I do not accept Quine’s account of the nature of knowledge, which is essentially first person and Cartesian. But I do find congenial Quine’s resolutely third person approach to epistemology, and to the extent that the naturalization of epistemology encourages such an approach, I am happy to count myself a naturalized epistemologist. (Davidson 1991, 193)
1 Introduction
While deeply influenced by W. V. Quine’s work on radical translation and naturalized epistemology, Donald Davidson has continually criticized what he sees as the central defects with Quine’s approach. Many of these criticisms resurface in the recently discovered 1986 Stanford conversations (hereafter ‘conversations’), where Davidson reviews the skeptical and subjectivist implications of Quine’s epistemology (Marschall 2026). These conversations then serve as a key episode in the long-standing philosophical debate between Davidson and Quine. There is reason to think that Davidson’s criticisms played an influential role in Quine’s modifications to his view of observation sentences and use of private stimulus meaning (Marschall 2026; Kemp 2022; Quine 1998). However, the conversations also highlight an important transition in Davidson’s critical perspective and his own positive epistemological views. Here, he focuses more specifically on Quine’s use of stimulus meaning and its role in translation, arguing that rather than sensory stimulation at our surfaces, Quine should favor distal objects and events as central for the determination of content. Davidson’s animadversions against stimulus meaning provide the key critical framework that he will continue to use to distinguish his view from Quine’s, what I will call his ‘proximal-distal analysis’. The conservations then serve, I argue, as a key formative influence in Davidson’s evolving critical stance towards Quine’s view, and in furthering his own externalist approach to epistemology. In order to offer further support for these claims, Section 2 begins with a summary overview of Davidson early criticisms of Quine, specifically his introduction of the third dogma of empiricism and its implications for understanding evidence. Section 3 then considers several key exchanges from the 1986 conversations that signal Davidson’s critical focus on Quine’s use of stimulus meaning and his further distinction between the proximal and distal views of meaning and evidence. Turning to Section 4, we see this distinction become the central framework for Davidson’s continued criticism of Quine’s theory of knowledge and the springboard for his own externalist approach to epistemology. This approach is furthered with his introduction of ‘triangulation’, which adds an irreducible social element to content determination and human knowledge. Lastly, using further points made in the conversations, I conclude by wondering to what extent this view can been seen as a contribution to naturalized epistemology, suggesting that contrary to Davidson and other commentators, it points to a more fundamental break with Quine than he seems to recognize or allow.
2 Early Criticisms and ‘A Coherence Theory of Truth and Knowledge’
By the time of the conversations in 1986, Davidson had already made several well-known criticisms of Quine’s epistemology. These center on his introduction of the third dogma of empiricism with its relativist and skeptical implications, as well as its connection to the problem of evidence and epistemic justification. In order to chart the changes in Davidson’s critical approach to Quine in the conversations, it will be useful to review these early criticisms. In addition, I will outline the constructive alternative of knowledge that falls out of these criticisms, where Davidson increasingly relies on his content externalism as providing the proper basis for the theory of knowledge. We will see that while these early criticisms continue to inform Davidson’s critical perspective, it is only in the conversations that he explicitly challenges Quine’s use of stimulus meaning and its role in translation.
Davidson’s initial criticisms are found in his classic ‘On the Very Idea of a Conceptual Scheme’ where he introduces the third dogma of empiricism, or scheme-content distinction, which posits ‘an organizing system, and something waiting to be organized’ (1974, 189).1 Various passages from Quine’s work are seen as emphasizing this distinction, with Davidson noting especially the following claim:
… we can investigate the world, and man as a part of it, and thus find out what cues he would have of what goes on around him. Subtracting his cues from his world view, we get man’s net contribution as the difference. This difference marks the extent of man’s conceptual sovereignty—the domain within which he can revise theory while saving the data. (Quine 1960, 5)
Here he detects a clear division between a scheme, depicted as a ‘totality of beliefs’ or ‘world view’, and content, described as ‘experience’ or ‘data’. Quine’s specific formulation of this dogma centres on his chief concern over how, on the basis of our sensory contact with the world, we are able to arrive at our current scientific theory. This results in the ‘evidence’ for both the meaning of our words and our knowledge of the world as consisting of the stimulation of our sensory organs. The further issue concerns exactly why this distinction should be seen as detrimental to Quine’s epistemology. For Davidson, the main problem concerns the view of justification and evidence that is suggested by Quine’s position and which emerges from this emphasis on the division between scheme and content:
The general position is that sensory experience provides all the evidence for the acceptance of sentences (where sentences may include whole theories). A sentence or theory fits our sensory promptings, successfully faces the tribunal of experience, predicts future experience, or copes with the pattern of our surface irritations, provided it is borne out by the evidence. (Davidson 1974, 193)
Davidson’s suggestion is that accepting a division between theory and data as seen in the third dogma motivates a search for the justification of our beliefs through comparison with this data. Doxastic justification is understood in terms of answerability to whatever is given in experience, in Quine’s case, the sensory promptings that serve as our ultimate ‘take’ on what happens in our surrounding scene. Consequently, Quine’s advocacy of the third dogma results in what Davidson calls the ‘ultimate evidence’ for our knowledge of the world consisting of the stimulation of our sensory surfaces (1989b, 161–62).
Davidson’s key critical point wonders whether sensory input can properly play this evidential role.2 These issues are further explored in Davidson’s misleadingly named ‘A Coherence Theory of Truth and Knowledge’ (1990a), (hereafter ‘Coherence’) a central transitional piece in Davidson’s philosophical development.3 Here, Davidson provides his most thorough discussion of how the scheme-content distinction encourages taking the invariant content or the epistemic ‘given’ as having some special epistemic role. He considers various attempts to ground our beliefs on what is given in the senses, explaining that they are motivated by the seemingly obvious claim that sensations connect our beliefs with the world, and are able to justify such beliefs because we are aware of them. However, Davidson notes a central problem with such accounts: ‘The trouble … is that the justification seems to depend on awareness, which is just another belief’ (1990a, 124). Since it is only beliefs that provide information about the world, it is they that serve to justify other beliefs, leaving sensations without any justificatory role. The relationship between sensations and belief must be understood solely in causal terms, and cannot without confusion be taken as a justificatory relation. Davidson sums up these points in the following way:
This causal relation cannot be a relation of confirmation or disconfirmation, since the cause is not a proposition or belief, but just an event in the world or in our sensory apparatus. Nor can such events be considered in themselves evidence, unless, of course, they cause us to believe something. And then it is the belief that is properly called the evidence, not the event. (Davidson 1982a, 486)
Now it is here that Davidson claims Quine’s naturalized epistemology makes a fatal error, since by claiming that knowledge is epistemologically grounded in some ultimate source of evidence it wrongly assigns a justificatory role to sensation. Quine’s use of observation sentences, sentences keyed to stimulations that all speakers will assent to in response to the same stimulations, is precisely for Davidson an attempt to utilize sensations so they can serve as justificatory vehicles.
However, this runs counter to the point we have just seen, that sensations only play a causal role in the production of beliefs. His argument for this claim is a familiar one, and emphasizes how sensory input lacks propositional content, and cannot then stand as a potential reason for a belief:
The reason sensations, percepts and sense-data cannot provide epistemic support for beliefs is simple: reasons have to be geared conceptually to what they are reasons for. The relation of epistemic support requires that both relata have propositional content, and entities like sensations and sense-data have no propositional content.4 (Davidson 1997, 22)
Appeals to the sensory given, even in terms of Quine’s more scientifically acceptable neural intake, does not provide the needed content required for serving as a reason. Neural input remains epistemologically irrelevant for the question of the evidential support of our beliefs. Davidson then concludes that any attempt to ground knowledge on some form of non-propositional ‘experience’ is untenable, and since Quine’s naturalized epistemology fails to recognize this, it too fails to adequately deal with the question of epistemic justification.5
In ‘Coherence’, Davidson continues to explore the consequences of this criticism for the theory of knowledge. With his emphasis on coherence, Davidson seeks to distance himself from empiricist alternatives, since ‘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’ (1990a, 123). He further argues that we can avoid the pitfalls of the third dogma by reflecting on the nature of interpretation and its consequences for how our beliefs come to have the contents they do. Correctly understood, this will reveal that our beliefs cannot be massively in error and that belief is in its nature veridical. By elaborating on these considerations, we can recognize the increasing role externalism plays in the continued development of Davidson’s epistemology.
Davidson argues that his proposed methodology of interpretation, what he calls ‘radical interpretation’, rules out the possibility that an individual could be mostly wrong about their beliefs. His argument has two parts. The first emphasizes that correctly interpreting the speech and thought of another reveals that most of that person’s beliefs must be true, thereby providing good reason to conclude that any single belief that coheres with the rest will also be true. Secondly, he claims that by noting that anyone with thoughts, who is then capable of wondering if their beliefs are correct, must know what a belief is, and how such beliefs are to be detected and interpreted. Given that these are general facts that everyone must appeal to in order to successfully interpret others, we have a reason for the overall truthfulness of our own beliefs. He then further explains it becomes pointless to ask for a further reason that can then be added to our web of belief, concluding with this surprising claim about our interpreter, ‘All that is needed is that he recognizes that belief is in its nature veridical’ (1990a, 128). In order to clarify the proper philosophical import of this potentially misleading conclusion, we need to note the way beliefs and their content are determined through interactions with others and the external environment. Stated briefly, it is the truth of externalism that supports Davidson’s veridical nature of belief thesis.
In order to appreciate this, we must remind ourselves of some important details of Davidson’s account of radical interpretation.6 Here we are introduced to the radical interpreter, an individual who knows how to communicate but who has no knowledge of the language that he is currently attempting to understand. The interpreter needs to take advantage of the events in the surrounding environment that prompts the interlocutor’s utterances. In this way, the interpreter uses prompted assent as basic by focusing on the causal relations between utterances and the world, taking an event as causing a specific utterance and then later prompting assent from the speaker. Davidson further explains, ‘This is a fair place to start the project by identifying beliefs and meanings, since a speaker’s assent to a sentence depends both on what he means by the sentence and on what he believes about the world’ (1990a, 128). The way in which the interpreter is able to isolate evidence for the interpretation of the speaker’s utterance is by an appeal to what is known as the principle of charity. Here, the interpreter takes the assertions made by the speaker as correct in an effort to provide an interpretation that preserves truth. Such charity counsels us to interpret by reading our own standards of truth into the pattern of sentences held true by the speaker in order to make that speaker intelligible to us. Without an appeal to such standards of rationality, we would be unable to develop a theory of interpretation because it is the appeal to charity that enables the events that prompts a speaker’s utterances to serve as evidence for our interpretation. The empirical happenings surrounding an utterance can now be seen as relevant to the truth of what a speaker says and therefore can be used in the production of hypotheses concerning what the speaker means by an utterance.
In this way, the principle of charity serves as a necessary pre-condition for the possibility of communication whatsoever and plays a crucial role in establishing the claim that we cannot be massively wrong about our beliefs: ‘It is an artifact of the interpreter’s correct interpretation of a person’s speech and attitudes that there is a large degree of truth and consistency in the thought and speech of an agent’ (Davidson 1990a, 131). At this point the obvious reply emphasizes that this was achieved through the use of the interpreter’s standards of truth and consistency, and in that case, we cannot rule out the possibility that both interpreter and speaker understand one another on the basis of shared but largely incorrect beliefs.
What is further needed is a characterization of the causal patterns that emerge through our interaction with the world and others. Here we make explicit how the actual situation that we find ourselves in, the complex causal nexus of objects and persons that interact with us, reveals that we cannot generally be wrong about what others think or about what the world is like. In ‘Coherence’ Davidson begins to move in this direction with his increasing emphasis on semantic externalism and its epistemological consequences.
This commitment to externalism is most explicit with this passage: ‘What stands in the way of global scepticism 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’ (1990a, 132).7 Here, externalism emphasizes that the content of our most basic beliefs and our verbal expressions of them are directly tied to the external objects that they describe. It is here that we can finally recognize the upshot of Davidson’s remark about the veridical nature of belief. There can be no general worries about the correctness of our perceptual beliefs precisely because their content is, at the very start, ostensively tied to objects and events in the world. The ability to communicate is then dependent on the causal lines from object to interpreter that converge on the objects in the world that determine the subject of conversation. This provides an answer to someone who reflectively wonders what reason he has to think that most of his beliefs are correct. Since the objects of beliefs must be taken as the causes of belief, beliefs are by their very nature the sorts of things that accurately depict what the world is like. We thus have a reason to take it that our beliefs cannot be radically mistaken, since the nature of successful interpretation requires that our most basic beliefs be accurately fixed to the world they are about.
This does not claim that there is a special class of beliefs that can never be questioned, since even our ostensive beliefs may be questioned once additional learning has provided the context needed to question them. Rather, what is being emphasized here is that these initial beliefs need to be directly anchored to the world in order to be able to develop a system of concepts within which they can later be questioned. These basic beliefs cannot be questioned as to their general correctness, because from within this account of interpretation, words are now being given a meaning for the first time, and there is then none of the further context needed that is required for the possibility of error. Davidson’s reflections on interpretation are then intended to establish the general externalist thesis that the content of our thoughts and utterances are partly determined by causal interaction with our local environment (1991, 201). We will see that his later introduction of the device known as ‘triangulation’ addresses an additional question left open here and in the conversations, concerning how social interaction helps to further locate the usual or normal cause of a speaker’s basic thoughts and utterances.8
3 The Stanford Conversations: The Problem with Stimulus Meaning
These early criticisms of Quine’s naturalized epistemology and the increasing emphasis on distal objects and events as both the cause and content of our beliefs play an important role in Davidson’s exchanges with Quine during the 1986 Stanford conversations (with notable assistance from the two other discussants Burton Dreben and Dagfinn Føllesdal). This section reviews some relevant parts of these conversations, where Davidson repeats well-known criticisms, but also gives further indication of his own emerging epistemological views. The conservations themselves cover a lot of philosophical ground. I will be especially interested in highlighting a key change in his critical approach, where he increasingly focuses on what he sees as Quine’s problematic introduction of stimulus meaning and its relation to evidence. This in turn paves the way for his later proximal-distal analysis, which serves as the critical framework for his ongoing assessment of the merits and problems with Quine’s views of meaning and knowledge. In surveying these discussions, my concern is with locating the precise role Davidson’s critical discussion of stimulus meaning plays in his proximal-distal analysis that is central for the later development of his epistemology.
First consider the following remarks, where Davidson lays out his agenda for their discussion:
… This is an old thing that we’ve discussed many times, but I want to start it again, that is the question what the ultimate evidence is and what we should use as the basis for our theory of meaning and our epistemology … I think it’s a mistake to translate or interpret somebody on the basis of similarities of patterns of stimulation, and I’ll say why I think it’s a mistake. But the conclusion is that’s the wrong place to look. I think I’m as much of a naturalist as Quine. I mean I start out assuming that science is more or less right, that what we think about the world is more or less the way it is. And my purpose in talking about these things is to put them in such a way as not to get into trouble … I think myself that if one uses something like stimulus meaning and one starts out with science as we have it and one gets into a position that invites skepticism, sort of science undermining itself, but one can be a naturalist and not take quite the same tack in one’s analysis of meaning and evidence. Now, many passages in Quine suggest a different course than the one that I think is the right one, and that is that [in] interpreting somebody else what is relevant to interpreting and understanding isn’t the pattern of stimulations basically. At a certain sophisticated point, we can take that into consideration, but at the beginning what we look for is ordinary things in the outside world like cats and books. And if we find that the presence of a book is what prompts somebody to assent to “That’s a book”, then our best first guess is that the right way to understand that is, it’s a sentence which is true if and only if there’s a book there. Now this says nothing about patterns of stimulation whatsoever. And I’ve got a list of passages over the years in which you have said just this. The publicly discernible thing that both speaker and interpreter can easily observe is the thing, so to speak, on which you look for agreement, books and cats, and says nothing about patterns of stimulations. (Marschall 2026, 189–90)
Here, Davidson highlights once again his concern with the search for ‘ultimate evidence’ within the study of knowledge or translation, with Quine favoring sensory stimulation at our surfaces, or stimulus meaning as the final support for our translations. He further discusses his own alternative, one that he claims is consistent with Quine’s naturalist approach in seeking the basis for the meaning of sentences in the external objects that prompt them. This is the view we have seen Davidson defend in ‘Coherence’, where successful interpretation depended on exploiting the connection between utterances and external objects and the fact that this connection in part determines the content of those utterances. Davidson further notes these two views of meaning and evidence can both be found in Quine’s writings. He is then providing a clear distinction between what he will soon call the proximal and distal views of meaning, which figures more prominently in his later critical discussions of Quine. The discussion continues in this later exchange where Davidson’s perspective is further developed as they continue to assess the merits and weaknesses of Quine’s continued focus on stimulus meaning:
Davidson: Well, I want to say two things. First, we can’t say when two people’s pattern of stimulation are the same except by referring to the book. Similarity is not visual patterns or anything like that at all.
Quine: What I suspected was that the word “pattern” was perhaps causing misunderstanding. I don’t think we should use the word. I have used it I’m afraid I’ve also avoided it in various places. It’s the class of all the distributions of activated nerve endings. Each one of those totalities of activated nerve endings at a moment being a stimulation. The class of all those stimulations that would prompt assent to the question, “Is that a book?” would be the stimulus meaning of “That’s a book.” And that’s what I was trying to say in Word and Object.
Da: Van, this seems to me to be hopelessly circular if we’re really going to put it that way. Surely what Word and Object supposes is that you can find out that there’s some similarity in those stimulations which would cause you to assent to so and so and the class of totalities of stimulations that would prompt me to do this. We can’t begin by assuming that they have the same meaning. Stimulus meaning is something I thought that could be specified without talking about the book.
Q: No. Stimulus meaning is not something that could be specified. I can talk about them yes. I can describe them without mentioning books, but the stimulus meaning for “That’s a book” can’t be specified in any way that I can imagine without bringing in the book.
Da: I see, well that certainly puts it into light. In that case I guess since we can’t say what makes these classes the same classes for two people except by referring to the book, I don’t see myself what work the notion of stimulus meaning is doing.
Q: What puts them together in respect to relevant similarity is concurrence of witnesses. And that’s the distinctive feature of what I call observation sentences. And as for the [?] witnesses, they’re sharing the same class of stimulations. This fellow is getting one stimulation and this one another. But if they were over here they would still be getting stimulations in that same class. It’s the class of all the stimulations that would prompt assent to members of the community.
Da: OK, but now it seems to me quite obvious that the notion of stimulation is idle here. What we have is the presence of books activating a certain response. The notion of stimulation, then, as far as I can see, is doing no work at all. We have a causal relation between books on the one hand and assent on the other hand. There’s a causal relation between these two things. What the intermediate stages of that are is of course very interesting to science, and there’s lots more to it than just the tips of the sense organs.
Dreben: That’s the whole point. This is all part of naturalized epistemology.
Da: I’m not saying anything that goes against naturalized epistemology. My point is, once we understand what it is to say the stimulus meaning for you and for me is the same is explained only in terms of the causal role of the book, then I don’t see any reason to pick out any talk about stimulation. The point is we’ve got a causal chain there that begins with a book and ends with a certain attitude towards or a certain response to a sentence. And how that causal chain goes is certainly for science to discover. But I don’t see how stimulation plays any particularly important epistemological role compared to anything else in the causal chain. (Marschall 2026, 195–96)
The background question to this discussion concerns the role of stimulus meaning in translation. Davidson had thought that successful translation was dependent on similarity of stimulus meaning, roughly, similarities between neural stimulation. Quine here corrects this misunderstanding explaining that the stimulus meaning of an utterance, ‘that’s a book’, cannot be specified independently of reference to the book. This response fits Davidson’s own favored position, where translation is determined through interaction with external objects. It then becomes unclear, however, what role Quine’s emphasis on sensory stimulation plays in translation. Dreben insists that it still plays an important role in Quine’s naturalized account of knowledge, suggesting that it remains part of a scientific explanatory account of how humans are capable of moving from sensory stimulation to scientific knowledge. Davidson sees the value of a scientific account of the causal chain from the book to our utterances and insists that he is not undermining Quine’s naturalized epistemology. However, given that the distal object is required for determining the meaning of the utterance, he no longer understands why the appeal to proximal stimulation retains any central role in such an epistemological account.
This last exchange builds on this final issue where Davidson further clarifies what is at issue with his emphasis on the lack of epistemological significance found in Quine’s lingering commitment to stimulus meaning or proximal sensory stimulation:
Q: My reason for taking the nerve endings is proximity. Of course there are lots of causes out there, and the farther out you go the more complicated it gets. I would just as soon take things in the thalamus or God knows where. If I knew just what cells mattered for triggering these things that would be better still. But there are two reasons not to. One is not knowing those things and other people not knowing them either. And the other is the matter again of possible irrelevant difference between people, irrelevant because the impacts on the sensory surfaces are in a rough and ready way verifiable, controllable by one’s neighbor. And it’s from one’s neighbor that one gets the language. And one’s neighbor doesn’t blame you for not assenting to an observation query for instance because you’re facing the wrong way. He sees that the relevant sensory endings aren’t being hit. That amateur physical psychology is at work constantly and teaching language. But, on the other hand, the more removed from immediacy of the surfaces the causes are, the more chance there is of the slip between cup and lip.
Da: Right. That’s where I said the [?] would play. However, I want to keep repeating that I obviously don’t disagree about the interest of the scientific account of all of that. What I object to is assigning an epistemological significance to someone or part of it, and that’s the part that more or less corresponds to the empiricist tradition.
Dr: And what does “epistemological” mean here?
Da: Taking the neutral thing which is unorganized and which we all have to work on as sensations, something like that. This informal notion of evidence located more or less at the skin, calling that what we have to go on where it seems that it could be equally well said that we have to go on what’s out there or what’s happening in our brain. If that just is a scientific term about the causal chain, that’s not giving it any epistemological significance. You haven’t made it play any special role in translating other people or therefore in explaining what they mean. I think the clearest way to put this is to say you’ll give it some part of this causal chain, and epistemological significance is to fasten on that part of the chain as the place where we look to check up on or explain translation, meaning, understanding if that’s also the point at which we explain knowledge. As Quine has said again and again, theory of knowledge and theory of meaning have the same starting point. Now as far as I can see, it is as we’ve been saying, just the difference between something like phenomenalism or something like Russell’s empiricism or even Hume’s for that matter. There is this subjective stopping point. And what Quine shows us is it can be subjective without being mentalistic. It’s physicalistic, and it’s just a matter of where you take it. I think that’s always the sharpest way to put the difference. To give it an epistemological significance just means that’s the point you’ve hit upon as the thing to pay attention to in transaction given your amount[?] of knowledge. (Marschall 2026, 367–68)
Here, Davidson expands on an issue that becomes increasingly important for his later critical assessments of Quine’s epistemology. He connects the epistemological significance of Quine’s appeal to proximal stimulation and private stimulus meaning to his earlier criticism of the third dogma of empiricism with its search for ultimate evidence. While this might seem like a straightforward implication of Quine’s naturalized empiricism, it is an emphasis that Davidson had not explicitly made in his earlier criticisms. As we have seen, the distinction between scheme and content encourages the location of some neutral aspect of our experience that serves as the final arbiter for empirical knowledge. In Quine’s case this takes the form of sensory stimulation, now including the use of stimulus meaning, which takes on a special role in the theory of knowledge and translation in terms of what ultimately tests our translations or more generally serves to justify our knowledge of the world. Davidson returns to the key question involving why Quine continues to assign a special epistemic importance to sensory stimulation. If stimulus meanings themselves cannot be determined independently of the external objects that prompt a speaker’s assent, they become irrelevant for epistemology. Once again, the causal chain from object to utterance retains its scientific importance, but once sensory stimulation is no longer viewed as central for determining meaning or justifying belief, it loses its importance for the theory of knowledge and meaning. All these points will now form the central framework through which Davidson criticizes Quine’s view, the proximal-distal analysis, and which furthers his approach to semantic externalism as providing the proper basis for both meaning and knowledge. The exchanges we have seen from the conversations show Davidson’s increasing focus on Quine’s use of stimulus meaning as a key catalyst in his further explicit turn to this proximal-distal analysis and his externalist view of knowledge.9
4 Davidson’s Critical Framework: The Proximal-Distal Analysis and Triangulation
Central for Davidson’s later development are the two semantic views we have seen him find in Quine’s reflections on radical translation. This viewpoint plays an important role in the conversations and it continues to be revisited in his later work on Quine’s epistemology. So much so, that it is fair, I think, to describe it as the general interpretive framework that informs Davidson’s later critical discussion of Quine’s philosophy. Evidence of this is clearly on display in his ‘Meaning, Truth and Evidence’ (1990b) which represents perhaps his most sustained criticism of Quine’s naturalized epistemology. This essay highlights the change in Davidson’s critical approach to Quine that we have witnessed in the conversations.
Rather than focus on the third dogma and the problem of ‘ultimate evidence’, Davidson here focuses more specifically on how Quine’s use of stimulus meaning attempts to ‘tie meaning and content to the firings of sensory nerves’ (1990b, 68). He begins by accepting Quine’s general characterization of empiricism as containing the following two cardinal tenets: ‘… whatever evidence there is for science is sensory evidence … [and] all inculcation of meanings of words must rest ultimately on sensory evidence’ (Quine 1969, 75) and then undertakes an exploration into what consequences such a view has for the determination of semantic content.
Two further points are made that are importantly connected to the shift in critical perspective that was noted in the conversations and which are, arguably, influenced by them. First, he explains how he will not misread Quine’s use of the word ‘sensory’ in the above passage as something psychological and subjective, but as simply noting the sensory source of our information about the outside world. And, second, he mentions how despite various passages where Quine talks of data and evidence, it seems that ‘evidence’ is not a concept basic to Quine’s theory of evidence. He then concludes, ‘… it may be nothing in Quine’s official doctrine quite plays the role of evidence or ultimate data’ (Davidson 1990b, 69). This admission on Davidson’s part reflects, I suggest, what he learnt from the 1986 conversations, where these exchanges clarify Quine’s use of evidence as meaning ‘information’ as in the activation of nerve endings.10 While Davidson never fully rejects his earlier critical attribution of the third dogma to Quine’s epistemology with its further search for ‘ultimate evidence’, their conversations have convinced him that the old empiricist search for evidence does not so easily apply to Quine’s naturalized epistemology.11 He appears willing to place less importance on these earlier criticisms and now turns to what he increasingly recognizes as the core problem with Quine’s lingering empiricist approach, namely, his use of stimulus meaning.
Davidson wants to know exactly ‘… how sensory stimulations determine the meaning-the content-of observation sentences’ (1990b, 71). In pursing this question he discerns two different positions in Quine’s writings, exactly as we have seen them described in the conversations, here called the ‘proximal’ and ‘distal’ theories of meaning and evidence respectively. The proximal theory takes sameness of meaning for observation sentences to be found with the sameness of patterns of neural input. This view is suggested by Quine’s declaration that sensory input serves as the evidence for observation sentences and in this way serves as a basis for radical translation. With the proximal theory, meaning and evidence are tied to the proximal stimulus, our neural input, and observation sentences are taken to be semantically equivalent if they have the same stimulus meaning, that is, if the same patterns of stimulation prompt assent and dissent to the sentence. However, in addition to this proximal theory, Davidson also detects a distal theory in Quine’s remarks, which takes sameness of meaning as defined in terms of shared public objects or events. On the distal account, meaning and evidence are connected to distal objects in the world, so that sameness of meaning is the result of interpreters and speakers correlating their responses to their mutually shared causes in that world. By using stimulus meaning as a way to tie meaning to evidence, Quine has made evidence primary for his account of meaning, and this concept of ultimate evidence unavoidably leads him to the intolerable consequences seen with the proximal account. The key issue here concerns the way the proximal view ties meaning to sensory evidence and as a result, fails to locate the proper source of content and successful communication.
As we have seen from both ‘Coherence’ and the conversations, Davidson himself accepts the distal account of meaning and evidence, in line with his continued commitment to external objects as the cause and content of our utterances. Later articles discussing Quine’s view make many of the same critical points and continue to urge the general superiority of the distal account for the theory of meaning and knowledge (1994, 2003). In ‘Externalisms’, Davidson mentions how despite modifications in Quine’s views on meaning, brought about in part by Davidson’s critical remarks, his epistemology ‘remains resolutely individualistic’ (2001b, 7). This results in a tension or what Davidson calls an ambivalence in Quine’s philosophy, since his epistemology continues to be based on private sensory evidence, while his theory of meaning takes advantage of the social, externalist resources that help fix the content of our sentences. This general line is repeated in Davidson’s reply to Quine’s ‘Where do we Disagree?’ (1999). In order to recognize the central role the proximal-distal analysis plays in his ongoing criticisms of Quine it will be worthwhile to take a brief look at this reply.
Here, Davidson reconsiders Quine’s commitment to empiricism, and wonders if it creates a remaining difference between them over epistemology. The central problem surrounds Quine’s continued insistence that the proximal stimulus is crucial for his naturalized epistemology. Davidson remains puzzled, and wonders ‘what role the proximal stimulus plays in epistemology or in determining the content of observation sentences’ (1999, 83). No one can reasonably doubt that our sensory organs play a crucial role in perception, and that what happens at our sensory surfaces is part of a causal chain from perceived objects to beliefs. Just as in the conversations, Davidson continues to wonder why Quine takes such empirical facts as having any epistemological significance.
In attempting to uncover why the proximal stimulus remains of central epistemic importance for Quine, Davidson focuses on the notion of perceptual similarity, which serves as the foundation of all learning and our current science. He wonders how the triggering of receptors could be related to this perceptual similarity, noting the difficulties in taking Quine’s statements that we ‘infer’ science from neural input literally. Davidson continues, ‘In spite of a heroic effort on Quine’s part, there seems to me no interesting way to construct, from patterns of neural firings and behavioral responses, anything like the propositional contents of observation sentences’ (1999, 84). If, as Davidson insists, it is a necessary feature of using words that the content of sentences is connected to external features of our local surroundings then a successful theory of content must appeal to such external connections.
Just as he discusses in the conversations, Davidson finds two accounts of content in Quine’s philosophy, a form of semantic externalism that emerges with his emphasis on the distal scene within translation, and a subjective internalism that remains a crucial aspect of his naturalized epistemology with its emphasis on the proximal stimulus. Davidson remains puzzled by the apparent tension between these accounts and concludes, ‘Quine does not, I think, regard these views as opposed tendencies, but as features of different projects. I do not see how to reconcile them’ (1999, 84). Davidson’s emphasis on the differences between the distal and the proximal provides the backdrop for his continuing puzzlement over Quine’s epistemology. By retaining the proximal stimulus as important for his epistemology, Davidson thinks Quine has rejected the type of semantic externalism that is needed to account for how the content of observation sentences is determined. His proximal-distal analysis is a plea for Quine to move away from his use of stimulus meaning in determining the content of utterances and to accept the tacit externalism also present in his view and which Davidson thinks provides a sounder basis for the theory of meaning (2003, 293).
As we have seen Davidson indicate in the conservations, the central implication of these remarks against Quine’s use of the proximal account and its replacement with the distal account, is the explicit removal of any important role for sensory stimulation within the study of knowledge and meaning: ‘The distal theory of meaning removes the sense organs and their immediate activities and manifestations, such as sensation and sensory stimulations, from central theoretical importance to meaning and knowledge’ (Davidson 1990b, 76). Davidson concludes that what is then relevant to the correct translation or interpretation of another’s sentences is not what occurs at our sensory surfaces, but rather the shared distal events witnessed by both interpreter and native. Our ability to use and learn language is dependent on the fact that the content of our sentences is from the very start, causally anchored to shared objects, events, and aspects of our local environment. If correct, the content of our sentences contains elements that are not, like our neural input, contingently related to the environment since the distal object that is part of the shared context of learning is required for the acquisition of meaning.12
I will return to this critical point in my concluding remarks, as it speaks directly to the question of the status of Davidson’s approach to epistemology, and the issue concerning whether it is, as he claims, consistent with naturalism or a naturalized approach to epistemology. Davidson’s view of naturalism offers what I view as a much weaker constraint on philosophical reflection then that seen with Quine. In addition, his rejection of the philosophical import of our sensory equipment comes with the real implication that empirical, scientific work is irrelevant to philosophical inquiry. It would seem that this is a conclusion that a naturalized conception of knowledge could not accept.
At this point, we should return to the question of what Davidson’s proximal-distal analysis suggests concerning the precise relationship between meaning and knowledge. The individualist perspective of Quine’s approach presents us as building up our understanding of the world through stimulation at our sensory surfaces, with the later option of learning to express this understanding through communication with others. Davidson, as we have seen, opts for his externalist alternative, which he thinks has a basis in Quine’s own reflections on translation. Here, the content of our beliefs and sentences is determined by events in the world that prompt them with our knowledge built on that basis. Thought, knowledge and meaning are all depicted as dependent on linguistic interaction with others (Davidson 1994, 192).
Davidson’s emphasis on the problems with Quine’s appeal to stimulus meaning, his favouring of the distal object over proximal stimulation and his further rejection of causal stimulation as relevant to the philosophical study of meaning and knowledge are all rooted, I have argued, in the Stanford conversations. Most important perhaps, is the explicit critical shift made in conversations concerning stimulus meaning, which leads to his renewed emphasis on semantic externalism as providing a sounder basis for epistemology. This results in a number of further issues that are treated in the later stages of Davidson’s philosophical development. Here, his introduction and use of the device known as ‘triangulation’ looms large.13 By appealing to the social and causal dynamics present in such triangular interactions between others and the world, Davidson attempts resolve a wide array of philosophical issues, including how thought depends on language and how thought is possible. We, however, are most concerned with the connection between externalism and triangulation, especially with its relevance for Davidson’s development of his own epistemological views. Here, the main interest is in the way triangulation helps Davidson clarify his externalist thesis in terms of locating the typical or common cause that determines the content of our utterances and thereby makes knowledge possible.
With the distal account, causes that are external to the speaker are of key importance for securing meaning. The radical interpreter has no choice but to take such causes as central to the process of interpretation. A lingering problem for this kind of semantic externalism is that it relies on the idea of the common cause of utterances. Davidson notes that there are too many possible candidates for this all-important common cause of an interpreter’s utterance (1990b, 60). The solution to this difficulty is that the relevant cause is the one that is salient for the speaker and interpreter, where salience is defined in terms of similarity of responses. Davidson states that, ‘What makes communication possible is the sharing, inherited and acquired, of similarity responses’ (1990b, 77). In such a way, the interpreter is able to correlate his own responses to those of the speaker by noticing the shared salient causes in the world of which they speak. These central points of Davidson’s distal theory are based in the basic elements of his further use of triangulation. It is only once another individual enters the picture that the common cause of their shared utterance can be determined, thus fixing the content of the utterance and making possible interpretation.
These ideas are further outlined in Davidson’s summary essay ‘The Conditions of Thought’ where triangulation plays as a central role in Davidson’s externalism (1989a). He begins by noting his commitment to externalism; ‘… in the simplest cases the events and objects that cause a belief also determine the contents of that belief’ (1989a, 195). However, we have the familiar problem of locating the relevant stimulus that causes a belief and confers content. Davidson addresses the question by considering the situation of a child who is learning a language. An important part of such a process involves the child noticing certain objects as relevantly similar to others; for example, the child finds tables similar. Here the general problem of locating the appropriate stimulus is made explicit, since from our external vantage point there is no reason to pick tables as the stimulus as it could be patterns of stimulation at the child’s surface that prompt his reactions. Davidson explains how the solution to this difficulty involves noting that it is simply natural for us to pick the table as stimulus. That is, our discriminatory abilities are such that they tend to focus on objects in the world as the relevant stimulus. The situation now before us contains three classes of objects found similar by the child, and by the parent. The parent finds tables similar, the child also finds tables similar and the parent recognizes the child’s responses to tables as similar. These three patterns of response pinpoint the relevant stimuli of the child’s responses and this interaction emerges as a form of triangulation with a line going from child to table from parent to table and between parent and child. With triangulation, we can now make sense of a stimulus having an objective location in shared space, since the convergence of two individual perspectives pinpoint the relevant location of the stimulus.
Moreover, Davidson thinks that this triangular arrangement constitutes a necessary condition for the existence of conceptualization, thought and language. This is seen by noting that in isolation there can be no question of what a creature thinks or means since there is no way to pinpoint the object that gives content to a creature’s utterances to thoughts. It takes at least two individuals sharing reactions to common stimuli in order to determine the location of the object that prompts these reactions. One significant conclusion from this investigation is then this. Without another person with which to triangulate, an individual’s thought has no content; that is to say, there is nothing recognizable as thought. This is, of course, a direct consequence of Davidson’s externalism since it is the external stimulus that confers content on our perceptual beliefs and if there is no way of locating the relevant stimulus then there is no way to detect anything recognizable as thought.14
It is, then, linguistic transactions concerning what is going on in the world that pinpoints the content of our thoughts and thereby provides, in a much fuller sense, the idea that we are creatures having thoughts about an objective world independent of those thoughts. Davidson claims that our grasp of the concept of objectivity then emerges through the recognition, provided by linguistic communication, of another individual sharing thoughts about objects whose properties exist independently of those thoughts. Causal connections are describable in intentional terms, thereby opening up space for error, only when causal lines converge by means of social interaction, and only once the responses to such a convergence are mutually salient to the creatures involved in this interaction. The intentional status of the thoughts of a speaker, which emerge through the process of linguistic communication, are thus dependent on the causal and social dynamics present in triangulation.
Davidson’s later philosophical development then includes an ambitious extension of his basic externalism into a wide array of topics in language, mind and knowledge. The rejection of Quine’s emphasis on stimulus meaning provides a key turning point in Davidson’s utter dismissal of first-person empiricist approaches to knowledge in favor of the externalist third person perspective that anchors the content of thought and language directly to the world they are about. Triangulation adds a central social dimension to Davidson’s externalism, highlighting further how our interactions with a shared world provide the source of objective propositional thought and ultimately empirical knowledge.
5 Conclusions
From the account sketched above, we can see that the Stanford conversations played an important formative role in Davidson’s later critical approach to Quine’s work. These interactions also appeared to help focus his own further efforts to articulate his externalist commitments and their further implications for his own epistemology. By turning his critical attention on Quine’s use of stimulus meaning, Davidson is able to develop the proximal-distal analysis that highlights his debt to Quine while moving in his own externalist direction. A central consequence of this critical perspective is that proximal sensory stimulation is of no relevance for the theory of knowledge. This further results in Davidson’s complete break with first person empiricist epistemology in favor of third person knowledge as the basis for knowledge in general. Here the device known as triangulation not only helps to locate the common cause of the content of thought but highlights the irreducible social element to all forms of empirical knowledge. In these concluding remarks, using comments found in the conversations, I would like to wonder further about the status of Davidson’s approach to knowledge, especially in terms of its relationship to Quine’s naturalized epistemology.
We can start by noting that Davidson often portrays his criticisms and resulting externalist epistemology as consistent with Quine’s own naturalism or naturalized approach to epistemology. In laying out his alternative distal view in contrast to Quine’s mistaken emphasis on the proximal stimulus, we saw Davidson claim that he too is a naturalist with his suggestion that he starts from within the largely correct view provided by science. In another passage from the conversations, he explains how his externalism can be seen as a form of naturalized epistemology despite its rejection of the epistemological import of the senses:
Using this method there’s always going to be a presumption that with respect to the simplest things we’re both in agreement and furthermore we’re both right … So far as naturalized epistemology is concerned, explaining how it is that people come to think the things they do and work up this theory of the world and so forth, I can do that without the notion of evidence because I’m working with these causal relations between the world and the person’s attitudes and responses, as is Van. So in the causal chain I don’t have something that I’m going to give a special epistemological significance to, call it perceptions, call it patterns of stimulations (those are different but call it any of those things). I am actively concerned to eliminate what I call epistemological intermediaries, not causal intermediaries. Of course the sense organs and all sorts of stuff are in between. That I do see as the big difference. (Marschall 2026, 241)
With his later introduction of the proximal-distal analysis he, once again, explains that favoring the distal over the proximal in no way goes against Quine’s naturalist commitment (Davidson 1990b, 48).
Another telling passage is given in his ‘The Problem of Objectivity’ when he describes his own philosophical starting point in the following way: ‘This attitude and method have sometimes been called naturalism because naturalism starts by accepting common sense (or science) and then goes on to ask for a description of the nature and origins of such knowledge’ (1995, 5). Here, once again, Davidson presents naturalism as beginning from a general acceptance of what science tells us about the world and then attempts to show how such empirical knowledge is possible.
In carrying out this epistemological project, he argues that the appeal to sensory stimulation is unhelpful. This returns us to Davidson’s key critical point against Quine’s use of stimulus meaning: ‘Although sensation plays a crucial role in the causal process that connects beliefs with the world, it is a mistake to think it plays an epistemological role in determining the contents of those beliefs’ (1989a, 46). For Quine this denial that sensory stimulation has any role in the theory of knowledge amounts to a rejection of a naturalized approach to knowledge. Still, Quine’s view is not the sole form a naturalized conception of knowledge may take, so perhaps there remains room for aligning Davidson’s externalism with naturalism. However, there is another broader implication of Davidson’s criticism, even if it is directed solely at Quine, one that has not been given sufficient attention by Davidson or others. It appears that in rejecting sensation, or our sensory equipment, he appears to be also questioning the idea that empirical scientific work is relevant for the study of human knowledge. Empirical studies concerning sensory reception, perception, and neurophysiology while, of course, legitimate scientific projects offer findings and results that are not directly relevant for the philosophical study of knowledge. If this is a correct implication of Davidson’s criticism, then it amounts to a more sweeping rejection of Quinean inspired projects in naturalized epistemology.
Given such a rejection, why does Davidson present his view as consistent with a naturalist outlook? There appear to be two basic understandings of naturalism at work within the Quine-Davidson debate over epistemology. Quine understands naturalism as both encouraging and motivating the use of scientific work within philosophy. Here, empirical science is brought to bear on philosophical issues, further meaning that philosophy must defer to the sciences for the central empirically support results that bear on the solutions to its problems. Davidson appears to view naturalism as a more general constraint on philosophical theorizing, where accepting the broad truth that science provides us, we further develop an account of knowledge that shows how we could have achieved such knowledge. We work within a scientific view of the world but develop a philosophical project aimed at describing how we have arrived at this scientific view without any need to appeal to scientific, empirical work. As a result, in Davidson’s hands, naturalism becomes a much weaker constraint on philosophical inquiry. Science sets the background but does figure prominently in the philosophical solution to the problems of knowledge. Davidson’s own externalist epistemology is elaborated in precisely these terms, where he begins by accepting that we have a mostly correct view of our local environment and its inhabitants, and then proceed to provide a description of the nature of that knowledge.15 I suggest that when Davidson sees his externalism as consistent with naturalism, it is this conception of the naturalist view that he is using.
There is another exchange in the conversations that bears on these issues:
Q: Well that’s not clear to me because this seems so much like my notion of evidence. And these sentences that these people tend to agree on immediately, the way you say, when this common cause is out there affecting them… these seem to me like my observation sentences except for this matter of not grouping them according to what looks like a rabbit but grouping them according to what is a rabbit.
Da: But that’s a big difference because one ties it to a very old general idea of empiricism or an idea of something quasi-subjective. It’s not that you’re using introspection. But you’re making the epistemology start with something very close in.
Q: That’s quite right. That’s my substitute for phenomena.
Da: Now, I am saying of course that those things are in the causal chain, but I’m not attaching any epistemological significance to them except subsequently, that is, after the theory gets going. I’ve guessed as well as I can on the basis of the evidence that you’re talking about books or rabbits. Then if it works very well for a whole lot of things I try to take care of the cases where you think there’s a rabbit and I don’t by noticing things about the causal chain that connects us and make corrections. But the point is I don’t begin with what’s here and build out. I begin out there and build in.
Q: Yes, I think that’s the switch. I have to make that adjustment too, in reverse. I do that when I get into analytical hypotheses, and then I go back and end up with, for instance the term “rabbit” or the sentence “There’s a rabbit” as being true just in case it’s not a counterfeit.
Da: Right. There are, so to speak, corresponding problems seeing from this point of view. But it makes a big difference to your epistemology because one can be described not exactly as phenomenalism but it has an element like that, that is, it starts out with something very close in. Mine is naturalistic; it’s physicalistic, but it starts out there and builds in the air later. And the result is that all kinds of skeptical questions that naturally arise, not just with Quine’s view but with any view that starts with something in here… you can’t even raise those questions.
Q: People say of my naturalized epistemology so-called, “He called it epistemology. That’s not epistemology.” Now I say of yours, “You call it epistemology. That’s not epistemology.”
Da: Good. I agree. Absolutely. It’s a naturalized theory of knowledge, but old fashioned epistemology has dropped. (Marschall 2026, 244–45)
This last exchange is important for what it suggests about the radical nature of both Davidson’s and Quine’s epistemological proposals and it further points to a divergence that questions Davidson’s acquiescence in naturalism or naturalized epistemology. We have already seen the basic difference between their respective views mentioned in this discussion. Here Davidson claims that Quine’s proximal view is similar to classical empiricism with this focus on individual phenomenal experience. Davidson prefers to start with physical objects in the local environment and reconstructs first person knowledge on that basis. This he considers to be naturalistic, where we should read him as I sketched above, accepting our basic picture of the physical world described by science. Quine mentions that he is often accused of not doing epistemology, since many have claimed that he has eliminated any concern with epistemic justification in favor of a causal account from stimulus to science. But here he makes the further interesting point that Davidson’s project is not epistemology either since he rejects the phenomenalist starting point of empiricism. Davidson agrees with this interpretation claiming that he too offers a naturalized epistemology that departs from the tradition.
This suggests that both Davidson and Quine offer revolutionary approaches to the study of knowledge. Quine remains an empiricist, but this view is naturalized, with the key result being that he no longer seeks the sort of empirical foundation for knowledge that Davidson’s third dogma criticizes. He still operates within an empiricist outlook, but in dropping experience in favor of sensory stimulation, his naturalized epistemology is concerned with giving a scientific explanation of how humans have developed scientific theories. Davidson also breaks with empiricism, and we saw that he thinks this too makes him a naturalized epistemologist. However, he rejects the search for ultimate evidence within experience and argues that language and thought are ostensively tied to the objective world from the start. The result is, as he says, a complete revision of a first person, Cartesian, and empiricist viewpoint on human knowledge. But rather than a type of naturalized epistemology, conceived of as a scientific explanation of the empirical steps leading to knowledge, his inquiry looks more like a conceptual project aimed at establishing basic truths about the interconnections between thought, language and our knowledge of the external world. I have suggested that this is a project that departs from the scientific explanatory aims of Quine’s naturalized epistemology.
Davidson’s critical focus on stimulus meaning in the conversations helped to clarify his increasing divergence from Quine’s proximal view and further motivates his externalist approach to epistemology. This is illustrated in the last exchange we considered, where his externalism represents a fundamental departure from traditional empiricism. It also highlights the degree to which his epistemological interests diverge from Quine’s approach to knowledge. The Stanford conservations then serve as an important catalyst in creating this separation between Davidson and Quine. They further suggest that Davidson’s third person epistemology represents a more fundamental break with Quine’s naturalism than he fully recognized or was willing to accept.
Acknowledgements
I would like to thank the editors, Benjamin Marschall and Andreas Føllesdal, for their work on this special volume. I am especially grateful for their patience. This work was supported by JSPS KAKENHI Grant number JP24K03434.
In that article Davidson’s chief target is conceptual relativism, which he thinks gains support by the division made between scheme and content. By rejecting the content side of this dichotomy in the form of an unprocessed source of evidence, Davidson thinks that we can also drop the entire distinction and thus render unintelligible conceptual relativism (Davidson 1984, xviii).↩︎
Elsewhere, Davidson mentions that it is this point that constitutes his central argument against empiricism, which is why it is my main focus here (2001a, 286). I thus pass over Davidson’s further worries about the skeptical consequences of Quine’s adherence to the third dogma.↩︎
In his ‘Afterthoughts, 1987’, Davidson explains ’If it had not already been published, I would now change the title of “A Coherence Theory”…’ (1990a, 134). He further remarks that his emphasis on coherence was to make a negative point, that all that counts as the justification for a belief must be other beliefs. As we will see, the crucial point of the article argues for the claim that belief is in its nature veridical. It is for these reasons that Davidson concludes, ‘My emphasis on coherence was misplaced; calling my view a “theory” was a plain blunder’ (1990a, 135). He expresses similar misgivings about this paper in the conversations (Marschall 2026, 210).↩︎
In Davidson (2001a), he puts the point this way: ‘… the central argument for rejecting empiricisms that base knowledge on something unconceptualized is that nothing that is unconceptualized can serve as a reason for an empirical belief … since the relation of a reason to what it supports is conceptual and so demands that the reason have a propositional content’ (2001a, 286).↩︎
For a defence of Quine’s view, one based in Quine’s own response to Davidson’s criticism, see Sinclair (2007).↩︎
He develops this account in Davidson (1984). For related discussion and a defense of his overall approach see Sinclair (2002).↩︎
Davidson gives this definition of externalism: ‘a theory or position is externalist if it entails that a person’s beliefs and what he means by what he says are not completely determined by the physical state of his brain’ (2003, 292). As we will continue to see his externalism then takes distal objects to determine the content of our utterances, and he further argues that this view follows from Quine’s reflections on translation. Also see his comments about the importance of externalism for his view in Davidson (2001a, 293–94).↩︎
On the importance of this addition to Davidson’s view see Verheggen (2017).↩︎
Davidson explains that he had long held suspicions about Quine’s use of stimulus meaning and his lingering empiricism well before their 1986 conversations (1994). However, his remarks further suggest that it was only in the conversations that he was able to first fully articulate his critical attitude to stimulus meaning. This transition in his critical focus then results in his further proximal-distal analysis.↩︎
See the exchange between Davidson, Quine and Dreben in the conservations, where Quine clarifies his use of evidence in terms of ‘information’ (Marschall 2026, 319–20). Later in explaining his attitude towards evidence Quine offers this clarification: ‘My stated overall problem has been the quasi-epistemological problem, within natural science, of man’s construction of natural science on the datum base of neural intake. The intake is not what we are aware of and infer from, but it does encompass our “information,” in the computer engineer’s sense, as to what is going on around us. It is perhaps a candidate for the title of evidence, but it does not meet Davidson’s dictum that only a belief can be evidence for a belief. The runner-up for the title is the observation sentence that has been conditioned to that neural intake. Evidence in this sense meets Davidson’s condition’ (1997, 575–76).↩︎
Quine agrees with Davidson that in his theory of evidence ‘evidence’ is not clarified and plays no role (Quine 1990). These considerations motivate his discussion at the start of Pursuit of Truth when he claims we can examine the evidential support of science without appealing to ‘evidence’ as a technical term (1992, 2). In a later response, Quine notes that Davidson’s critical remarks on his use of evidence led him to ‘fight shy of the word’ (1994, 502).↩︎
On this point consider his further remark: ‘It is an empirical accident that our ears, eyes, taste buds, and tactile and olfactory organs play a causal role in the formation of beliefs about the world’ (1989b, 45–46). He goes on to mention that the causal connections fixed through sensation could be established in other ways without it making any difference to the content of beliefs or to the veridical nature of our beliefs.↩︎
This idea is first introduced in Davidson (1982b) and becomes a major theme in Davidson’s later work as seen in Davidson (2001c). For a key overview and defense see Verheggen (2013). Two contrasting interpretations of triangulation are given in Fergestad and Ramberg (2011) and Sinclair (2005). Central criticisms of this idea are found in Bridges (2006) and Pagin (2001).↩︎
For further discussion concerning the problems associated with locating the common or usual cause, see Verheggen (2017).↩︎
Davidson outlines his own method and philosophical starting point in such terms both in the conversations and more explicitly in Davidson (1995).↩︎