By Ryan Simonelli

Winner of the Gerritt and Edith Schipper Undergraduate Award for Outstanding Undergraduate Paper at the

59th Annual Meeting of the Florida Philosophical Association

Ryan Simonelli, New College of Florida

I. Introduction

Donald Davidson argues that the very nature of belief ensures that, if we have any beliefs at all, most of them must be true. He takes this to show that the traditional Cartesian sort of epistemological skepticism is fundamentally mistaken. Many find this response to skepticism to be lacking. Richard Rorty claims that this line of reasoning does not amount to answering the skeptic, but rather it amounts to “preventing him from asking his question.” 1 Similarly, Henry Jackman remarks that Davidson’s argument cannot be conceived of as a “straight” answer to the skeptic, “as a proof that we do, in fact, have true beliefs about the world around us,” and Barry Stroud insists that it can only function “to block a familiar route to skepticism, not to show that skepticism is false.” 2 These point to the claim that, rather than actually refuting the skeptic, Davidson is simply “telling him to get lost.” 3

While the notion that Davidson is not actually answering the skeptic has held considerable influence in the literature surrounding his arguments, I will argue that Davidson’s line of thought— once extended—is able to provide a legitimate and satisfying “straight” answer to the epistemological skeptic. I first show how we ought to construe Davidson’s argument transcendentally, and I lay out Barry Stroud’s challenge to this standard construal. In response to this challenge, I draw from recent work by Rebecca Kukla and Mark Lance and attempt to give Davidson’s argument a newfound force by applying it to our acts of ostension, of pointing others to features in our shared environment. I believe the sort of ostensive transcendental argument that I extract from Davidson is a promising response to epistemological skepticism that remains largely unexplored in skepticism’s vast literature. It also allows us to think of Davidson’s argument in a new light: as a direct realist response to skepticism. Reading Davidson this way is rather unorthodox, especially given his apparent acceptance of coherentism at some places in his writing, but I argue that this is the strongest way to read his epistemological work, providing a response to Stroud’s

objection to his arguments.

II. Davidson’s Arguments against External-World Skepticism

The external-world skeptic, or Cartesian skeptic as we will call him, argues that we cannot rule out the possibility that all or mostly all of our beliefs about the external world are false, and so we lack justification in believing that most of them are true. 4 In the standard case, the Cartesian skeptic makes this argument by putting forward some skeptical scenario in which we might be radically deceived about all of our beliefs such as the possibility that we are dreaming or brains in a vat. Since these scenarios are purportedly subjectively indistinguishable from a scenario in which we are not deceived, and it appears that most of our beliefs about the world are false in these scenarios, the possibility of these scenarios seems to imply the possibility of all our beliefs being false. Davidson argues, however, that this is not a genuine possibility; given the nature of belief, having massively false beliefs is impossible, and so external-world skepticism of this sort must be confused.

To get a grip on Davidson’s basic way of thinking about belief, consider the following scenario: I am shipwrecked on an uncharted island with an unknown native population. I stumble across one such native whose language is entirely foreign to me. To understand what a member of this population is saying and, correlatively, what his beliefs are, I must take an “interpretive stance” towards him and see how he responds to stimuli in his environment. This three-part relation between interpreter, speaker/believer, and shared stimulus is what Davidson calls “triangulation.” 5 To use Quine’s example, I might hold up a rabbit, and he might (presumably understanding that I am trying to learn his language and not, for example, trying to attack him with the rabbit), exclaim “Gavagai!” 6  In trying to understand the language he is speaking, I also will attribute beliefs to him. Suppose that I take ‘gavagai’ to mean rabbit. I then will attribute to the native the belief that a rabbit is present. I do this because this belief is appropriate, given that a rabbit is in fact present and clearly visible. In order to make sense of the native’s beliefs at all, I must attribute to him the beliefs that he ought to have, and in doing this, I must attribute mostly true beliefs to him. Generalizing this point, in order to understand anyone as having beliefs, I must interpret their belief states charitably, as being mostly true.

Davidson extends this insight from radical interpretation to language learning where it takes its full force in highlighting the connection between the cause of an utterance or a belief and its meaning. On Davidson’s account of meaning, meanings of the most basic kind correlate directly with the objects that cause them. He writes, “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.” 7 When learning a language, a child learns to say certain words and sentences in the situations for which they are appropriate. For simple object words, some of the first words learned by a child (and the sorts of words we can associate with empirical beliefs about one’s immediate environment), the appropriate situations in which they are to be uttered are when the object is present; in the paradigm cases, object words are to be spoken as a response to objects. Though this is a rough and oversimplified picture as Davidson admits, a child learns the word “rabbit,” for example, by positive reinforcement whenever a rabbit is present and negative reinforcement when one is not. The same sort of triangulation that occurs in radical interpretation also occurs in language learning. There is a three-way interaction between a language- learner, a language-teacher, and a stimulus in the world. Learning a language consists in learning how to use words appropriately. Accordingly, to know the meaning of “rabbit,” one must understand the appropriateness of saying the word as a response to rabbits. 8

Davidson is by no means modest about the epistemological implications of his considerations regarding belief and meaning. He concludes, “The fallout from these considerations for the theory of knowledge is revolutionary. If words and thoughts are, in the most basic cases, necessarily about the sorts of objects and events that commonly cause them, there is no room for Cartesian doubts about the independent existence of such objects and events.” 9 The idea here is relatively straightforward. The concepts in light of which my beliefs and thoughts make sense arise in a social context that essentially involves other language users and shared objects and events to which language-speakers are responsive. In order for us to have beliefs about rabbits, and for it to make sense to talk about these beliefs, there must really be rabbits. More so, these rabbits must really be the things causally responsible for our beliefs about rabbits. 10

Both the argument from interpretation and language-learning rest crucially on the idea that triangulation is essential for making sense of intentional states, whether this is interpreting the intentional states of others or learning how to undertake the appropriate intentional states for oneself. Without shared awareness of the objects in the world causally responsible for our beliefs, there is no way of evaluating the appropriateness of intentional states, and without the possibility of this evaluation, there can be no ascription of intentional states at all. The argument from interpretation shows that we must interpret any intentional agent as having largely appropriate beliefs, and the argument from the causal connections shows that these beliefs have to be, in the paradigm cases, about the objects that cause them. Together, these arguments purport to show that any believer at all must have mostly true beliefs about the objects and events in the world, and thus the epistemological skeptic who claims that all of these beliefs might be false is mistaken.

III. The Transcendental Construal and Stroud’s Objection

While it is clear that Davidson’s arguments, if correct, show that our theory of belief and language will exclude global skepticism as a serious theoretical possibility, many doubt that these arguments suffice as a direct refutation of skepticism. The first intuitive objection is that Davidson’s arguments seem to rely on broadly empirical observations about language learning and linguistic practice, and thus, they already assume that we are largely right about the way the world is. According to this objection, relying on the way language actually works to argue that most of our beliefs must be true is to take too much for granted; it assumes the very conclusion for which it is attempting to argue. In short, the worry is that Davidson begs the question. Michael Williams puts it as follows:

The appeal to charity turns out to involve the idea of unproblematic access to certain causal relations between speakers and objects in the world. If, in the context of the sceptic’s question, we grant ourselves this access, the game is over before it begins. 11

It is easy to see the concern raised here. We cannot simply assume the very thing the skeptic is calling into question in our attempt to answer the skeptic. Traditionally, the way of satisfying this criterion of adequacy was to appeal to an a priori foundation for knowledge—something the skeptic can’t possibly call into question—and infer outward, from. However, it is clear that Davidson does not want to pursue this strategy. He claims, “Empirical knowledge has no epistemological foundation, and needs none.” 12

Still, just because Davidson’s arguments are not foundationalist ones, it does not mean that they problematically assume empirical knowledge of the world. One popular way of reading Davidson is to see his arguments as transcendental ones that, rather than providing an epistemological foundation from which we can infer the existence of the objective world, entirely erase the need for such a foundation. 13  This clearly seems to be what Davidson is up to when he says, for example:

If one can show, as I think is possible, that in order to have a thought, even a doubt, one must already know that there are other minds and an environment we share with them, then this amounts to saying that it is impossible seriously to doubt these things—we cannot give a coherent content to such doubts. 14

We should not think of this as some sort of Cartesian strategy involving an inference from inner certainty outward to the world, since, if Davidson really thinks that empirical knowledge has no foundation, no such strategy is possible. Rather, we should think of the move as a fundamentally elucidatory one. That is, once we lay out what the concepts of thought and belief really mean and illuminate the sorts of things that must be in play for these concepts to even make sense, we understand that there must be an objective world about which we are mostly right. Conceived in this way, it is clear that Williams’ worry is misplaced. Davidson does not assume that we have “unproblematic access to certain causal relations between speakers and objects in the world,” but rather he aims to show that this access is necessary to even consider the skeptic’s questioning of our beliefs.

The goal of Davidson’s transcendental argument is to show that a coherent conception of belief necessarily involves triangulation. If triangulation obtains and we are mutually forming our own beliefs and evaluating each other’s beliefs in response to the same set of objects, then external- world skepticism is a non-starter. On Davidson’s view, all three points of triangulation must rise or fall together. Thus, we can employ the transcendental argument to start at the most uncontroversial point of triangulation (our own beliefs) and end up with beliefs about other agents and the environment we share. The possibility of this subjective starting point has led some philosophers like Thomas Nagel to liken Davidson’s argument to Descartes’s cogito, but when laid out in this way it becomes clear that this starting point is merely rhetorical, and the subjective side of things has no actual epistemological priority. 15

As I will construe it here, Davidson’s basic transcendental argument has two essential premises:

1.) Havingbeliefsonlymakessenseifthosebeliefscanbeintersubjectivelyinterpretedand evaluated for correctness.

2.) This sort of interpretation is only possible if the beliefs are systematically about the objects that cause them.

Davidson inherits the first point from the later Wittgenstein, and we can think of it in terms of the normative nature of belief and understanding. 16  In order to have a concept of having a belief, one must also have a concept of believing correctly or incorrectly, and the only way one can have this concept if they have an understanding of the intersubjective evaluation to which their belief could be subjected. This Wittgensteinian point is not entirely uncontroversial, but it is relatively mainstream, and I won’t give an extended argument for it here. 17 The second point is the crucial one that provides the main thrust of Davidson’s transcendental response to skepticism. If our beliefs can be

interpreted at all, they must be, at least for the most part, world-involving. That is, we cannot make sense of ourselves as having beliefs at all if we do not also take into account the objects in the world that causally constrain them. Without causal constraint, interpretation is impossible. This constraint, Davidson believes, constitutes the objective dimension of our epistemic practices—the crucial third point of triangulation.

I think this transcendental construal is the strongest way to frame Davidson’s argument. Still, even when construed in this fashion, Barry Stroud argues that Davidson’s argument is not sufficient to show that we must have mostly true beliefs. He claims that the only conclusion we can draw from this argument is that any interpreter must interpret a speaker’s beliefs as mostly in accord with his own beliefs. This would give us the truth of the conditional, “If an interpretation is carried out by interpreters whose beliefs about the world are mostly true, then the attributed beliefs must be mostly true,” but it would not give us the consequent alone, that any believer must have mostly true beliefs, and that is what Davidson claims to show. 18  Davidson seems to think that the second premise of his transcendental argument, the causal premise, gives us this consequent. According to Stroud, however, employing this premise gets us no further. Stroud agrees (at least for the sake of argument) that mutual interpretation is required for the possibility of belief attribution, and that this requires believers thinking each other’s beliefs are caused by the same objects, but he claims we are left without certainty that objects we take to be causing our beliefs really are the objects that cause them. 19

One way of putting Stroud’s criticism is that he accepts the necessary connection between the first two points of triangulation (subjectivity and intersubjectivity), but resists the move to the third point of triangulation (objectivity). Stroud takes Davidson’s account of meaning to imply that, in order to have the concept of rabbit, we must interpret each other as having beliefs that are causally constrained by rabbits. Still, this does not necessarily imply that our beliefs really are causally constrained by rabbits. We can be wrong about what we take to be systematically causing our beliefs, so long as we are wrong together.

I think this is a serious and powerful criticism, and Davidson’s argument, as it stands, does not have the resources to counter it. In what remains of this paper, I will attempt to show that genuine epistemic access to features of our environment really is necessary, and thus Davidson’s argument is successful even when read in the strong way, as a “straight” response to external-world skepticism.

If Davidson’s argument from triangulation is correct, then shared epistemic access to features of the environment must be necessary to understanding ourselves as having any beliefs at all. The most direct and powerful way of demonstrating this access, I believe, is by looking at our ostensive capacities, our practices of pointing out features of our environment to which we are mutually responsive. We can demonstrate the thrust of this strategy with the following exchange:

SKEPTIC: The external world does not exist!
BELIEVER: Ok, have a seat on that couch across from me and I’ll try to change your mind.
SKEPTIC: (Sits down.) So, let me hear your argument.
BELIEVER: (Holds out one hand in between them.) Here is a hand!
SKEPTIC: Oh, no! I won’t fall for that one! You’re simply begging the question. BELIVIER: Am I? So, I see my hand right now, and I presume yousee my hand as well. It’s right there in front of our faces.
SKEPTIC: No, but I’m saying that it could be the case that you don’t have any hands. BELIEVER: Ok, if you don’t want to call this thing that I am holding out a “hand,” that’s just fine with me. But “hand” seems like a perfectly suitable word for this thing, and that’s what we’ve been calling these things throughout the years. SKEPTIC: But that’s not an argument!
BELIEVER: Nope. It’s better than an argument. It’s a hand!

It seems that a response of this sort would be quite powerful against the skeptic. Still, someone with Stroud’s sympathies to the skeptical possibility will insist that the practical applicability of a response to skepticism does not ensure the correctness of that response, even if it makes the skeptical position practically impossible to assert. The skeptic might insist that it’s still possible that it only seems as if there is a hand to which I am ostending, but I am in fact mistaken. If that is possible, he might continue, it is possible that I am mistaken about all of my ostensive acts.

To respond to this apparent possibility and show that we really do have epistemic access to shared environment, a modified version of Davidson’s transcendental argument that shows the necessity of successful ostension is in order. This argument mirrors the structure of the basic Davidsonian transcendental argument, and runs as follows:

1a.) We cannot understand ourselvesas believers at all unless we also recognize other believers who interpret and evaluate our beliefs.
2a.) We cannot recognize other believers without simultaneously recognizing shared features of our environment to which we are mutually responsive.

I see (1a) and (2a) of this argument as corollaries of (1) and (2) of the basic transcendental argument. Now, consider first the relationship between (1) and (1a). If belief and understanding would only make sense within the context of the intersubjective interpretation and evaluation of beliefs, then, in order to understand myself as having beliefs, I must recognize others who can interpret and evaluate these beliefs. I must see myself as situated in what Wilfrid Sellars calls “the space of reasons” in which I am held to public standards of correctness in the formation of beliefs. 20 Locating myself in this space requires me to understand the possibility of other believers correcting my beliefs. Davidson’s arguments already give us this much, but, if we look closely at what is entailed by the possibility of this correction, we can see an interesting line of argument that Davidson seems to overlook.

If someone is to correct a belief of mine, a mutual recognition of each other as believers must take place. She must recognize me as a believer, and I must recognize her as recognizing me as a believer. Only then will I be responsive to any sort of belief-correction she might issue. In “Yo!” and “Lo!”, Rebecca Kukla and Mark Lance identify a distinctive sort of speech act that functions to initiate this mutual recognition. They call it a vocative, or a “Yo!” act. 21 To perform a “Yo!” act is to, on the one hand, express recognition of a person, and, on the other hand, initiate mutual recognition by instituting a normative relation that requires the person to whom the act is directed to reciprocate the act (to “Yo!” back). Kukla and Lance argue that without successful performance of these acts, intersubjective communication and correction of belief is impossible, since they “draw us into normative relations with one another and place us in social space.” 22  If Davidson is correct that intersubjective correction of belief is a requirement for belief is itself, then, in order to be a believer, I must be something that can be “Yo!”ed and which can “Yo!” back. Without these acts, the normative relations required for belief-correction cannot arise. Thus, we can conclude (1a), that the successful performance of “Yo!” acts, which recognize other believers, is necessary to our own status as believers.

Once we have established the necessity of “Yo!” acts, Kukla and Lance show that it is a short step to (2a). Another set of speech acts they introduce are “Lo!” acts, ostensive observatives. These acts express first-personal recognition of features of one’s environment and call others to do the same. These acts are crucial for the purposes of this anti-skeptical argument, since they “establish our responsiveness to a shared empirical world and make possible reasonable debate about that world.” 23 Kukla and Lance argue that “Lo!” acts are necessary for any communication at

all by tying them directly with “Yo!” acts.


ur interactions with others are embodied interactions,

and we cannot interact with one another unless we can recognize one another in an ostensive and

perceptual sense, where “recognizing” involves a “literal receptive encounter.” 24  Therefore, “Yo!”

acts, which we have seen are essential to recognizing oneself as a believer at all, simultaneously

function as “Lo!” acts. 25 We can only communicate with people whom we recognize as believers,

and we can only do this if we mutually recognize each other as embodied agents interacting with

objects in a world that we share with them.

In short, in order to be a believer, I must be at the very least able to successfully perform “Yo!” acts, to recognize other believers with whom I can communicate and who can interpret and evaluate my beliefs. Of course, it is possible that in some cases I “Yo!” to what I think is another person, but is in fact a cardboard cutout or something of the sort. However, the notion that I might have a global failure in the recognition of other believers is not possible, since it is only by way of this recognition that I can understand my beliefs as being interpreted, evaluated, and thus bound to standards of correctness in the first place. Since global failure in the recognition of other agents is not possible, global failure in recognizing objects in my environment is also impossible. This is because recognizing another believer requires interpreting them as such, and thus recognizing them as an embodied believer, recognizing and interacting with various objects in the environment that I myself recognize. To understand others as believers, we must understand them as having beliefs that are both causally and rationally constrained by objects that we are aware of ourselves. 26 Thus, shared awareness of these objects is necessary to understanding ourselves as rational agents, as beings with beliefs that can be evaluated for correctness.

Now that we have established a shared world to which we are mutually responsive in the formation of our beliefs, we can reemploy Davidson’s argument regarding the connection between causation and content with newfound force to say that the objects in response to which our beliefs are formed also must constitute the content of these beliefs. When we communicate about worldly matters, we are communicating about these ostensible objects, the objects by which we interpret and evaluate the beliefs of our fellow agents. In order to be believers at all, we must be mutually aware of objects in our environment, and it is these objects that our words and beliefs must be about. There is simply no other empirical content on which our beliefs could take their basis. In Davidson’s words, for an interpreter or language learner, “whatever is seen or thought to be ostended is the only content he has to work with, though further contexts can reinforce or modify this content. There is no crack here where the sceptic can drive a wedge between belief and reality.” 27  If we look at argument in this order, with the ostensive step as a necessary component, Stroud’s objection no longer has any force.

III. A Commitment to Direct Realism

I have reconstructed Davidson’s argument as a transcendental one and fashioned an ostensive version of it that I believe is immune to Stroud’s critique of the original argument. In this context, I believe we must see Davidson as committed to a form of direct realism regarding perception, even though he never explicitly endorses such a view. Direct realism, as I use it here, is the thesis that we are directly aware of objects in the world themselves, and, in standard cases, we form our perceptual beliefs in virtue of our awareness of these objects rather than awareness of some mental phenomena. Here, I mean “in virtue of” in an epistemological sense rather than a causal sense. Davidson states, “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.” 28  There are all sorts of causal intermediaries between the object of an utterance or belief and the utterance or belief itself (such as the set of light rays bouncing off an object, the stimulation of the eyeballs, the electrical signals going down the optic nerve, etc.). However, Davidson insists that the only causal connection of serious epistemologicalsignificance for us is that between the object itself and the belief about that object.

My insistence that Davidson is committed to a form of direct realism might seem strange to some readers since, at least in some places in Davidson’s writing, it seems as if he not only wants to restrict mental sensations from being the epistemic basis of our perceptual beliefs, but objects as well. In “A Coherence Theory of Truth and Knowledge,” he endorses coherentism, the epistemological view that “nothing can count as a reason for holding a belief except another belief.” 29 Such coherentism, according to John McDowell, leaves us with the threat of “spinning frictionlessly in the void.” 30  If Davidson actually endorsed this view, it would be rather mysterious, since the sort of “frictionlessness” that McDowell thinks coherentism permits seems wholly other to Davidson’s approach which insists on the primacy of triangulation. However, Davidson later rejected the use of the term “coherentism,” asserting “My emphasis on coherence was properly just a way of making a negative point that ‘all that counts as evidence or justification for a belief must come from the same totality of belief to which it belongs’.” 31  If this was Davidson’s main point, then, rather than saying, “nothing can count as a reason for holding a belief except another belief,” he ought to have said, “Anything’s counting for a reason for holding a belief can only be understood as having this normative significance within the context of our totality of beliefs.” If he had said the latter, it would not preclude objects from counting as reasons. It would be completely in-line with the Davidsonian point to say that, within the context our general set of beliefs about the world and our norms of belief-formation, objects themselves become reason-giving.

This is, in fact, the way that Kukla and Lance propose thinking of the reason-giving nature

of objects in the world that we perceive. They write,

The world can demand nothing of us and holds us to nothing, but our normative practices are, generally speaking, intersubjective and collaborative, and other people with whom we engage in practices are indeed agents who hold us to all sorts of things. Add to this the fact that, as we explored above, our practices are typically essentially object-involving, and we earn the result that other people ensure that worldly objects and events have specific normative significances for us that we cannot simply choose to ignore. 32

As an illustration of this point, consider the arcade game Whack-a-Mole. In this game, animatronic moles come out of holes, and the player’s objective is to strike them with a foam mallet as they come out. Of course, the player has a disposition towards whacking the moles, and the moles are causally responsible for this whacking as they come out of the holes. But there is also a norm at play. The player ought to whack the moles when they come out. This is because she is playing Whack-a- Mole. Her whacking of the moles is justified (in a non-epistemic sense) given the context in which she finds herself. The moles exert both a causal and normative force here. Of course, nothing intrinsic about the moles exerts this normative force; they only attain this normative significance through their contextualization in a game in which it is correct to whack them.

We can say the same sort of thing about how objects in the world provide a rational constraint for belief. Like the act of whacking in Whack-a-Mole, the general act of observing is an act couched in normative social practices. The suggestion then is this: given a certain shared understanding of what the world is like and what the standards of rationality are, certain norms for dealing rationally with our encounters with objects are put in place, so that we can say that objects provide rational constraint for our beliefs. On this account, I might be aware of a rabbit, but the rabbit itself is not what exerts the normative force on me to believe that a rabbit is present. Rather, this normative force is exerted by my particular location in a community with certain epistemic norms. There is of course, a difference between the Whack-A-Mole norms and perceptual norms. I might opt out of playing Whack-A-Mole, and so there would be no sense in which I would be bound by the norms of the game. But, given that abiding by perceptual norms is essential to our status as having any beliefs at all (as I argued in the previous section), we cannot opt out of this normative commitment in the same sort of way.

On this picture, our own understanding of things, the set of conceptual and discursive norms that we share, and the reason-giving nature of the world itself, are all inextricably intertwined. Once we look at things in this fashion, the conceptual and epistemological barrier between

subjective, intersubjective, and objective is completely broken down precisely in the way that Davidson demands. No conceptual or epistemic priority is granted to self-knowledge, knowledge of others, or knowledge of the external world. Without this barrier, there is no way for skeptical doubts to get off the ground.

IV. Conclusion

It is now clear how an argument from radical interpretation can make sense as a response to a skeptical concern. The very fact that interpretation is coherent with respect to our beliefs shows that we must be embodied within a world about which we have mostly true beliefs. And, if we reflect a bit, just look at us! We are embodied agents interacting with the world in various interpretable ways. Now that we have gone through how this sort of ostension to the world and ourselves is a necessary requisite for our ability to make sense of anything at all, it no longer simply begs the question against the skeptic. Thus, contrary to popular opinion, I believe that Davidson’s arguments, when framed in the way I have laid them out here, are able to provide a straight response to the external-world skeptic. 33


Brandom, Robert. “Perception and Rational Constraint.” Philosophical Issues 7 (1996): 241-259. Davidson, Donald. “Afterthoughts.” Subjective, Intersubjective Objective. Oxford: Oxford University

Press, 2001a. Pp. 154-157.

Davidson, Donald. “A Coherence Theory of Truth and Knowledge.” Subjective, Intersubjective Objective. Oxford: Oxford University Press, 2001b. Pp. 137-153.

Davidson, Donald. “Reply to Barry Stroud” The Philosophy of Donald Davidson. Ed. Lewis Edwin Hahn. Chicago: Open Court, 1999. Pp. 162-165.

Davidson, Donald. “The Myth of the Subjective.” Subjective, Intersubjective, Objective, pp. 39-52. Oxford: Oxford University Press, 2001c. Pp. 39-52.

Davidson, Donald. “The Problem of Objectivity.” Problems of Rationality. Oxford: Oxford University Press, 2004. Pp. 3-18.

Davidson, Donald. “Three Varieties of Knowledge.” Subjective, Intersubjective, Objective. Oxford: Oxford University Press, 2001d. Pp. 205-220.

Heath, Joseph. Following the Rules. Oxford: Oxford University Press, 2008.
Kukla, Rebecca and Lance, Mark. ‘Yo!’ and ‘Lo!’ The Pragmatic Topography of the Space of Reasons.

Cambridge: Harvard University Press, 2009

_______. “Intersubjectivity and Receptive Experience.” The Southern Journal of Philosophy 52.1 (2014): 22-42.

Maker, William. “Davidson’s Transcendental Arguments.” Philosophy and Phenomenological Research 51.2 (1991): 345-360.

McDowell, John. Mind and World. Cambridge: Harvard University Press, 1994.
Nagel, Thomas. “Davidson’s New Cogito.” The Philosophy of Donald Davidson. Ed. Lewis Edwin

Hahn. Chicago: Open Court Publishing, 1999. Pp. 195-205.

Quine, W.V.O. Word and Object. Cambridge: The MIT Press, 1960.

Rorty, Richard. “Pragmatism, Davidson, and Truth.” Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson. Ed. Earnest LePore. Oxford: Blackwell, 1986. Pp. 126-150.

Sosa, Ernest. “Knowledge of Self, Others and World.” Contemporary Philosophy in Focus: Donald Davidson. Ed. Kirk Ludwig. Cambridge: Cambridge University Press, 2003. Pp. 163-182.

Sellars, Wilfrid. Empiricism and the Philosophy of Mind. Cambridge: Harvard University Press, 1997.

Stroud, Barry. “Radical Interpretation and Philosophical Skepticism.” The Philosophy of Donald Davidson. Ed. Lewis Edwin Hahn. Chicago: Open Court Publishing, 1999. Pp. 139-161.

Williams, Michael. Unnatural Doubts: Epistemological Realism and the Basis of Skepticism. New York: Blackwell, 1991.

_______. “Responsibility and Reliability.” Philosophical Papers 37.1 (2008): 1-26.





  1. Rorty 1986, 344.
  2. See Jackman 1998 and Stroud 1999.
  3. These are the words Davidson uses to describe Rorty’s suggestion in Davidson 2001a, 154.
  4. It is important to note from the outset that the external-world skeptic, while still a global skeptic, is not the most radical skeptic we can conjure up since he does not doubt the very possibility of an anti- skeptical proof by doubting our epistemic norms generally. While Davidson’s arguments will not be wholly irrelevant in response to this more radical sort of skeptic, I believe a different sort of anti- skeptical approach might be required to deal with him sufficiently. For more on the distinction between skepticisms see Williams 2008.
  5. Davidson 2001d, 212.
  6. Quine 1960, 28.
  7. Davidson 2001b, 151.
  8. Of course, this is not to say that this responsiveness is all that is required to know the meaning of a word, only that it is a necessary requirement.
  9. Davidson 2001c, 45.
  10. At least in the paradigm cases, since, of course, a rabbit-shaped rock might sometimes be causally responsible for a belief about a rabbit.
  11. Williams 1991, 313.
  12. Davidson, “The Myth of the Subjective,” p. 46.
  13. Davidson himself rarely uses the term “transcendental” in reference to his argument, but is a relatively standard reading in the secondary literature. See Maker 1991 and Sosa 2003.
  14. Davidson 2004, 6.
  15. See Nagel 1999.
  16. Davidson makes this explicit connection to Wittgenstein as I do here in Davidson 2001d, 209.
  17. For a particularly thorough overview of this Wittgensteinian point and its place in philosophical history see Heath 2011.

  18. Stroud 1999, 147.

  19. Stroud 1999, 153.
  20. Sellars 1997.
  21. Kukla and Lance 2009, 138.
  22. Kukla and Lance 2009, 210.
  23. Kukla and Lance 2009, 210.
  24. Kukla and Lance 2009, 211.
  25. At least in the core cases, since there are circumstantial outliers in which this is not the case. Imagine giving a talk and thinking that there might be a hand of a person asking a question in the back of the room. You might say “Is that a hand back there?” calling out a person with a “Yo!” act if there is in fact a hand raised back there, but not committing yourself to an observative “Lo!” act. Cases like these, however, are deviations from the norm in which a “Yo!” also functions as a “Lo!”
  26. The language I use here, in speaking of perceptual awareness as a conjunction of causal and rational constraint, is drawn from Brandom 1996.
  27. Davidson 1999, 165.
  28. Davidson 2001c, 46.
  29. Davidson 2001b, 141.
  30. McDowell 1994, 11.
  31. Davidson 2001a, 155.
  32. Kukla and Lance 2014, 30.
  33. I would like to thank Aron Edidin, Rebecca Kukla, Jacob Berger, Bob Pasnau, and the two anonymous reviewers from the Florida Philosophical Review for their helpful feedback on earlier versions of this paper.  

Ryan Simonelli

Ryan Simonelli, works primarily in epistemology and philosophy of language. He graduated from 

New College of Florida in May, writing his undergraduate thesis on an approach to epistemological 

issues drawing from Robert Brandom and Donald Davidson. Ryan has also spent some time 

studying at Oxford University and University of Colorado at Boulder. He recently submitted his 

Ph.D. applications and is now just anxiously waiting to find out where he’ll spend the next several 

years of his life.