table” correctly or punished for using it wrongly, has begun the laborious process of acquiring the sense of what it is to be committed, in Brandom’s terms, to speaking correctly and acting rightly; committed to acknowledging that sentences such as “The table is heavy”, or “The lion is hunting a gazelle” have truth-values that are independent of the child’s uttering it. As Brandom often emphasizes, it is the hallmark of sapient creatures that we are committed out of a free will to whatever reason demands from us, say, in a discourse, regardless whether abiding with that commitment should lead us to acknowledge that what we have believed in, as individuals or as collectives, is false, ill-conceived, or without credibility. Acknowledging this fundamental normative character of triangulation as a practical, social-dynamic figure implies that we must see it as embedded in any directive response given to a child who is beginning to acquire a world through the transformation of sounds into words and concepts, but also that it is always and only up to the child, as it is entering the space of reasons, to actually undertake the commitment. If the training doesn’t succeed in making it do so, there is nothing more the mother or any other triangulating person can do. This acknowledgement is the final element in our triangulation of Conservative Davidson with Brandom. The point that the original triangulation figure must be fundamentally normative if we are to see it as a condition of concept constitution, seems to have taken us further than Davidson himself

248

would have been inclined to go. As we have acknowledged earlier, Davidson never intended to give some form of explanation of the actual acquisition of thought and speech; he just wanted to show that it must be conditioned by a more basic and primitive interrelation or interaction between the acquirers and their surroundings. Still we think that even an account of the most basic and primitive pattern of interaction, be it triangular or not, between two or more creatures and their environment must demonstrate the possibility of normativity if it is intended to say something about the conditions for concept constitution. 5. At the beginning of our inquiry we introduced the idea that there are two antagonistic tendencies in Donald Davidson’s philosophy. His naturalist propensity brings him close to analytical philosophy in a broad sense and to Quine’s version of empiricism in a more specific sense. We have called this tendency the Conservative Davidsonian position. The “radical” position, while more concealed in his work, shows us a Davidson of a more Continental, rationalistic, we might even say transcendental bent, insisting on the insuperable role of reason, as what we today would call a social phenomenon, for the understanding of how it is possible for human beings to acquire thoughts, beliefs, and knowledge of an objective world. We believe this Davidson, R-Davidson, is more closely aligned with Brandom’s modern version of philosophical rationalism than with Quine’s scientistic version of empiricism, and this is why we have tried to expose the radical Davidson by triangulation of Davidson’s use of the triangulation figure with Brandom’s inferentialist understanding of the constitution of our capacity for thought and speech. We have attempted to show that common criticisms of Davidson’s triangulation figure take aim at an understanding of triangulation that, while typically and understandably attributed to Davidson, expresses commitments that we would like to filter out. Critics attack Davidson’s claim that triangulation primarily depicts a causal nexus, linking sensory stimuli from the world with two or more creatures’ mutually responsive behavior. Davidson holds that this figure is an externalist condition for the possibility of creatures to acquire the capacity for thought and speech, hallmarked by the ability to distinguish between true and false, right or wrong, with the re-

249

quirement, though, that a consistent argument can be constructed to the effect that the triangular arrangement enables a distal identification of what in the stream of stimuli are the salient objects and events. We have agreed with critics that the attempt to explain a basic structure of individual acquisition of thought and beliefs by appealing to a nexus of causal regularity will not succeed. In our triangulation of Conservative Davidson with Brandom we have contested, however, the need to maintain this view of the explanatory point and philosophical purpose of the triangulation figure. Our preferred philosopher, R-Davidson, never sought to explain how individuals acquire thought and speech; he uses the triangulation figure to say something about how the world becomes our world through more basic forms of responsive, communicative behaviors between us speakers and between us and the world. The consequence of this view is that we ought to see triangulation as pertaining to concept constitution, rather than as providing insight into how individual creatures acquire the capacity for thought. We contested the presumption by asking whether we can maintain Davidson’s claim that the triangulation figure expresses a condition for concept constitution if we downplay the role assigned to causal regularity, and look instead at the actual descriptions and examples that Davidson provides when illustrating the implementation of the triangulation figure. What we see then are descriptions of how children and sentient creatures actively communicate in responding to the impressions of the world and to the responses of the other triangulating creature(s). Causality clearly plays a role in this pattern of responsive communication, but must it play the central role that Conservative Davidson and his readers claim that is should? We attempted to answer this question through our discussion of the first phase of our triangulation of Davidson and Brandom. Our proposal was to downplay the role of causality by accepting Brandom’s acknowledgement of the role of causality and sensory receptivity as a natural fact on which we simply have to rely and without which concept formation would be unthinkable, but which does not play any kind of constitutive role in the formation of conceptual content. In Brandom’s philosophical universe the formation of concepts is fully determined in inferential scorekeeping exchanges between members of a linguistic community, and then not as the rules or proprieties of those discourses, but in the actual per-

250

formance of the speech acts; in what these members are saying and doing. Brandom thus considers concept formation—the acquisition of thought and speech—as taking place through a particular form of discursive practice, in which we stand committed to—in a Brandomian tone—the “material logic” of those practices, i.e., committed to whatever premises that entitle an interlocutor to say whatever she is saying and to whatever consequences saying it may have, regardless of whether the premises and consequences support or undermine our own beliefs and knowledge. This takes us much further, however, than Davidson ever intended to go with the triangulation figure. We simply wanted to use Brandom’s perspective to see whether it could be possible to look upon triangulation as a practice; as something we actually do in a Brandomian sense of the word. We found support for this view in Davidson’s examples and descriptions of actual triangulation more than in his general descriptions, since his emphasis in these examples are precisely what creatures do when triangulating. Not only did we find that this is possible; we also found that Davidson’s examples enable us to maintain triangulation as a more basic or primitive figure than one of merely reasoned, discursive practice, as in Brandom, allowing us to emphasize that the corporeal things of the world have to be fully integrated in this practice, understood as a practice in which we, as children, acquire the world as a world, or more precisely as our world. Triangulation, then, is a figure outlining the pattern of a process in which a child is being initiated into acquiring the world as its world; beginning with a conditioning practice or game in which objects and events in the world, which at the outset have no particular meaning to it such as tables, birthdays, or washing its hands, are being made known and salient to it through the game, and where the use of sounds in connection with the “object-” or “event-game” is but one of many activities of that game. The game of accustoming a child to the space of reasons involves directly in the learning process also those corporeal, lumpy parts of the world that the sounds we are conditioned to use correctly, come to be words, sentences, and concepts about. Somewhere in this process the conditioning fades and the child comes to grasp the sounds as having a particular role and pride of place in the game, in the form of words and concepts. How or when this transition happens cannot be reconstructed, but we have argued that the final element needed from Brandom to fulfill the triangulation with the con-

251

servative Davidson is the understanding that the hallmark and prime characteristic of the state and attitude of the child, once having gone through the transition, is its grasp of concept use as a fundamentally normative practice. One may reasonably doubt whether we are entitled to the claim that in constructing R-Davidson, we have actually also located a radical Davidson. We conclude by briefly addressing four grounds for such doubt. We have not, first of all, been able to maintain the status of the triangulation figure as a necessary condition for the acquisition of thought and speech, in so far as it is taken to relate two creatures not yet into thought and speech with their common environment. The impossibility of doing so has already been pointed out by many critics, among them Peter Pagin and Jason Bridges, and used by them to reject Davidson’s triangulation enterprise in total. This line of criticism is fair enough, in view of Davidson’s frequent claim that his ambition was to show a condition of the acquisition of the capacity for thought and speech. R-Davidson, however, remains sanguine on this point, since his concern is not to explain how a creature’s capacity for intentionality may emerge out of a state of non-intentional behavior.4 In the triangulation processes Davidson actually does describe, particularly the examples of a child learning the use of words, the role of the mother or teacher is not to transfer her intentionality—her beliefs, her knowledge, her desires—on to the child; it is to be the sanctioning, corrective part of the triangulating learning-process. She may, in principle, use any sounds and any sanctions she wants in playing the games of “table practices”, “washing practices”, “birthday practices”, etc. with the child, as long as she uses them consistently.5 The main or principal difference be-

4

Since Davidson frequently states that his account of triangulation is not a reductive account, but often seems to describe it as if it were, he is at least not very clear on this point. But that creates a room for an understanding of the radical Davidson from whom we may take it literally that triangulation does not invite reductivism. 5 She might for instance use completely wrong sounds for tables, washing, or birthdays, and would then introduce the child to a slightly different world if it, as another Kaspar Hauser, grew up only with this one triangulator, provided, of course, that she managed to be fully consistent in her use of the “wrong” sounds (which would then be the right sounds to the child). But normally the child would be quickly corrected even

252

tween the learning-child examples and the animal examples is not that the triangulator of the first examples is the only one who possesses intentionality, but that the child acquires the world, or actually is being initiated into the world, as a conceptual world; a world that can be understood and expressed in language.6 The child’s beliefs, desires, and knowledge may differ considerably from the mother’s or teacher’s beliefs, desires, and knowledge, but their conceptual world is the same, since they are both taking part in language games that reach far beyond the individual beliefs etc. they acquire. Thus, we suggest that what we must face up to, is the error of the very idea that triangulation between non-intentional creatures constitutes a condition of the possibility of acquisition of intentional states. If we abandon this idea, we can maintain that the radical Davidson uses the triangulation figure to help us understand a basic condition to the acquisition of thought and speech in the sense of concept constitution; of how the world emerges to the child as conceptually constituted and determined. Secondly, and closely related to the first concern, we have seriously modified the significance of Davidson’s use of animal examples or similes. We have already explained why we think the R-Davidson must regard these as nothing but pure illustration, evincing the philosophical insight that the communicative pattern displaying a child’s learning process as a process of acquiring a world (or of being initiated into a world), can also be used to make sense of how animals acquire or are being initiated into a world. The reason is not that this is an accurate description of how the animals experience being initiated to their world, but that it is the only way we, the observers, can make sense of their being initiated to it. And then we can, as philosophers, say that triangulation offers us a grasp of a basic condition to concept constitution in how we, as children, are being initiated into our world through basic practices or games of initiation. The triangulation figure—as a figure of such initiating practices or games—takes on the status of expressing or exposing an a priori, externalist condition for then, when entering into triangulating table-, washing-, and birthday practices with other children and families. 6 With Hegel (and Brandom) we might say that the child is being initiated to the Concept; with Gadamer we might say that it is being initiated into a linguistic cum cultural tradition.

253

the possibility of a capacity for thought when the latter is regarded as the passage to a conceptually accessed world. The third and fourth concerns challenge the claim that R- Davidson may be extracted from the philosophical works of Donald Davidson. The first of these stems from doubt that Davidson may accommodate the final point of the triangulation with Brandom; namely that the triangulation figure must be consonant with the fundamental role of normativity in any practice that deserves the label conceptual (or simply thinking and speaking). Even if triangulation construed as a causal nexus does not make a normative pattern explicit, we believe that Davidson’s examples of triangulation as practice demonstrate that normativity is embedded as a potential of those practices from the outset. What begins as the sanctions of a conditioning practice in fact embeds the possibility of a normative practice, but is only realized as such once the learner becomes capable of taking on the commitment of complying with the sanctions, in the sense of the proprieties of the language games into which she is being initiated. Davidson’s examples and account stop before normativity becomes an issue, but we cannot see any obstacle to the claim that his description of basic triangulating practices or games, such as the game of initiating a child to tables, or hand washing, or birthdays, admits a potential for these to become normative practices of speaking correctly about tables, hand-washing, or birthdays, or in a more Davidsonian tone, of complying with the truthconditions of the use of those words. Finally, one may suspect that we have relied too much on the example of the child learning to speak, leaving aside not only the animal examples (as anything more than illustrations), but even Davidson’s general accounts of triangulation. Can this sole example bear the weight of the burden we have loaded onto the triangulation figure? We think so, but this requires an explanation. Davidson said that the idea of the importance of the triangulation figure dawned upon him when the importance of ostensive learning to a child not yet into thought and speech sank in (see 2001a: 292), so this example has a particular status from the beginning, even if we think that Davidson exaggerates the role of ostension. We might say that the child example is a paradigmatic example that reaches beyond what it specifically says about children learning to speak, and that it, as a description of triangulation as a conditioning cum learning practice, has a general signifi-

254

cance for an understanding of any acquisition of or initiation into new aspects of the world. Through the eyes of R-Davidson we can see that the triangular figure or pattern of how a child is being initiated into the world through a multitude of initiation practices or games is not something Davidson has “discovered”, but a general figure that enables us to get a grasp of how the constitution of the world as conceptual can be seen to emerge out of such basic initiating games. In that sense, and only that sense, it makes sense to see Davidson’s triangulation figure as a form of transcendental condition to our understanding of concept constitution. Thus, we feel entitled to say that the triangulation figure is not only about a child learning to grasp the world as conceptual; it is about any acquisition of a new aspect of the world, or any initiation into a new such aspect. As with the child, all persons participate regularly in triangulating learning games in which our responses to any new experience is being triangulated with the similar responses of others and with whatever in the world we are experiencing, into the learning of new saliences, be it an experience that invites the use of a new theory or metaphor to make something hitherto unknown become something salient (such as the subconscious emerging from the work of Freud and others), or be it the result of discovering something new for which we still have no relevant concepts, such as dark energy in outer space. Even if these games do not start as conditioning games, it may still be helpful to see them as triangulating games in the sense that the persons who are involved in trying to conceptualize these new aspects of the world only may do so gradually, by participating in language games where the new aspects can be experienced again and again, and where the responses of any single person to the experience are being corrected or sanctioned by the others’ responses, until the concept emerges—such as the subconscious from experiences of hysteria or the muon from experiences of observing hooked vapor trails in large hadron colliders. We think R-Davidson survives the four challenges. No doubt, RDavidson breaks with well-founded construals of Davidson’s texts, but we still believe he represents a genuine spirit of those texts. He is concerned with our capacity for thought and speech as a question of understanding the conceptual, and not as a determination of the emergence of thought in individual minds. There is a distinct parallel here with the perspective on

255

the principle of charity as a condition for the possibility of reason and rationality, more than as a methodological device of the interpretation of individual speakers. Triangulating Davidson with Brandom brings to the fore a resemblance or relationship between the two that may be stronger than what is often acknowledged, by resolving the tension in Davidson’s work between the remnants of a reductive naturalism and an emerging pragmatic rationalism in favor of the latter. It thus opens for a reading of Davidson that brings him closer to Continental hermeneutic figures and distances him from the post-empiricist naturalist tradition after Quine. To insist, as we do, that Davidson’s work embodies deep aspirations in real tension may appear to some to be a serious criticism. To our way of thinking, it is an acknowledgement of his greatness as a philosopher for our times. References Brandom, R. 1994, Making It Explicit. Cambridge: Harvard University Press. Brandom, R. 2000, Articulating Reasons. Cambridge: Harvard University Press. Brandom, R. 2009, Reason in Philosophy. Cambridge: Harvard University Press. Bridges, J. 2006, “Davidson’s Transcendental Externalism”. Philosophy and Phenomenological Research, 73 (2), 290-315. Davidson, D. 1975, “Thought and Talk”. In S. Guttenplan (ed.) Mind and Language. Oxford: Clarendon Press, 7-23; repr. in Inquiries into Truth and Interpretation. Oxford: Oxford University Press, 1984 (from which I quote). Davidson, D. 1992, “The Second Person”. In P. French, T. Uehling, et al. (eds.), The Wittgenstein Legacy, Midwest Studies in Philosophy, 17. Notre Dame, Ill.: University of Notre Dame Press, 255-267. Davidson, D. 1994, “The Social Aspect of Language”. In B. Mcguinness (ed.), The Philosophy of Michael Dummett. Dordrecht: Kluwer, 1-16; repr. in Truth, Language, and History. Oxford: Oxford University Press, 2005 (from which I quote).

256

Davidson, D. 1997, “The Emergence of Thought”. Erkenntnis, 51, 7-17; repr. in Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001 (from which I quote). Davidson, D. 1999, “Reply to Føllesdal”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. La Salle, Ill.: Open Court, 729-732. Davidson, D. 2001a, “Comments on Karlovy Vary Papers”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 285-308. Davidson, D. 2001b, “Externalisms”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 1-16. Dretske, F.I. 1981, Knowledge and the Flow of Information. Cambridge: MIT Press. Føllesdal, D. 1999, “Triangulation”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. La Salle, Ill.: Open Court, 719-728. Pagin, P. 2001, “Semantic Triangulation”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 199-212. Ramberg, B.T. 2001, “What Davidson Said to the Sceptic”. In P. Kotatko, P. Pagin, et al. (eds.), Interpreting Davidson. Stanford: CSLI, 213-236. Stroud, B. 1999, “Radical Interpretation and Philosophical Scepticism”. In L.E. Hahn (ed.), The Philosophy of Donald Davidson. La Salle, Ill.: Open Court, 139-161.

TRIANGULATION AND PHILOSOPHY: A DAVIDSONIAN LANDSCAPE Jeff MALPAS (University of Tasmania, Hobart, Australia)

Triangulation … [… n, of action from *triangulare to TRIANGULATE. So F. triangulation (1835 in Dict. Acad.)] The action or process of triangulating. 1. The tracing and measurement of a series or network of triangles in order to survey and map out a territory or region. (Oxford Dictionary of the English Language).

1. Introduction: Triangulation and a Philosophical Landscape Although the term may not appear prior to “Rational Animals” (1982), a form of triangulation is clearly present in Davidson’s work from very early on. It can already be discerned in Davidson’s development of the idea of radical interpretation: the assignment of content to attitudes and utterances is dependent on the holistic interconnections of attitudes, of attitudes with utterances, and of both with action; this interconnection must encompass speakers taken together as well as singly; and it must also depend on the interconnection between attitudes, utterances, actions and the real-world objects to which they relate (both causally and intentionally). Moreover, triangulation can be viewed as adumbrated in the Quinean description of the three-way encounter between field linguist, native speaker, and environmental stimulus that is at the heart of radical translation (all the more so given Davidson’s conversational quip that it was Quine who, at the time of the writing of Word and Object, convinced him of the truth of externalism1). Within contemporary analytic thought, the reception of the Davidsonian account of triangulation has been, in general, fairly unsympathetic. By this, I do not only mean that Davidson’s account of triangulation has been viewed as implausible or unconvincing, but also that it has usually 1

A quip made, in my hearing at least, at a conference in Aix-en-Provence in 1999.

258

been read in ways that make very little effort to come to a genuine understanding of that account—such readings typically fail to heed the Collingwoodian advice (which amounts to nothing more than a basic principle of hermeneutic adequacy) that when one reads an author the first aim should not be to arrive at a judgment of the absolute truth or falsity of the author’s claims, but rather to come to the position at which one could see how those claims could be true.2 The fact that the analytic consensus on Davidsonian triangulation has been negative may itself be thought somewhat surprising given the way in which triangulation itself connects with a much wider body of thought within both contemporary philosophy and its history (although that may say more about the readers of Davidson’s work than about the work itself). One might well argue that what Davidsonian triangulation does is to present, within a specifically analytic idiom, ideas that have a much more widespread currency within pragmatist, idealist, phenomenological, and hermeneutic ways of thinking. What I aim to do here is to explore some of these broader connections, conceptual as well as historical, and to locate triangulation within a broader landscape than is usually the case. The aim is thus to “triangulate” triangulation, and to provide a brief sketch, through this triangulative process, of the territory or region within which Davidson can be located—a territory or region that, in my own work, is encompassed within the idea of “philosophical topography” or “topology” (see, for instance, Malpas 1999a). Undertaking such a task is valuable not only because of the light it may shed on the Davidsonian position as such, but also because of the way in which it opens up a different mode of philosophical

2

See Collingwood (1978: 27). (Collingwood is here referring to a principle underpinning his own teaching practice, but see also his discussion in Collingwood 1978: ch. V, 29-43, and ch. VII, 53-76.) It should be noted that, in spite of the apparent difference in the role given to truth here, Collingwood’s advice is quite consistent with the Davidsonian principle of charity, and can be viewed as a variation on it. Like Davidson, Collingwood emphasises the need to assume the rationality of one’s interlocutor’s utterances, and so to arrive at an interpretation that makes the best overall sense of those utterances (on this account, an inability to make sense of one interlocutor’s utterances, which includes an inability to see how those utterances could be held to be true, must itself count against the viability of the interpretation being employed).

259

proceeding than is common within the analytic tradition that dominates so much of contemporary Anglo-American thought. 2. Triangulation, Topography, and Some Geometry In its original form, the practice of triangulation is essentially underpinned by the principles of Euclidean geometry together with some basic trigonometry: from one angle plus the lengths of two sides of a three-sided figure one can always determine the length of the third side. When applied in topographic surveying and mapping (an application pioneered by the Dutch mathematician Willebrord Snell in the 17th century), triangulation enables the determination of relative location—given two locations that each stand at a known distance to a third location, and given also the angle that subtends the two lines connecting each location to the third, one can then determine the distance between the first and second locations—and through repeated triangulations one can map out a series of locations that make up an entire region or territory. The need for triangulation to operate through repetition (triangulation is not something that can be completed all at once) indicates that topographic triangulation has an essentially temporal dimension—in this respect, triangulation may even be said to provide a demonstration of the necessary co-implication of space with time—although it is perhaps better to say that it is essentially dynamic or processual. The locations upon which triangulation operates in its topographic employment may be identical with the locations of particular entities or geographical features (natural formations or human constructions), and yet the locations themselves will be determined as locations only relative to the other locations with which they are connected within the triangulative structure—the determination of location in triangulation is thus essentially relational, and this is a feature of such locations and not only of our knowledge of them. Finally, since it is indeed Euclidean geometry that underpins the process of triangulation, so triangulation can only operate across a single plane—across a single surface. The relations that are determinative of locations and that are worked out dynamically or processually are also, therefore, relations that themselves belong to a single surface—there is nothing to the relationality in question that exists either beneath that surface or, indeed, above it. Process, rela-

260

tionality, and surface are three ideas that lie at the heart of triangulation, not only as it operates in geographical mapping, or topographical surveying, but also in its philosophical employment, of which Davidson provides an especially significant, but not the only, exemplar. 3. Triangulation, Knowledge, and Content In its philosophical employment, triangulation is not intended to deliver, nor does it depend upon, calculations of angle or distance as these apply in a strictly geometric context. In his original presentation of the idea in “Rational Animals”, Davidson describes triangulation as he deploys it as follows: If I were bolted to the earth, I would have no way of determining the distance from me of many objects. I would only know that they were on some line drawn from me towards them. I might interact successfully with objects, but I could have no way of giving content to the question where they were. Not being bolted down, I am free to triangulate. Our sense of objectivity is the consequence of another sort of triangulation, one that requires two creatures. Each interacts with an object, but what gives each the concept of the way things are objectively is the base line formed between the creatures by language. The fact that they share a concept of truth alone makes sense of the claim that they have beliefs, that they are able to assign objects a place in the public world (1982: 105).

If triangulation establishes the distance between two points that each stand in relation to a third point, then, in Davidson’s exposition here, the relation between the first two points and the third is equivalent to the relation between each of two speakers and a third object, and the relation between the two speakers—the “base-line” that is objectivity—is what is worked out within the triangle. As Davidson employs it, however, triangulation does not serve only to establish the relation between speakers. Indeed, in the passage just quoted, Davidson first introduces the notion of triangulation as a means by which the locations of objects can be determined relative to an observer. While he does not always clearly distinguish between the different aspects of triangulation at work here (and sometimes writes as if there were quite different forms of triangulation at issue), Davidson actually

261

seems to use triangulation to encompass a single complex structure within which content is determined, and within which subjectivity, intersubjectivity, and objectivity are mutually determined and defined. In “Three Varieties of Knowledge”, the essay that he himself presents as providing an overall summary of his position, Davidson articulates a triangulative structure that unites the three forms of knowledge at issue here—knowledge of self, knowledge of others, knowledge of world—and in uniting them also makes them possible («the three sorts of knowledge form a tripod: if any leg were lost, no part would stand», Davidson 1991: 220). Although the triangle at issue here exhibits, one might say, certain structural relations between concepts, it can nevertheless also be seen as a sort of generalised, and perhaps more statically rendered version, of the same structure that is evident, in dynamic form, in the triangulation that unites a speaker with another speaker and a common object. This is particularly clear in “Three Varieties of Knowledge”. The central theme of the essay is indeed the relation between subjective, intersubjective, and objective, which is presented as forming a triangular structure, but at the same time Davidson describes triangulation, in particular, in a way that echoes his initial explication in “Rational Animals”. Triangulation is thus understood in terms of the determination of location, but location is itself the analogue (the very close analogue3) for content: It takes two points of view to give a location to the cause of a thought, and thus to define its content. We may think of it as a form of triangulation: each of two people is reacting differentially to sensory stimuli streaming in from a common direction. Projecting the incoming lines outward, the common cause is at their intersection. If two people now note each other’s reactions (in the case of language, verbal reactions), each can correlate these observed reactions with his or her stimuli from the world. A common cause has been determined. The triangle which gives content to thought and 3

I say “very close” because the determination of content is so much tied to the determination of cause; causes are always causes “under a description”; and the identification of cause that enables the determination of content involves the identification of the appropriate description along a continuum of descriptions that can be understood as extending from the speaker to the object. Content and cause are thus tied together, but so too are they tied to a form of notional “location” along a line from speaker to object.

262

speech is complete. But it takes two to triangulate (Davidson 1991: 212213).

The Davidsonian account of triangulation—mirrored in the triangular relation between the three forms of knowledge—is structurally analogous to triangulation as it operates in geometry: in particular, it employs knowledge of the relation between two objects as they each stand in relation to a third in order to determine the relation between the first two objects.4 Perhaps more significant, however, is the fact that Davidsonian triangulation also exhibits features that are characteristic of triangulation as it operates in topographical surveying or mapping: it is essentially oriented to process, relationality, and surface. The processual or dynamic character of Davidsonian triangulation is a feature of triangulation in two ways. First, in a way that directly parallels the structure of topographic triangulation, Davidsonian triangulation is not a process completed all at once, but is instead a process of ongoing interaction between speakers, and between speakers and the world. Second, the interactive character of triangulation in the determination of content is itself underpinned by the interactive character of the engagement between speakers, and between speakers and the world that resides in their everyday activity—it is thus through their behavior that speakers are drawn into relations with others and with the world, and it is on the basis of this behavioral interaction that the triangulative relation that makes for the possibility of content arises. The role of action here is crucial: the “common causes” that Davidson looks to as standing at the intersection of lines taken from each speaker to the world, and determined as common, not because of some mysterious lines of intentionality or perception, but through the causal interaction between speakers, and between speakers and objects, and this causal interaction is one that is perceptual and also behavioral. It is because speakers act in similar ways toward the same objects that those objects present themselves to the speakers in similar ways—typically in ways that interfere or

4

In the geometrical case, one might think of the “objects” at issue as the apexes of the triangle—related by the distances between them—or as the lengths of the sides.

263

overlap with one another through the interfering or overlapping modes of behavior of speakers (see Malpas 2011e). The interaction that underpins triangulation, and in which triangulation consists, is not merely an interaction between entities that are already determined in respect of the content that might be attributed to them. Triangulation is indeed the process by which content is constituted, and so, inasmuch as speakers are understood to have certain attitudes or to engage in certain actions (intentionally described) and inasmuch as objects are presented as having a certain meaningful character, then such attitudes, actions and meanings are formed only within the relational structure that is established by triangulation. Put in a slightly different way, one might say that it is only through the relational interaction between speakers, and between speakers and the world, that speakers come to be speakers (and so to have attitudes), that behavior becomes action, that objects become meaningful. The way in which Davidsonian triangulation exhibits a processual or dynamic character, is thus itself directly connected to its commitment to a fundamental relationality. The Davidsonian claim regarding the role of triangulation is, in this respect, not merely an epistemological, but an ontological thesis—it concerns the grounds for the possibility of content, or, to put the matter more directly, it concerns the very being of content. The failure to appreciate the ontological nature of triangulation is one key reason why triangulation has been so poorly received by most of the analytic readers of Davidson’s work: triangulation has been assumed to be a process that operates upon content (no matter whether content is understood in terms of the meaning of sentences, the contents of belief, or the intentionality of action) rather than being that which establishes content. Davidson’s position exemplifies what I take to be a form of “relational ontology” with respect to content: the ground for the possibility of content is given through the interaction between creatures possessed of certain cognitive and behavioral capacities (more specifically, creatures possessed of the capacity for language, and hence creatures who can take on the roles of speakers and agents), and between those creatures and the world. In referring to this as a relational ontology, I do not mean, however, that the existence of anything other than relations is refused or denied. The relational ontology at issue is one that applies to content, but it depends upon an un-

264

derlying causal ontology—an ontology, if we are to remain within the Davidsonian framework, of events (and perhaps also of things, inasmuch as any ontology of events can perhaps be redescribed, with some adjustments, in terms of an ontology of things). The relationality of content that emerges within the account of triangulation reflects the character of triangulation as it operates with respect to locations or landmarks within some territory according to which locations are determined as locations through the way in which they stand in relation to other locations, and so as part of a locational network rather than as independent elements. Locations thus have to be understood holistically in much the same way that Davidson claims beliefs, and meanings, also have to be understood holistically: to have one belief, there must be many beliefs; for there to be one meaningful sentence, there must be many meaningful sentences; for there to be one speaker, there must be other speakers and a world.5 The structure of triangulation makes very clear the way in which the holism that appears here is itself tied to externalism: holism does not concern merely the inter-relatedness of content as it is given within some “internalised” structure of content (with respect to a particular body of beliefs or utterances narrowly conceived), but encompasses the “externality” of other speakers and the world. On the basis of such holistic-externalist relationalism, not only can we not understand content as residing in autonomous, “internalised” states or structures, but we cannot understand the mental in such a fashion either. Minds emerge not on the basis of autonomous states, brains, or bodies as they stand apart from the world, but instead supervene on larger systems that encompass such states, brains, bodies and environments in much larger holistic interaction. It is this that is the real basis for the idea of anomalous monism—which can now be seen as essentially an expression of the commitment to triangulation, conjoined with certain other claims concerning the nature of law and cause, within the philosophy of mind.

5

How many speakers? Triangulation, taken literally, suggests that there must be at least one other speaker, but since speakers always belong to a community of speakers, so it seems safe to say that for there to be one speaker there must be many.

265

If the relationality of content can be seen to be underpinned by the processual or dynamic character of triangulation, then relationality can itself be seen to underpin the character of triangulation as giving priority to surface. In the Davidsonian context, the idea of surface appears most clearly in terms of a rejection of certain forms of foundationalism. The generally anti-foundationalist tendency of Davidson’s thought is clearly evident in Richard Rorty’s deployment and development of Davidsonian ideas (with which Davidson himself was largely in sympathy6), but we can see how such anti-foundationalism is a key element in Davidson’s own thinking by consideration of the structure of triangulation alone. Triangulation requires that we look to the reciprocal relations between the elements that together make up the structure of which they are part rather than to any single element as determinative of that structure. Triangulation thus makes for a unity that is based in the very same plurality of elements, and their relatedness, that triangulation also brings to light, and does not allow of any reduction or dissolution into anything simpler or more basic. Although it is only within the structure of triangulation that content, and so also action, mind, and meaning, come to appearance, triangulation itself depends on the interconnection of content as worked out in the interconnection between speakers, between speakers and the world, between attitudes, and between attitudes and behavior. Triangulation operates, in other words, through that which it also constitutes—hence the emphasis on what I have presented as the primacy of surface. The anti-foundationalism that is at issue here appears in many different ways in Davidson’s work. It is what underlies his onetime characterisation of his position as involving a “coherence” theory of truth and knowledge, and, more generally, underpins both his rejection of scepticism and relativism as well as his refusal to allow that knowledge can be grounded in any set of infallible procedures or indubitable “truths”. It is evident in his view of linguistic understanding as based, not in any system of rules or conventions, but in the ongoing communicative interaction between speakers (as 6

Although Rorty gave a much more pragmatist twist to Davidson’s position than Davidson himself would have accepted—something that is clearest when it comes to the question of truth. See Rorty’s response to my reading of Davidson on this point in Rorty (2000: 41-42, n. 22).

266

Marcia Cavell puts it «Meaning depends on successful communication rather than the other way around», 2005: xv). It leads to a rejection of the idea that there can be any single, privileged vocabulary that is uniquely adequate to the world so that truth is always plural (if there is one truth then there are many truths)—an idea itself crucial to the possibility of interpretation across difference (see Malpas 2011e)—and that also leads to Davidson’s adoption of an ontologically monistic, but descriptively pluralist approach to the relation between the mental and the physical (the position known as “anomalous monism” to which I referred briefly above). Although the material on which Davidsonian triangulation operates is different from that which is the subject matter for the geometer or topographical survey, still the same structure obtains, and process, relationality, and surface appear as key features of triangulation as it functions philosophically no less than in its geographical employment. What triangulation gives rise to in the philosophical sphere, however, is a position that may well appear quite divergent from more conventional accounts inasmuch as it does not allow what philosophy often seems to demand, namely, some privileged level of analysis on the basis of which all else can be extrapolated, explained or deduced. There is no such level here—there is only the complex interweaving that combines elements, and is formed in such combination, at the same time as it also determines them (a determination that is never complete, and so always allows for other determinations7). Moreover, not only does triangulation mitigate against reductive or foundationalist approaches, but because it does indeed position notions such as belief, action, meaning and mind within a larger interdependent framework of concepts, so it also requires that one rethink the key concepts at issue here. Belief, truth, action, world cannot mean the same within the structure established by Davidsonian triangulation, and the holisticexternalist relationalism with which it is associated, as these terms may mean elsewhere, and to read Davidson on the assumption that those terms do mean the same is to misread him. The rethinking required here (to which Davidson typically does not himself draw attention—the “subversion” in which he engages remains implicit) is usually not restricted just to 7

It is thus compatible with indeterminacy understood as the ever-present possibility of other determinations.

267

the concepts themselves, but ramifies across a range of philosophical categories and modes of approach. The idea of triangulation does not entail just a different conception of content, but a different conception of what philosophy itself might be about and how it might go about it. It is this different conception that seems to me to be captured within the idea of philosophical “topography” or “topology”. Such a topography or topology can be seen as an elaboration of Davidsonian triangulation, but it also draws on topological and cartographic conceptions of philosophy that can be found in thinkers such as Kant and also Heidegger (see Malpas and Thiel 2011; Malpas and Zöller 2011; and also Malpas 2006). 4. Triangulation and Some (Other) Philosophers With some exceptions, much of the discussion of Davidsonian triangulation within the analytic literature tends to ignore or to overlook the features of process, relationality and surface that are so central to the Davidsonian position—as a result there is a tendency to read Davidson as holding to a much more conventional position than is actually the case, and to import into such readings exactly the assumptions that the idea of triangulation contests. Moreover, the analytic treatment also tends to approach triangulation from within a fairly narrow philosophical frame—as if the idea were peculiar to Davidson alone. The tendency to treat it as idiosyncratic in this way undoubtedly contributes both to the difficulty many analytic readers have in understanding the Davidsonian position, and their generally negative judgment as to its philosophical cogency and significance. That latter tendency surely reflects the somewhat parochial character of much analytic thought—not only is there a residual suspicion of many forms of engagement with the history of philosophy, coupled with a tendency to prioritise philosophical thinking as it occurs in English, but there is also a more general and widespread inability to undertake any genuine philosophical engagement with ideas that go beyond certain conventional or familiar tropes and approaches.8 Thus analytic philosophy, for all that it 8

A tendency that Collingwood noted, again in his Autobiography (1978), as also a feature of the “realism” that characterised English philosophy in the early part of the twentieth century.

268

often presents itself in liberal and cosmopolitan terms, too often tends, at least so far as its own philosophical practice is concerned, to be relatively narrow-minded and intolerant—it insists on assimilating everything to its own familiar frames, shunning what resists such assimilation.9 Seen from a perspective that is broader than the analytic, however, Davidsonian triangulation, and the externalist-holist relationalism that it expresses, can be seen to be continuous with a larger tendency within European thought (one that carried into earlier traditions of British and American philosophy) that emphasises much the same “triangulative” structure—although often expressed in different ways—as constitutive of “mind” (broadly understood), content, or self as these stand in relation to world. In terms of contemporary thought, the work of Hans-Georg Gadamer provides an important and obvious point of comparison here.10 Like Davidson, Gadamer adopts a dynamic and dialogic approach to the character of understanding in general—something captured in Gadamer’s emphasis on 9

In its defence it might be said that it is only intolerant of what it takes to be philosophically dangerous or mistaken. One has to be careful here, however, since this is exactly the sort of defence that is always used to defend intolerance and prejudice. Judgment, whether negative or positive, has to be based on some form of understanding, and yet understanding is just what analytic philosophy, in its encounter with other modes of philosophising, often seems to resist and to refuse. 10 For a closer comparison of Davidson and Gadamer, see Malpas (2002), as well as Malpas (2011e); see also Hoy (1997), and Dostal (2011). It is worth noting that Dostal takes the Davidsonian position to be essentially epistemological rather than ontological (and explicitly opposing my own reading of the matter as set out in Malpas 2002). Such a claim seems to me to be indefensible on at least two grounds: first, on the grounds that Davidson clearly and explicitly commits himself to a set of ontological claims—claims about, for instance, the nature of mind, meaning, belief and knowledge; and second, on the grounds that if we take seriously the transcendental character of Davidson’s project, as Dostal seems himself to do, then we surely must treat Davidson’s account as ontological, and not merely epistemological, since the transcendental concern with grounds is not a concern merely with evidence, but with that which makes possible—the transcendental is indeed a mode of ontological inquiry, although one that also transforms the conception of what ontology might be. This is a particularly important point given my own topographical conception of the transcendental, of the ontological, and of the philosophical (on this matter, see especially the essays in the first part of Malpas 2011c as well as the opening and closing discussions in Malpas 1999a).

269

language as “conversation” (Gespräch)—and similarly eschews any idea that understanding can be based in some determinate “method” or rule (on Gadamer, see Malpas 2005, 2010). One of the key moves in Gadamer’s development of hermeneutic theory was to emphasis the nature of hermeneutic engagement as always oriented towards some thing—a particular subject-matter or Sache—whether understood in terms of a question that is at issue or a “text” that is to be understood. Every act of interpretive engagement involves a dialogue between conversational partners—even though sometimes one partner may be present only through the text—that is directed towards what Gadamer calls a “fusion of horizons” (Horizontverschmelzung). Reading Gadamer through the same triangulative lens employed by Davidson, one can say that the structure of hermeneutic engagement is one in which the fusion of horizons that is understanding, and which always takes place in language (leading Gadamer to talk of understanding, in a way that misleads Davidson in his own reading of Gadamer (see Davidson 1997: 275), as a matter of arriving at a common language), corresponds to the establishing of the base-line joining one speaker to another that Davidson sees as a key part of the triangulative process. That base-line is established through the way each speaker articulates their own relation (a relation that must be both causal and intentional11) to the interpretative object—the Sache in Gadamer’s terms, the “common cause” in Davidson’s.12 The comparison with Gadamer is itself suggestive of a further comparison that can be drawn between Davidsonian triangulation and aspects of the thinking of Gadamer’s one-time teacher, Martin Heidegger. Although Heidegger does not focus specifically on the structure of interpretation or understanding in the way that Davidson or Gadamer do, and also uses a different language from either of these later thinkers, he nevertheless 11

The fact that Gadamer does not thematise the causal relatedness here does not mean that it is irrelevant (although one might view it as trivial or truistic); Davidson, on the other hand, is explicitly concerned with the way in which intentional or rational engagement is necessarily underpinned by the causal. The latter, however, need not be incompatible with the former. 12 Notice that since Davidson takes reasons to be causes, the difference between his and Gadamer’s language here is less than one might first think. See Malpas (2008), in which some of the complexities of Davidson’s position are further explored.

270

presents an account in which one can discern a very similar picture of “content” or “meaning” as formed only through the interaction of thinking beings with one another and with respect to the entities that they encounter in their active engagement within the world (see Malpas 2011d). There is thus, one might say, already a triangulative structure that can be seen in, for instance, the analysis given in Being and Time that emphasises the character of Dasein as “being in the world” in a way that is determined both in relation to the being of equipment and in relation to the being of others. Equally significant, moreover, is the fact that Heidegger also strives, if not always successfully, to move toward an account that exhibits similar features of process, relationality, and surface to those that are evident in Davidsonian triangulation. Indeed, it is partly for this reason, that Heidegger’s thinking can be properly understood as constituting a form of “topology”—as, in his words, a “topology of being”.13 The work of Gadamer and Heidegger provides useful landmarks in relation to which the Davidsonian account of triangulation can be positioned. Yet as landmarks they are also positioned in a slightly different part of the landscape from that with which Davidson was most familiar. If one were to look for the philosophical landmarks closest to Davidson with specific regard to the question of triangulation—the landmarks by which he oriented and located himself on this matter—then the thinker whose work is most prominent (in addition, perhaps, to Quine and Wittgenstein) is undoubtedly G.H. Mead. It is Mead’s social account of the mind that provides the immediate antecedent to the Davidsonian position—one that Davidson himself acknowledged (in conversation and discussion, if not in print). Although Mead does not himself use the term “triangulation”, he nevertheless sets out an account that exhibits the same triangulative structure evident in Davidson. In Mind, Self and Society, Mead writes of the relation between selves and the world in a way that prefigures Davidson’s account in essays such as “Three Varieties of Knowledge”: When a self does appear it always involves an experience of another; there could not be an experience of a self simply by itself. The plant or the lower 13

On the topological reading of Heidegger alluded to here, see Malpas (2006, 2011c).

271

animal reacts to its environment, but there is no experience of a self. When a self does appear in experience it appears over against the other, and we have been delineating the condition under which this other does appear in the experience of the human animal, namely in the presence of that sort of stimulation in the co-operative activity which arouses in the individual himself the same response it arouses in the other. When the response of the other becomes an essential part in the experience or conduct of the individual; when taking the attitude of the other becomes an essential part in his behavior—then the individual appears in his own experience as a self; and until this happens he does not appear as a self (Mead 1934: 195).

Self appears, according to Mead, only in relation to the other, while the relation between the two depends on a mutuality of involvement in relation to some common “object”. What Davidson saw to be missing from Mead’s account, however, was argument, and it is partly the argumentative development of the account that Davidson provides. But Davidson does more than just take up an existing set of views: while Mead may be an important precursor to Davidson, what Davidson does is to embed certain key ideas that also appear in Mead within an explicitly philosophical, rather than sociological, analysis, and so in a way that addresses key philosophical problems concerning mind and meaning, self and other, truth and world. Although writing from within the American pragmatist tradition, and so connecting with the ideas of James and Dewey, Mead’s account also draws on a background of ideas that derive both from a developing naturalistic tendency indebted to Darwinian thought (whose effects are also clearly evident in much contemporary analytic thought including that of Davidson14) as well as from elements of the German idealist tradition as mediated through Mead’s teacher, Josiah Royce. Indeed, it is this latter tradition, rooted in the work of Kant and Hegel, that provides a common his14

It is not uncommon to find the commitment to naturalism construed in ways that make it automatically incompatible with, and opposed to, hermeneutic or phenomenological approaches. Such a construal is surely mistaken if only because the meaning of naturalism is itself contested and cannot straightforwardly be taken for granted. In the case of a thinker like Davidson, naturalism does not and cannot be taken as equivalent to any form of reductive physicalism or materialism nor as entailing the rejection of the language of rationality or intentionality (as Davidson’s commitment to anomalous monism should make clear).

272

torical underpinning to the thinking of Heidegger and Gadamer (although it is a tradition with which they take issue), and, even if indirectly, that of Davidson.15 It is in this tradition that one finds, for instance, a similar emphasis on the self as worked out always in relation to others, as well as to the world, and so also a commitment to a mode of analysis that looks not to reduce or to simplify, but to understand elements as mutually related within a single, unitary, yet complex, structure. The type of analysis at issue here appears as a key idea in the transcendental-critical project that is undertaken by Kant in the first Critique. Not only does Kant himself employ an explicitly topographical characterisation of his method and approach (see Malpas and Thiel 2011, and also Malpas and Zöller 2011), but he also looks to understand the structure of knowledge and experience in a way that emphasises the holistic and relational character of the structure by which they are made possible (see Malpas 1999b). In this respect, it seems to me entirely appropriate that Davidson sometimes characterised his position as “transcendental”, since the Kantian idea of the transcendental is itself inseparable from Kant’s own geographic-cartographic conception of his project. The transcendental inquiry is best understand as one that aims to map out the territory that is knowledge or experience in a way that exhibits the unity and limits of that territory, and that does so on the basis only of what is given within it.16 The structure of the inquiry is thus analogous to the structure of triangulation,

15

One may speculate that, in Davidson’s case, this is a tradition filtered through his early involvement with Whitehead—not that it is Whitehead’s own reading of this tradition that is operative here, but rather that through the involvement with Whitehead, Davidson assimilated a body of ideas that, even if not explicitly acknowledged, remained effective in his thinking. Thus Davidson once commented, on reading one of my own discussions of his work in relation to Kant, that he had not realised just how much Kant there was in his thinking, but that it probably was not surprising given how much he had read Kant as an undergraduate. 16 This is a conception of the transcendental that I have explored in a number of essays—in addition to the essays cited in the previous note, see also Malpas (1997, 2011b). The understanding of the transcendental in direct relation to the topographical or topological is an important element in my reading of Kant, as well as in Heidegger and Davidson, as well as being central to the idea of “philosophical topography”.

273

and it exhibits much the same key features—process, relationality, surface—as are evident in the Davidsonian account. It may be thought that what is often missing from Kant is an emphasis on the intersubjective and the social (although this does enter into Kant’s inquiries in the Third Critique), but such an emphasis is certainly present in Hegel in an explicit, if variously understood, fashion within his account of the logic of recognition: «Self-consciousness exists in and for itself when, and by the fact that, it so exists for another; that is, it exists only in being acknowledged» (Hegel 1807: § 178). The self comes to be as a self, according to Hegel, through the way in which it relates to other selves—a relatedness that takes one very particular form in the dialectical relation between master and servant as famously analysed by Hegel in the Phenomenology of Spirit (1807). Moreover, the self also finds its objectification in and through externality of language: «We only know about our thoughts, only have determinate, actual thoughts, when we give them the shape of objectivity, of being distinguished from our interiority, and so give them the form of externality, and indeed of such an externality as carries the stamp of the highest internality» (Hegel 1817: § 462, Addition). While the interpretation of Hegel on these matters is contested (particularly the theory of recognition as developed in the work of Alexandre Kojève), there can be no doubt that Hegel’s conception of the inter-related character of self, other and world, has not only exerted a powerful effect on the history of modern philosophical thought,17 but that it is also an account which, in certain key aspects, can be seen to be mirrored in Davidsonian triangulation. There is not the space here fully to explore the way in which Davidson’s thought may be mapped onto aspects of Kantian and Hegelian philosophy (although it should perhaps be noted that the way both McDowell and Brandom have attempted to position themselves in relation to Kantian and Hegelian thinking may serve to underline the significance of the con17

Although the Hegelian understanding of self as constituted within a “recognitive” structure has had an impact in many different areas and among a range of thinkers, it has been particularly influential in recent philosophy among theorists of the Frankfurt School including Jürgen Habermas, Karl-Otto Apel, and especially Axel Honneth (see, for instance, Honneth 1996—Honneth’s work also draws directly on Mead).

274

nection at issue here), but the key point for my discussion concerns not the details of such a possible mapping, so much as the direction that it suggests for better understanding the character of Davidsonian triangulation. Indeed, it is worth noting that on the one occasion when Davidson does explicitly position triangulation in relation to another thinker, it is not a representative of the analytic tradition whom he cites,18 but rather someone whose work was deeply immersed in a more historical, and implicitly hermeneutic approach, namely, R. G. Collingwood. Thus, in “Three Varieties of Knowledge”, Davidson quotes from Collingwood’s The Principles of Art: The child’s discovery of itself as a person is also its discovery of itself as a member of a world of persons […] The discovery of myself as a person is the discovery that I can speak; and am thus a persona or speaker; in speaking I am both speaker and hearer; and since the discovery of myself as a person is also the discovery of other persons around me, it is the discovery of speakers and hearers other than myself (Collingwood 1938: 248; quoted 19 by Davidson 1991: 219).

Here Collingwood’s thinking seems clearly to echo that of Hegel, but it also locates the idea of the interdependence of self and other in a context that explicitly connects this with the capacity for speech—a point that makes it particularly resonant with the Davidsonian account. The positioning of Davidsonian triangulation within this larger philosophical landscape is not intended as any sort of confirmation of the Davidsonian account—as if philosophical argument were a matter of establishing some form of philosophical consensus. The point is rather to show that, far from being an idiosyncratic approach, the Davidsonian development and deployment of the idea of triangulation stands within, and may well be taken to derive from, a well-established body of previous thought, and that locating Davidson’s thinking in this way may actually be important in understanding the nature of, and so also the possible grounds for, 18

Although Davidson does discuss triangulation in relation to analytic contemporaries such as Kripke and Putnam, it is not because he finds an analogue to triangulation in their work, but rather because of the overlapping nature of the problems at issue. 19 It seems likely that Davidson’s reference to Collingwood here is at least partly a reflection of his close friendship with Alan Donagan—a Collingwood scholar and the author of Donagan (1962).

275

that thinking. Of course, the larger tradition of thought with respect to which triangulation is here positioned is a body of thought toward which analytic philosophy has often taken a hostile and uncomprehending stance. Perhaps it is the latter that really underlies the antagonism so often shown toward the Davidsonian position, or at least, toward the thinking of the later Davidson as that centres around the idea of triangulation. Even if unintentionally, Davidson’s later thinking, particularly as expressed in triangulation, may appear to draw the analytic enterprise too much toward the supposed shoals of a European shore.20 5. Conclusion: Triangulation and Philosophical Topography The idea of triangulation comes originally, as I noted above, from geometry and geography. In geographical terms, it refers us, as the OED has it, to the “tracing and measurement of a series or network of triangles in order to survey and map out a territory or region”. When the territory or region at issue is, instead of a piece of land, a domain of concepts, or a certain region of existence, then triangulation implies a way of understanding that philosophical domain or region in terms that emphasise: the dynamic or processual character of such understanding, and therefore the inability of any “complete” account; the character of that understanding as given in and through the relational articulation of the domain or region, and so the mutuality and interdependence of the elements that make it up; and the focus of understanding on a grasp of the domain or region as given in and through its surface, and so the impossibility of any reductive or simplificatory account. 20

Since it is Davidson’s later thinking that seems to move more clearly in this direction, it is no surprise that analytic readers have tended to want to separate off the later thinking from the earlier—and to retrieve a version of Davidson’s thinking that is more narrowly focussed on the meaning-theoretic concerns evident in Davidson’s early essays. In fact, I see no reason for strongly separating Davidson’s later work from his earlier. There can be no doubt that there are shifts in his thinking, and important changes of mind, but just as the idea of triangulation is itself adumbrated even in Davidson’s early essays, so Davidson’s thought as a whole is marked more by its continuity than its discontinuity—moreover, it is a continuity that is best seen, so I would argue, from the late thought back to the early.

276

In these respects, triangulation captures the essence of what I have called philosophical topography or topology—a mode of philosophical analysis that looks to understand its subject matter in a way analogous to the way a surveyor aims to understand a territory or region: through engaging with and moving across the landscape at issue, mapping out the relations between landmarks, and so uncovering the structure of the region by an exploration of its surface. Such a process never results in a single map that stands as definitive of the region to be explored, but opens up the path for further exploration, for different mappings. Davidsonian triangulation provides both an exemplification of such a form of philosophical topography, while it also lays bear some of its essential features. Understanding Davidson’s thought in this way is to see Davidson as belonging rather less to the mainstream of conventional analytic thought, and, instead, like Richard Rorty, as more part of a very different, much more eclectic, and also more radical tradition. I would suggest that it is partly the attempt to interpret Davidson in a way that does indeed try to assimilate him back into the conventional analytic mainstream that so often results in readings that render key Davidsonian ideas, including that of triangulation, as implausible or obscure. Such an outcome, as I noted at the outset of my discussion here, is itself suggestive of the hermeneutic inadequacy of such readings, as well as of a philosophical narrowness that Davidson himself did not share—a narrowness which, it should be added, he did not himself see as intrinsic to analytic thought. In his foreword to a volume of essays engaging analytic with Chinese thought, Davidson commented on the way in which «the analytic method in philosophy […] when practiced with an open mind […] engenders dialogue» and thereby creates «mutual understanding, fresh insights, sympathy with past thinkers, and, occasionally, genuinely new ideas» (Davidson 2001: v).21 One only wishes that Davidson’s own thinking were more often approached in just such a spirit.

21

A fuller version of the comment appears as the epigram to Malpas (2011a).

277

References Cavell, M. (2005), “Introduction”, in D. Davidson, Truth, Language, and History. Oxford: Clarendon Press. Collingwood, R.G. (1938), The Principles of Art. Oxford: Oxford University Press. Collingwood, R.G. (1978), An Autobiography, with a new introduction by S. Toulmin. Oxford: Clarendon Press. Davidson, D. (1982), “Rational Animals”. Dialectica, 36, pp. 317-327; repr. in Subjective, Intersubjective, Objective. Oxford: Clarendon Press, 2001, pp. 95-106. Davidson, D. (1991), “Three Varieties of Knowledge”. Philosophy, 66, pp. 156166; repr. in Subjective, Intersubjective, Objective. Oxford: Clarendon Press, 2001, pp. 205-220. Davidson, D. (1997), “Gadamer and Plato’s Philebus”, in E.L. Hahn (ed.), The Philosophy of Hans-Georg Gadamer. Chicago: Open Court, pp. 421-432; repr. in Truth, Language, and History. Oxford: Clarendon Press, 2005. Davidson, D. (2001), “Foreword”, in B. Mou (ed.), Two Roads to Wisdom: Chinese and Analytic Philosophical Traditions. Chicago: Open Court. Donagan, A. (1962), The Later Philosophy of R.G. Collingwood. Chicago: University of Chicago Press. Dostal, R. (2011), “In Gadamer’s Neighbourhood”. In J. Malpas (ed.), Dialogues with Davidson: Acting, Interpreting, and Understanding. Cambridge, Mass.: MIT Press. Hegel, G.W.F. (1817), Enzyklopädie der philosophischen Wissenschaften, pt. 3 engl. transl. Encyclopedia III. Hegel, G.W.F. (1807), Phänomenologie Phenomenology of Spirit.

des

Geistes,

engl.

transl.

Honneth, A. (1996), The Struggle for Recognition: The Moral Grammar of Social Conflicts. London: Polity Press. Hoy, D. (1997), “Post-Cartesian Interpretation: Hans-Georg Gadamer and Donald Davidson”, in E.L. Hahn (ed.), The Philosophy of Hans-Georg Gadamer. Chicago: Open Court, pp. 111-128.

278

Malpas, J. (1997), “The Transcendental Circle”. Australasian Journal of Philosophy, 75, pp. 1-20. Malpas, J. (1999a), Place and Experience: A Philosophical Topography. Cambridge: Cambridge University Press. Malpas, J. (1999b), “The Constitution of the Mind: Kant and Davidson on the Unity of Consciousness”. International Journal of Philosophical Studies, pp. 1-30. Malpas, J. (2002), “Gadamer and Davidson on the Ground of Understanding”. In J. Malpas, U. Arnswald, and J. Kertscher (eds.), Gadamer’s Century: Essays in Honor of Hans-Georg Gadamer. Cambridge: MIT Press, pp. 195216. Malpas, J. (2005), “Sprache ist Gespräch: On Gadamer, Language and Philosophy”. In A. Wiercinski (ed.), Between Description and Interpretation: The Hermeneutic Turn in Phenomenology. Toronto: Hermeneutic Press, pp. 408-17. Malpas, J. (2006), Heidegger’s Topology: Being, Place, World. Cambridge, Mass.: MIT Press. Malpas, J. (2008), “On Not Giving Up the World—Davidson and the Grounds of Belief”. International Journal of Philosophical Studies, 16, pp. 201-215. Malpas, J. (2010), “The Beginning of Understanding: Event, Place, Truth”. In J. Malpas and S. Zabala (eds.), Consequences of Hermeneutics. Chicago: Northwestern University Press. Malpas, J. (2011a) (ed.), Dialogues with Davidson: Acting, Interpreting, Understanding. Cambridge, Mass.: MIT Press. Malpas, J. (2011b), “Ground, Unity, and Limit”. In Heidegger and the Thinking of Place. Cambridge, Mass.: MIT Press. Malpas, J. (2011c) (ed.), Heidegger and the Thinking of Place. Cambridge, Mass.: MIT Press. Malpas, J. (2011d), “Topology, Triangulation, and Truth”. In J. Malpas (ed.), Heidegger and the Thinking of Place. Cambridge, Mass.: MIT Press. Malpas, J. (2011e), “What is Common to All: Davidson on Agreement and Understanding”. In J. Malpas (ed.), Dialogues with Davidson: Acting, Interpreting, and Understanding. Cambridge, Mass.: MIT Press.

279

Malpas, J. and Zöller, G. (2011), “Reading Kant Topographically: From Critical Philosophy to Empirical Geography”. In R. Baiasu, A. Moore, and G. Bird (eds.), Contemporary Kantian Metaphysics Today. London: Palgrave Macmillan. Malpas, J. and Thiel, K. (2011), “Kant’s Geography of Reason”. In S. Elden and E. Mendieta (eds.), Reading Kant’s Geography. New York: SUNY Press. Mead, G.H. (1934), Mind, Self and Society. Chicago: University of Chicago Press. Rorty, R. (2000), Philosophy and Social Hope. Harmondsworth: Penguin.

CONTRIBUTORS

1. M. Cristina AMORETTI, University of Genoa, Italy. 2. Dorit BAR-ON, University of North Carolina, Chapel Hill, USA. 3. Mario DE CARO, Roma Tre University, Italy. 4. Ingvald FERGESTAD, University of Oslo, Norway. 5. Sanford C. GOLDBERG, Northwestern University, Evanston, USA. 6. Kirk LUDWIG, Indiana University, Bloomington, USA. 7. Jeff MALPAS, University of Tasmania, Hobart, Australia. 8. Gerhard PREYER, Goethe-University, Frankfurt am Main, Germany. 9. Matthew PRISELAC, University of North Carolina, Chapel Hill, USA. 10. Bjørn RAMBERG, University of Oslo, Norway. 11. Adina L. ROSKIES, Dartmouth College, Hanover, USA. 12. Fredrik STJERNBERG, University of Linköping, Sweden. 13. Claudine VERHEGGEN, York University, Toronto, Canada.

NAME INDEX

Abbott, E.A. 9. Adler, J. 163 Amoretti, M.C. 17, 19, 23, 51 Audi, R. 53 Bar-On, D. 21 Berneker, S. 53 Berkeley, G. 14 Block, N. 10, Brandom, R. 22, 221, 224, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 244, 245, 246, 247, 248, 249, 250, 255, 273 Bridges, J. 225, 226, 227, 244, 251 Briscoe, R.E. 18 Brown, J. 13 Burge, T. 10, 11, 12, 50, 116, 181, 182, 183, 184 Call, J. 113 Carruthers, P. 141 Cavell, M. 266 Champ, E. 113 Child, W. 18 Cheney, D. 113, 137 Chomsky, N. 114 Collingwood, R. G. 274 Conee, E. 53 De Caro, M. 18, 21 Davidson, D. passim. Descartes, R. 11, 62 DeRose, K. 163 Dewey, J. 271 Douven, I. 163 Dretske, F. 11, 53, 236

Dummett, M. 19 Edwards, S.D. 12 Eilan, N. 137 Engel, P. 18 Evans, G. 11 Fennell, J. 17 Fergestad, I. 22 Fitch, T. 114 Fodor, J. 11, 200, 201, 205, 210, 216 Føllesdal, D. 17, 228 Frege, G. 11 Freud, S. 254 Gadamar, H.-G. 268, 269, 272 Gauker, C. 166 Glüer, K. 110 Goldberg, S.C. 21, 161 Goodman, N. 207 Grice, P. 140 Hahn, L.E. 18 Harman, G. 11 Hauser, M. 114, 134 Hawthorne, J. 163 Heidegger, M. 267, 269, 270, 272 Heil, J. 18, 185 Hempel, C. 200 Hegel, G.W.F. 271, 273 Hüntelmann, F. 23 Hume, D. 11, 14 Hurford, J. 114 Husserl, E. 11

284 Jackendorf, R. 114 Jackson F. 11 James, W. 271 Kant, I. 31, 224, 267, 271, 272, 273 Kojève, K. 273 Kripke, S. 184 Kvanvig, J. 163 Lackey, J. 163 Lepore, E. 11, 17, 22, 179, 180, 187, 189, 200, 201, 203, 204, 205, 211, 216 Loar, B. 11 Locke, J. 14 Ludlow, P. 13 Ludwig, K. 11, 17, 20, 22, 179, 180, 187, 189, 203, 204, 205, 211 Macdonald, C. 12 MacFarlane, J. 163 Malcolm, N. 78, 189 Malpas, J. 22, 258, 263, 266, 267, 269, 270, 272 Martin, N., 13 McDowell, J. 11, 273 McGinn, C. 11, 12 Mead, G. H. 270, 271 Miklósi, A. 134 Montminy, M. 18 Nagel, T. 12, 62, Neurath, O. 214 Pagin, P. 17, 222, 223, 225, 251 Pinker, S. 114 Place, U.T. 9 Preyer, G. 22, 23 Priselac, M. 21

Putnam, H. 10, 11, 50, 62, 181, 182, 185, 186 Ramberg, B. 22, 39 Rorty, R. 265, 276 Roskies, A. L. 20 Royce, J. 271 Ryle, G. 213 Schlick, M. 214 Searle J. R. Sellars, W. 9, 200 Seyfarth, R. 113, 137 Sinclair, R. 18 Sosa, E. 40, 60 Smart, J. C. 9, 10 Snell, W. 259 Stanley, J. 163 Stjenberg, F. 18, 20 Stroud, B. 32, 33, 36, 37, 38, 39, 40, 41, 42, 43, 228, 229 Thiel, K. 267, 272 Tomasello, M. 113 Unger, P. 163 Verheggen, C. 17, 19, 101 Quine. W.v.O. 9, 98, 155, 200, 201, 205, 206, 208, 212, 213, 216, 248, 255, 257, 270 Weiner, M. 163 Williamson, T. 163, 164 Wittgenstein, L. 16, 41, 131, 181, 184, 213, 270 Wright, C. 13 Zöller, G. 227

SUBJECT INDEX

Anit-skeptical argument 58 American philosophy of mind 9 Accessibilism 53 Cartesian (-ism) 12, 13 first person thinking 13 intuition 13 ontological 12 skeptic 31 Charity 155-156 epistemological 155 Common cause 82 Concept constitution 230 Content-determination argument 115 see also triangulation Concept possession 76 Davidson’s coherentism 54, 55 Dilemma in the theory of knowledge 214 Dualism of scheme and content 190

weak 12 First person authority 200, 203 Frege’s turn 14 Foundationalism 47-48, 265 Functionalism 10 Fusion of horizons (Horizontverschmelzung) 269 Internalism 9-10 token 12 type 12 Holism 204 semantic 200-201 Justification coherence theory of 48, 53 Linguistic division of labor 186 Mentalism 53

Epistemic restrictions 22 Expressive behavior 21, 123, 132 Externalism 9-10 diachronic 11 epistemic 52 language-grounded 205, 212 non-reductive 11 orthodox 50 perceptual 182 physical (causal) 11 reductive 11 semantic 33-34, 179, 180 social (linguistic, interpretative) 181 strong 12 synchronic 11 type 12

Objectivity argument from 110, 112 and communication 109-110 concept of 16, 73 Omniscient interpreter 187, 192, 199 and Swampman 192 Philosophical topography 23, 258 Physicalism token 11 Principle of coherence 71-72, 198-199, 201 of correspondence 71, 198-199, 202 Publicity 214-215

286 Radical interpretation (RI) ambitious version of 198, 205 epistemic restrictions of 205 modest project 205 Reliabilist epistemology 235 Second person 170 Self-conception 168 Thought experiment Swampman 187 Twin Earth scenario 185 Third person global 197 Token-token-identity theory 10 Transcendental argument 84 Triangulation 15, 69-70, 81, 105-106, 109, 122, 207, 259

and communication 37, 70-71, 107 epistemological dimension of 153 intermediated 131 and language (and thought) 38, 222 line of 209 and location 261-262 and mental content 17 and objectivity 100-101 and philosophical skepticism 32, 42 and proto-objectivity 123 and propositional thought (and language) 15 reflective 123, 131 Type-type-identity theory 10

PhilosophischeAnalyse PhilosophicalAnalysis 1 Herbert Hochberg Russell, Moore and Wittgenstein The Revival of Realism

8 Rafael Hüntelmann Existenz und Modalität Eine Studie zur Analytischen Modalontologie

2

9 Andreas Bächli / Klaus Petrus Monism

ISBN 3-937202-00-5 334 pp., Hardcover € 94,00

Heinrich Ganthaler Das Recht auf Leben in der Medizin Eine moralphilosophische Untersuchung

ISBN 3-937202-01-3 167 pp., Hardcover € 58,00

3 Ludger Jansen Tun und Können Ein systematischer Kommentar zu Aristoteles’ Theorie der Vermögen im neunten Buch der „Metaphysik“ ISBN 3-937202-02-1 302 pp., Hardcover € 70,00

4 Manuel Bremer Der Sinn des Lebens Ein Beitrag zur Analytischen Religionsphilosophie ISBN 3-937202-03-X 134 pp., Hardcover € 58,00

5 Georg Peter Analytische Ästhetik Eine Untersuchung zu Nelson Goodman und zur literarischen Parodie ISBN 3-937202-04-8, 332 pp. Hardcover € 94,00

6 Wolfram Hinzen / Hans Rott Belief and Meaning Essays at the Interface ISBN 3-937202-05-6 250 pp., Hardcover € 58,00

7 Hans Günther Ruß Empirisches Wissen und Moralkonstruktion Eine Untersuchung zur Möglichkeit von Brückenprinzipien in der Natur- und Bioethik ISBN 3-937202-06-4 208 pp., Hardcover € 58,00

ISBN 3-937202-07-2 189 pp., Hardcover € 58,00

ISBN 3-937202-19-6 340 pp., Hardcover € 70,00

10 Maria Elisabeth Reicher Referenz, Quantifikation und ontologische Festlegung ISBN 3-937202-39-0 ca. 300 pp., Hardcover € 89,00

11 Herbert Hochberg / Kevin Mulligan Relations and Predicates ISBN 3-937202-51-X 250 pp., Hardcover € 74,00

12 L. Nathan Oaklander C. D. Broad's Ontology of Mind ISBN 3-937202-97-8 105 pp., Hardcover € 39,00

13 Uwe Meixner The Theory of Ontic Modalities ISBN 3-938793-11-2 374 pages, Hardcover,€ 79,00

14 Donald W. Mertz Realist Instance Ontology and its Logic ISBN 3-938793-33-3 252 pp., Hardcover, EUR 79,00

15 N. Psarros / K. Schulte-Ostermann (Eds.) Facets of Sociality ISBN 3-938793-39-2 370 pp., Hardcover, EUR 98,00

16 Markus Schrenk The Metaphysics of Ceteris Paribus Laws ISBN 13: 978-3-938793-42-8 192pp, Hardcover, EUR 79,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier

PhilosophischeAnalyse PhilosophicalAnalysis 17 Nicholas Rescher Interpreting Philosophy The Elements of Philosophical Hermeneutics

25 Laird Addis Mind: Ontology and Explanation Collected Papers 1981-2005

18 Jean-Maurice Monnoyer(Ed.) Metaphysics and Truthmakers

26 Hans Bernhard Schmid, Katinka Schulte-Ostermann, Nikos Psarros Concepts of Sharedness Essays on Collective Intentionality

ISBN 978-3-938793-44-2 190pp., Hardcover € 89,00

ISBN 978-3-938793-32-9 337 pp., Hardcover € 98,00

19 Fred Wilson Acquaintance, Ontology, and Knowledge Collected Essays in Ontology ISBN 978-3-938793-58-9 XX, 726., Hardcover, EUR 159,00

20 Laird Addis, Greg Jesson, and Erwin Tegtmeier (Eds.) Ontology and Analysis Essays and Recollections about Gustav Bergmann ISBN 978-3-938793-69-5 312 pp., Hardcover, EUR 98,00

21 Christian Kanzian (Ed.) Persistence ISBN 978-3-938793-74-9 198pp., Hardcover, EUR 79,00

22 Fred Wilson Body, Mind and Self in Hume’s Critical Realism ISBN 978-3-938793-79-4 512pp., Hardcover, EUR 139,00

23 Paul Weingartner Omniscience From a Logical Point of View ISBN 978-3-938793-81-7 188pp., Hardcover, EUR 79,00

24 Simone Gozzano, Francesco Orilia Tropes, Universals and the Philosophy of Mind Essays at the Boundary of Ontology and Philosophical Psychology

ISBN 978-3-938793-86-2 289pp., Hardcover, EUR 79,00

ISBN 978-3-938793-96-1 306pp., Hardcover, EUR 89,00

27 Holger Gutschmidt , Antonella Lang-Balestra, Gianluigi Segalerba (Hrsg.) Substantia - Sic et Non Eine Geschichte des Substanzbegriffs von der Antike bis zur Gegenwart in Einzelbeiträgen ISBN: 978-3-938793-84-8 565pp., Hardcover, EUR 149,00

28 Rosaria Egidi, Guido Bonino (Eds.) Fostering the Ontological Turn Gustav Bergmann (1906-1987) ISBN 978-3-86838-008-8 274pp., Hardcover, EUR 89,00 29 Bruno Langlet, Jean-Maurice Monnoyer (Eds.) Gustav Bergmann Phenomenological Realism and Dialectical Ontology ISBN 978-3-86838-035-4 235pp., Hardcover, EUR 89,00 30 Maria Elisabeth Reicher (Ed.) States of Affairs ISBN 978-3-86838-040-8 219pp., Hardcover, EUR 79,00 31 Richard Schantz (Hrsg.) Wahrnehmung und Wirklichkeit ISBN 978-3-86838-042-2 252 Seiten, Hardcover, 89,00 EUR

ISBN 978-3-938793-83-1 196pp., Hardcover, EUR 69,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier

PhilosophischeAnalyse PhilosophicalAnalysis 32 Javier Cumpa & Erwin Tegtmeier (Eds.) Phenomenological Realism Versus Scientific Realism Reinhardt Grossmann – David M. Armstrong. Metaphysical Correspondence

39 Pedro Schmechtig, Gerhard Schönrich (Hrsg.) Persistenz, Indexikalität, Zeiterfahrung ISBN 978-3-86838-112-2 408 Seiten. Hardcover, EUR 119,00

ISBN 978-3-86838-051-4 139pp., Hardcover, EUR 69,00

33 Christan Kanzian Ding – Substanz – Person. Eine Alltagsontologie

ISBN 978-3-86838-057-6 353 Seiten, Hardcover, EUR 39,00

34 Uwe Meixner Modelling Metaphysics The Metaphysics of a Model ISBN 978-3-86838-060-6 274pp., Hardcover, EUR 79,00

35 Roberto Poli (Ed.) Causality and Motivation ISBN 978-3-86838-068-2 192pp., Hardcover, EUR 79,00

36 Manuel Bremer Universality in Set Theories A Study in Formal Ontology ISBN 978-3-86838-071-2 125pp., Hardcover, EUR 69,00

37 Meinard Kuhlmann The Ultimate Constituents of the Material World In Search of an Ontology for Fundamental Physics ISBN 978-3-86838-072-9 278pp., Hardback, EUR 98,00

38 Franck Lihoreau (Ed.) Truth in Fiction ISBN 978-3-86838-092-7 305pp., Hardcover, EUR 98,00

EditedBy • HerbertHochberg • RafaelHüntelmann ChristianKanzian • RichardSchantz • ErwinTegtmeier