An Ecumenical Mooreanism

Abstract

The purpose of this paper is to get clear on how we should think about Mooreanism. I will argue that Mooreanism is best understood as a metaphilosophical response to skepticism rather than a particular position on specialized debates in first-order epistemology. This ecumenical understanding of Mooreanism implies that a broad array of epistemologists is free to be Moorean. In Sect. 2 I discuss several non-Moorean responses to skepticism. In Sect. 3 I provide an exposition of Mooreanism itself. In Sect. 4 I show that most epistemologists are free to be Mooreans. This is important for the following reason: to the extent that we want a non-concessive reply to the skeptic – rather than a reply that is partially or potentially concessive – we should be highly attracted to Mooreanism.

1 Introduction

The purpose of this paper is to get clear on how we should think about Mooreanism and to offer up for our consideration an ecumenical understanding of Mooreanism. Mooreanism, I think, is best understood as a metaphilosophical response to skepticism rather than a particular position on specialized debates in first-order epistemology. Thus, we should not identify Mooreanism with liberalism in the epistemology of perception, nor with the view that Moore’s proof of the external world doesn’t suffer from transmission failure. These very specialized, local conceptions of Mooreanism are problematic for two reasons. First, they fail to capture the commonsensist metaphilosophy at the heart of Moore’s response to radical skepticism. Second, they imply that it is very easy and relatively uncostly to reject Mooreanism. I offer here, in place of these understandings, a conception of Mooreanism as a commonsensist metaphilosophy. This ecumenical understanding of Mooreanism implies that a broad array of epistemologists is free to be Moorean. In Sect. 2 I discuss several non-Moorean responses to skepticism. In Sect. 3 and its various subsections I provide a lengthy exposition of Mooreanism itself.

The upshot of all the foregoing is twofold. First, we will gain a much clearer picture of just exactly what Mooreanism really is; second, we will see that very many philosophers, with very many different views, are free to be Mooreans. This latter point is philosophically important and not simply terminological. As we will see shortly, responses to radical skepticism can be divided into four categories: concessive, partially concessive, potentially concessive, and non-concessive, this latter being the Moorean response. Insofar as one wants a non-concessive response, and insofar as right thinking about skepticism demands such a response, Mooreanism emerges as the best, maybe the only, choiceworthy option on this menu. It’s not as though we have a bunch of good non-concessive responses to skepticism, Mooreanism being just one option among others. In this sense, most of us should be Mooreans, whether we want to or not. That is, we should all agree on the fundamental Moorean point that the controversial philosophical principles needed by the skeptic are rationally inferior to the truths of commonsense and thus should be rejected for that reason; having agreed on this, we are free to disagree about why the skeptic’s argument fails and what the proper requirements are for positive epistemic statuses like justification and knowledge.

2 Non-Moorean Responses to Skepticism

Arguments for radical skepticism have seemingly plausible premises but implausible conclusions. We may distinguish four sorts of replies to such skepticism: concessive, partially concessive, potentially concessive, non-concessive. The concessive reply is represented by skepticism itself, an option most of us won’t be very congenial to – most of us are quite partial to the idea that we know we have hands. I would wager that most of us function like Mooreans here, implicitly accepting the basic Moorean point that the controversial philosophical principles needed by the skeptic are rationally inferior to the truths of commonsense and thus should be rejected for that reason.

Three examples of partially concessive replies to skepticism are denying closure, contextualism, and contrastivism. One premise in the brain-in-a-vat (BIV) skeptical argument says that S can know that S has hands only if S can know that S is not envatted; this is an instance of the closure principle that S can know that P and that P entails Q only if S can know that Q.¹ Closure deniers reject this, arguing that I can know I have hands even if I can’t know that I’m not envatted.² Closure denial is a partially concessive response in that it retains mundane hands-knowledge but concedes that we lack anti-skeptical knowledge.³

A second partially concessive response is represented by contextualism. Contextualism is the view that the truth-conditions of knowledge attributions vary with the ascriber’s context. On this view, the ascriber’s context determines what the standards for knowledge are in that context. Applied to BIV skepticism, contextualism says that in mundane contexts the ascriptive statement “S knows that she has hands” is true, and this because S’s reasons for thinking she has hands are sufficient to meet the standards associated with the word “knowledge” which are operative in mundane contexts. However, in the skeptical context the standards associated with “knowledge” are higher, and here ordinary reasons no longer cut it, hence in the skeptical context the ascriptive claim that “S knows that she has hands” would be false. Contextualism is a partially concessive response to radical skepticism because it tells us that sometimes the claim “S know that S has hands” (made by a handed person who sees her hands) is true and sometimes it isn’t.⁴

A third partially concessive reply to radical skepticism is the contrastivist one. Contrastivism says that knowledge is a three-place (ternary) relation rather than a two-place (binary) relation. That is, whenever S knows that P, S knows P rather than some contrast proposition Q.⁵ This opens up the possibility that S knows P rather than Q but not P rather than R. And this is exactly how contrastivism handles the problem of BIV skepticism. The contrastivist says that I know I have hands rather than stumps but insists that I don’t know I have hands rather than vat-images of hands. Contrastivism is a partially concessive reply because it retains hands-knowledge but gives up anti-skeptical knowledge.⁶

A potentially concessive reply to skepticism is something we can call “hyper-rationalism.” Hyper-rationalists deny that we can use the Moorean truths of commonsense to rule out the skeptic’s argument. They pride themselves on being properly philosophical and chide Mooreans for being un-philosophical. Their approach to skepticism says that we cannot reject the skeptic’s argument unless we can give a successful philosophical argument against her premises. Stroud (2000), for instance, betrays a kind of hyper-rationalism when he says that skepticism beats common sense “hands down” regarding the “philosophical question of the possibility of knowledge,” accusing Moore of a mere assertion of common sense that betrays a “kind of uncomprehending dogmatism” (29); for Stroud, Moore’s commonsensism is “unsatisfying philosophically” (30). Technically, skeptics would be hyper-rationalists who say that there is no such argument. Some, perhaps most, hyper-rationalists, however, will not be skeptics because they think there is a good philosophical argument against skepticism. Many of those philosophers who reject Mooreanism do so, I suspect, because they are hyper-rationalists. The Moorean differs from the hyper-rationalist in thinking that we can use commonsense truths to rule out the skeptic’s premises – even if we cannot philosophically diagnose and respond to the skeptic’s error. For the Moorean, the epistemic propriety of rejecting the skeptic’s argument does not depend on having a successful philosophical argument against the skeptic. The Moorean looks askance at the hyper-rationalist’s repeated insistence that her commonsense beliefs are in epistemic jeopardy until and unless they can be redeemed by an anti-skeptical philosophical counter-argument. As Bergmann (2006) says, in a similar context, the skeptic’s insistence that we can retain our mundane commonsense knowledge only if we can philosophically answer the skeptical challenge “is an indication of a philosophical temperament gone awry” (236).

Does Mooreanism’s incompatibility with hyper-rationalism imply that, on the Moorean view, commonsense beliefs are totally immune to skeptical arguments? For two reasons it does not. First, Mooreans are generally happy to allow that at least some items of commonsense can be overturned by scientific arguments; what they insist on is that they cannot be overturned by mere philosophical arguments (more on this later). Second, Mooreanism is compatible with the following idea: if S thinks that her belief B in commonsense proposition P has been defeated by philosophical skeptical argument A, or if S can’t say whether or not B has been defeated by A, then B has been defeated. What the Moorean will say about this is that B has been defeated only because S made the following mistake: she overestimated A’s epistemic credentials whilst underestimating P’s epistemic credentials. If she had appraised her epistemic situation correctly, she would’ve seen that P is epistemically superior to A.

I make no attempt here to argue against the partially and potentially concessive replies. I have brought them up for two reasons. First, I think that one can best understand the Moorean view by situating it in its proper epistemological context, which is alongside competing responses to skepticism. Second, I think that one can best appreciate the merits of the Moorean view by seeing what its competitors say to the skeptic.⁷ Because our partially and potentially concessive responses are concessive they are less than ideal responses to skepticism. What we should like, I think, is a response to skepticism that is more robust, one that preserves both mundane hands knowledge and anti-skeptical knowledge. If we cannot get a robust, non-concessive reply then we shall, perhaps, have to make do with one of these other replies. One advantage of Mooreanism over these partially and potentially concessive replies is that it is a robust, non-concessive reply to skepticism.

3 Mooreanism

3.1 Mooreanism’s Two Theses

According to Mooreanism, there is a class of commonsense propositions that are known by almost all normal, cognitively mature, properly functioning human cognizers and enjoy the following epistemic status: we have more reason to believe them than we do to believe the conjunction of a philosophical skeptical argument to the contrary. Call this the “Commonsense Thesis” (CT). Also according to Mooreanism, in virtue of CT we can use our knowledge of these commonsense propositions to provide a rational basis for rejecting (conjunctions of) skeptical premises.⁸ Call this the “Transmission Thesis” (TT). Mooreanism, understood as a commonsense metaphilosophy, is the conjunction of CT and TT.⁹

Philosophers refer to the things the Moorean thinks are safe from the skeptic’s premises in a variety of ways. Lemos (2004) refers to them as “common sense propositions,” Kelly (2005) as “Moorean facts,” and Armstrong (2006) as “Moorean truths.” One of the most important things to note about CT and TT is the relationship between them: namely, that it is the superior epistemic status of the Moorean truths (CT) which allows us to reason from them to the negation of the conjunction of the skeptic’s premises (TT). We need to follow Lemos (2004:6–14) here by underscoring, as he does, that Moorean truths have their anti-skeptical powers by virtue of their epistemic status rather than by virtue of their popularity or ubiquity. Merely being widely known is not what makes a proposition a Moorean fact. First, some Moorean facts might not be widely known; second, not all of our knowledge is on the same epistemic level. Moorean facts are those we have an especially strong justification for, strong enough to swamp whatever justification we might also have for skeptical arguments to the contrary. It is because we have more justification for Moorean truths than we do for conjunctions of skeptical premises that we can reject the latter on the basis of the former. Thus, when there is a conflict between the conjunction of the skeptic’s premises and a Moorean truth, it is more rational to stick with the Moorean truth and reject the skeptical conjunction.¹⁰

3.2 More on Moorean Facts

To get a better grip on Mooreanism we need to say something about Moorean facts or truths, what Lemos calls “common sense propositions” – these things it is more rational for us to believe than it is for us to believe the conjunction of the skeptic’s premises. Clearly, Mooreans stand in the commonsense tradition in philosophy, a tradition that includes Reid, Moore, Chisholm, and many others.¹¹ I will not here attempt to give a definition of commonsense propositions, but I should like to give some examples and then say some things about them.¹² In terms of examples, consider the following partial list, originally from Reid and Moore and nicely distilled and paraphrased by Boulter (2007:29–30).

• 1. There are in the Universe an enormous number of material objects (e.g. our bodies, other people, animals, plants, stones, mountains, rivers, seas, planets, tables, chairs, etc.).

• 2. Human beings have minds inasmuch as we have a variety of mental states, including acts of consciousness. We see, hear, feel, remember, imagine, think, believe, desire, dislike, will, love and so on.

• 3. Material objects can and do exist when we are not conscious of them.

• 4. My well-functioning memory is reliable if not infallible when concerned with recent events.

• 5. By attentive reflection a man can have a clear and certain knowledge of the operations of his own mind.

• 6. Moral judgments are true or false.

  Boulter adds that it is also part of commonsense sense to say that we know propositions such as (1) – (6) to be true. Other examples of commonsense are not hard to find; consider the following:

• 7. Dogs are not made of glass.

• 8. I have hands.

• 9. There are other human beings.

• 10. Rocks can’t do calculus.

• 11. Nothing can be a square and a circle.

• 12. Newborns can’t run marathons.

I would also add, in line with Boulter and many other commonsensists, that it is part of commonsense that we know (7) – (12) to be true; or, if that knowledge claim is too strong, then I would say at least that we have more justification to believe (7) – (12) than we do to believe the premises of any skeptical philosophical argument to the contrary.

I will now make a few cursory remarks on Moorean facts. These are commissive, in a sense – though minimally so, in my view – and might even be too commissive for someone attracted to a more minimalist kind of Mooreanism, or to a Mooreanism which is commissive in different ways. Readers who find the previous examples sufficient and the following remarks unnecessarily commissive are free to, as it were, delete these cursory remarks from Mooreanism as they conceive of it (or, alternatively, to supplement the examples with their own different commitments).

Moorean facts, though readily accessible to “the vulgar” – and thus not too complicated – need not be widely held. What percentage of the world’s population believes that you have hands? Nor need Moorean facts be psychologically irresistible (though perhaps many are): a philosophical argument might persuade someone to give up belief in a Moorean proposition, though this would be an epistemological mistake, according to Mooreanism. Moorean facts do not necessarily include anything anyone might think of as a matter of commonsense; thus, that heavier objects fall faster than lighter objects is not a matter of commonsense. The fact that something is widely believed in a given culture or society is not, by itself, enough to make it a Moorean truth.¹³ Moorean facts concern the common affairs of ordinary life, not the “expertise-admitting” (Nicholas Rescher’s phrase) domains of physics, economics, neuroscience, and so on.¹⁴ Moorean facts will typically be things that are believed in the basic way, i.e. noninferentially. However, we shouldn’t rule out the possibility of an inferentially believed Moorean fact.¹⁵ And Moorean facts, though typically believed in the basic way, can be argued for, it’s just that in the typical case this needn’t be done, the reason being that the premises of the argument for a Moorean truth would be less obviously true than the Moorean truth itself.¹⁶

Moorean propositions can be necessary truths (e.g. 1 + 1 = 2) or contingent (e.g. I have hands) – and a priori or a posteriori, as the parenthetical examples show. Though Moorean truths need not be widely held, as I indicated above, many will be. Many believe, for instance, that there are other minds and that 1 + 1 = 2. Moorean truths may not be uttered very often, mainly because uttering them would communicate that which everyone already knows, the taken-for-granted truths of ordinary life. If we think, with Audi (1999), of a self-evident truth as a truth that one comes to know just on the basis of understanding it, then we can say that some Moorean facts will be self-evident, but others won’t. Moorean truths can be singular propositions (e.g. Harry is a human) or general propositions (e.g. rape is bad). And, as the second parenthetical example in the last sentence (i.e. rape is bad) indicates, Moorean facts can be evaluative or non-evaluative (e.g. I am not made of glass), though this is controversial of course.¹⁷ Moorean facts, when not explicitly acknowledged, will be things we nearly always take for granted, such as that material objects continue to exist when unperceived by us. They are things we are disposed to believe, things we would readily assent to were they to be brought before the mind’s eye – things we do assent to when they are brought before the mind’s eye. The denial of a Moorean fact will come off as absurd, and even those who profess to deny them will (typically, at least) act as if they believe that they are true. In fact, we may often find ourselves questioning the cognitive competence of those who don’t believe Moorean truths: something has clearly gone wrong, epistemically speaking, with someone who affirms that she no longer exists.¹⁸ Importantly, Moorean facts have more going for them, epistemically speaking, than the conjunction of the skeptic’s premises.¹⁹ We might say, with Rescher (2005), that Moorean truths are, due to their solid epistemic status, “beyond reasonable doubt” (29). Because of this, we can say that Moorean truths should be accepted, in the normal case, even if, for some strange reason, they aren’t widely accepted. Finally, Moorean truths – or some of them, at least – though epistemically superior to the skeptic’s premises, are not infallible or immune from any refutation whatsoever: they can be defeated.²⁰ I could, for example – in the manner of Neo from the Matrix movies – get excellent evidence that the external world as I have known it is an illusion, though this is quite unlikely (more on this below).²¹

3.3 Mooreanism is Metaphilosophical

Mooreanism so understood is a metaphilosophical claim that can be endorsed by epistemologists who have different views about such things as the nature of justification and the proper resolution of the Gettier problem.²² As I will discuss later on, the metaphilosophical nature of Mooreanism is not always recognized by epistemologists, some of whom identify Mooreanism with other, first-order epistemological claims. Metaphilosophically understood, Mooreanism says that commonsense propositions (of the sort outlined in the previous subsection) are more plausible than skeptical premises to the contrary and hence that it wouldn’t be rational to abandon the former in favor of the latter; hence, when doing philosophy we should regard commonsense truths as data to accommodate and explain. Lemos (2004:7) notes the way that Mooreans often respond when faced with a skeptical argument that runs counter to a commonsense proposition: “(1) P is a common sense proposition that I and many others know. (2) Theory T implies that P is false. (3) Therefore, Theory T is false or unreasonable.” Strictly speaking, of course, Lemos’s specific argument strategy here, though common in Moorean circles, is not necessary for the pairwise plausibility comparison between Moorean facts and skeptical premises to the contrary.²³ The heart of the Moorean reply to skepticism is not Lemos’s argument per se, but rather the insight that we have more reason to believe commonsense propositions than we do to believe skeptical premises to the contrary. Coady’s (2007) summary of the Moorean strategy is spot-on: “Moore’s basic move is to challenge the idea that we could have anywhere near as much certainty about the philosophical arguments for scepticism and reductive metaphysics as we have about the common-sense beliefs they are meant to topple” (106). As a metaphilosophical response to skepticism, Mooreanism tells us what we (epistemically) should do when we encounter a skeptical argument that runs counter to a commonsense truth. The Moorean response doesn’t diagnose the malady afflicting the skeptic’s premises, nor prescribe a cure – these things need doing, of course, but they are not done by Mooreanism per se, and an epistemologist who is a Moorean and who offers a specific diagnosis and prescription does not do so qua Moorean, though she may do because she is a Moorean.²⁴ Thus, Mooreans may differ among themselves about just exactly where the skeptic goes wrong, and they may draw different epistemological lessons from the same skeptical argument. Mooreanism, understood as a commonsense metaphilosophy, does not commit one to substantive views on most of the questions that epistemologists debate (more on this later).²⁵

3.4 The Superior Epistemic Status of Moorean Facts

One obvious question here is why we should think that Moorean truths are better off, epistemically, than the skeptical arguments that seek to overturn them. Why do we have more justification for a Moorean fact than we do for the conjunction of skeptical premises which entail the negation of that fact? Different Mooreans may answer this question in different ways, depending of course on how they think about epistemic justification. Moore himself seems congenial to an evidentiary answer, i.e. an answer that appeals to evidence.²⁶ In his “Proof of an External World,” Moore (1939) contends that he has “conclusive reasons” and “conclusive evidence” (149) that is he awake and not dreaming. Moore concedes that he can’t say just exactly what this conclusive evidence is, but this needn’t imply that he lacks this evidence, nor that he doesn’t know that he has it. As Lycan (2007: 98) explains, Moore doesn’t need a theory of evidence to know that he’s got evidence for P. If we follow Moore himself, then, we should be led to say that we have much better evidence for commonsense propositions than we do for the conjunction of the skeptic’s premises. Having said that, it seems clear that in saying these things Moore is not necessarily committing himself to evidentialism; it seems equally clear that Mooreanism as a commonsense metaphilosophy is not committed to evidentialism.

Moorean truths are of different kinds and thus admit of different sorts of epistemic support. I believe that I have hands, a belief I hold on the basis of my perceptual evidence. I also believe that 2 + 1 = 3, a belief I hold on the basis of rational intuition. As I close my eyes and then raise my left hand over my head, I can tell by proprioception that my left hand is over my head. The Moorean epistemologist may of course go on to develop a theory of evidence and may therefore have quite a few things to say about evidence in general and about evidence in these three examples. Alternatively, some Mooreans may reject evidentialist views about justification and explain the lofty epistemic status of Moorean facts in terms of some alternative theory of epistemic justification, such as reliabilism or proper functionalism.²⁷ But the Moorean qua metaphilosopher needn’t have such a theory at her disposal to be able to say that she does indeed have strong epistemic support for her belief that she has hands, that 2 + 1 = 3, that her left hand is over her head, and so forth. In fact, and following the spirit of Moore’s remark that he knows he has conclusive evidence for commonsense truths even if he can’t say exactly what that evidence is, we can add that, for most Mooreans, no epistemological explanation is needed at all in order to know or justifiably believe in the truths of commonsense. My grandmother, for example (who is not a philosopher), doesn’t need to engage in epistemological explanation-giving to know that her dog is not made of glass and that this claim is more justifiable than philosophical skeptical arguments to the contrary.²⁸

In any case, when we turn our attention to the skeptic’s premises we notice, say Mooreans, that our evidence for the philosophical theses the skeptic needs to negate commonsense truths is paltry at best. Consider the principle that I am justified in accepting a belief formed by some method M only if I first know that M is reliable. Call this the “Methodist Requirement.”²⁹ What could the skeptic’s evidence for the Methodist Requirement be? It looks like it would have to be an intellectual seeming or rational intuition.³⁰ Now the Methodist Requirement entails, in conjunction with some other premises, that I don’t know I have hands. My evidence for my hands is, let’s say, a perceptual seeming that I have hands plus the testimony of others.³¹ Now, my own perceptual seeming, plus the testimony of others, is very strong evidence for my belief that I have hands. Insofar as I do have an intellectual seeming in favor of the Methodist Requirement, I do have some justification for it.

On the other hand, it seems clear that my total evidence favors my having hands over the truth of the Methodist Requirement. First, my seeming that I have hands is much stronger than my seeming that the Methodist Requirement is true. Second, I have testimonial confirmation that I have hands, but very little if any testimonial confirmation in favor of the Methodist Requirement. Third, the Methodist Requirement has, in conjunction with some other premises, very bizarre, counterintuitive consequences – such as that no one knows anything – while the claim that I have hands does not. Fourth, the Methodist Requirement is highly controversial and mostly rejected by the people who have thought most about it: epistemologists. Finally, I can confirm, by making predictions and then seeing if they come true, my belief that I have hands; I can’t do the same for the Methodist Requirement. At the end of the day, then, it seems that my belief that I have hands has much more epistemic support than does the Methodist Requirement. And, assuming – as Kelly (2005, 2008) does in this context – that I should believe what my total evidence indicates, it seems that the choice between I have hands and the Methodist Requirement is true is fairly easy: I should believe that I have hands, and reject the Methodist Requirement.³²

The problem gets worse for the skeptic when we take note of the fact that the Methodist Requirement and other like-minded principles entail radical skeptical claims only when they are conjoined with other propositions. What we are really comparing, then, are conjunctions of skeptical premises with Moorean truths. On our total evidence, say Mooreans, Moorean truths seem to come out way ahead of these skeptical conjunctions. I have much more evidence for my own hands than I do for any such skeptical conjunction. In general, then, when we encounter skeptical arguments, what we really have to do is make a “Moorean plausibility comparison,” as Lycan (2007:94) puts it, between the Moorean fact at hand and the conjunction of the skeptic’s premises. When we do this, the Moorean facts win. It should be kept in mind here that the philosophical explanation I have been trying to give for the epistemic superiority (on the Moorean view) of commonsense truths is not something that needs to be given by, say, an ordinary person (or even a philosopher for that matter) in order for that person to retain her knowledge or justified belief that she has hands in the face of a skeptical argument to the contrary. The plausibility comparison can be carried out without epistemological explanation. Again, on the Moorean view, my grandmother needn’t do epistemology to know, say, that her dog isn’t made of glass, a point that holds even if she comes into contact with someone who gives her a skeptical argument against this commonsense truth.

But will Moorean truths always win, or are they sometimes defeated? Mooreans don’t deny that Moorean truths can be defeated, though they do say, with Rescher (2005:59), that the “burden of proof” borne by the skeptic “can only rarely be discharged.” What Mooreans do typically deny is that purely philosophical arguments can defeat commonsense propositions. Let’s say that a skeptical argument is purely philosophical just in case none of its premises present any evidence that our faculties are malfunctioning or otherwise unreliable, or that the commonsense proposition in question is false. Purely philosophical skeptical arguments do their work by pointing to the conflict between Moorean truths and philosophical principles in metaphysics and epistemology. What Mooreans typically do affirm is that arguments utilizing, as Lycan (2001) puts it, “careful empirical investigation and scientific theorizing” (40), can sometimes overturn commonsense propositions.³³ Kelly’s (2008) explanation of this asymmetry between purely philosophical arguments and those that utilize empirical investigation and scientific theorizing is that scientific theories can use prediction to get confirmation in a way that abstract philosophical theories cannot. Whether one accepts this explanation of the asymmetry or not, it is not hard to see that there is a big difference between a BIV-style argument which utilizes empirical observations that point to my actually being a BIV and one that merely highlights the conflict between my belief that I have hands and a certain philosophical principle.

3.5 Mooreanism and the Commonsense Tradition

My conception of Mooreanism as a commonsense metaphilosophy consisting of CT and TT fits well with the commonsense tradition. Though I cannot give a satisfying overview of this tradition here, I would like to try and briefly highlight this fittingness. Boulter (2018) argues that the Scholastics had a commonsense metaphilosophy, but I shall begin with Reid (1764), who argued that commonsense has authority over philosophy:

In this unequal contest between common sense and philosophy the latter will always come off with both dishonour and loss; nor can she ever prosper until this rivalry is dropped, philosophy gives up encroaching on the territory of common sense, and a cordial friendship is restored; for, in reality, Common Sense holds nothing of Philosophy, nor needs her aid. But, on the other hand (if I may be permitted to change the metaphor), philosophy’s only root is the principles of common sense; it grows out of them, and draws its nourishment from them; when it is cut off from this root its honours wither, its sap is dried up, it dies and rots. (7)³⁴

What then does Reid do, when faced with a philosophical argument that contravenes a commonsense truth? He makes precisely the move that Lemos cites as characteristic of the commonsense philosophers: since philosophical theory T contradicts commonsense proposition P, theory T is false.

A traveller who has good judgment may mistake his way, and be led unawares onto a wrong route; and, while the road is fair before him, he may go on without suspicion, and be followed by others; but when the road ends at a coal-pit, he doesn’t need much judgment to know that he has gone wrong, and perhaps to find out what has led him astray. (11)

Reid, then, appears to be a Moorean – to speak anachronistically.

Moorean epistemology is perhaps most commonly associated with Moore’s (1939:146) proof of the external world, in which Moore reasoned from his having hands to the existence of the external world. Now just exactly what was Moore up to when he gave this proof? Greco (2002) argues – correctly, I think – that in giving this proof Moore is following Reid on a number of points, two of which are pertinent here. The first is that commonsense has authority – or “epistemic priority,” as Greco (2014) elsewhere puts it – over philosophy; the second is that, when there is a conflict between commonsense and a skeptical argument, we should examine the skeptic’s arguments to see where he has gone wrong. Moore explicitly argues this way in places, such as in his essay, “Four Forms of Skepticism” (1959), where he contends that his knowledge of mundane commonsense propositions is more certain than the skeptic’s premises.

What I want, however, finally to emphasize is this: Russell’s view that I do not know for certain that this is a pencil or that you are conscious rests, if I am right, on no less than four distinct assumptions: (1) That I don't know these things immediately; (2) That they don't follow logically from anything or things that I do know immediately; (3) That, if (1) and (2) are true, my belief in or knowledge of them must be “based on an analogical or inductive argument”; and (4) That what is so based cannot be certain knowledge. And what I can’t help asking, myself is this: Is it, in fact, as certain that all these four assumptions are true, as that I do know that this is a pencil and that you are conscious? I cannot help answering: It seems to me more certain that I do know that this is a pencil and that you are conscious, than that any single one of these four assumptions is true, let alone all four. (226)

On this Reidian interpretation of Moore’s approach, the real point of the famous proof is the epistemic priority that commonsense enjoys over skeptical philosophical principles.

Many philosophers in the last several decades, including a number of contemporary epistemologists, have concurred with Reid and Moore, such as Chisholm (1989: 3–4), Sosa (1999), Lycan (2001), Michael Huemer (2001: 43), Greco (2002), Lemos (2004), Rescher (2005), Armstrong (2006), Bergmann (2008), Black (2008), and Kelly (2005, 2008), among others. Moorean-friendly passages that fit well with my conception of Mooreanism as a commonsensism abound in philosophical discussions of skepticism; this conception of Mooreanism lines up with how those within the commonsense tradition typically think about the Moorean response to skepticism.

3.6 Mooreanism, Liberalism, and Conservatism

We can gain an even clearer understanding of Mooreanism by examining its relationship with recent debates between liberals and conservatives in the epistemology of perceptual belief. In many of these discussions Mooreanism is explicitly invoked, usually either to contrast it with conservatism or to make the point that Mooreanism is compatible with both liberalism and conservatism. I think this latter point is correct, and that getting clear on all this will help us to better understand Mooreanism itself. Conservatism says that you have justification to believe you have hands (or, that H is true) only if you also have some independent justification to believe that you are not in a radically deceptive skeptical scenario (or, that ~ S is true). Liberalism denies this, affirming that your perceptual experience that you have hands can, all by itself, give you justification to believe ~ S.³⁵

The Moorean response to skepticism, or at least Moore’s proof – (1) I have hands; (2) If I have hands there’s an external world; (3) So, there’s an external world – is sometimes regarded as a liberal response: the thought is that the Moorean denies that justification to believe that one has hands depends on having independent justification to believe ~ S.³⁶ Coliva (2012), for example, pits Wright’s “conservative” critique of Moore’s proof against Pryor’s “liberal” defense of Moore’s proof; the critique in question, contested by Pryor, is that Moore’s proof is epistemically circular because Moore’s perceptual experience of being handed provides justification to believe that he has hands (the first premise of Moore’s proof) only if he already has independent justification to believe the conclusion, which is that there is an external world. So, the proof is (epistemically) circular, on this view, because Moore has justification to believe the first premise only if he already has justification to believe the conclusion. Pryor disagrees with the charge on the grounds that Moore’s justification to believe the first premise does not depend on first having justification to believe the conclusion. The dialectic is set up in such a way that Mooreanism appears to be the liberal view, or at least a liberal view, and hence that any conservative view would thus be non-Moorean.

As a number of epistemologists – such as Silins (2008), Tucker (2010a), Neta (2010), and Willenken (2011) – have pointed out, however, Mooreanism is not incompatible with conservatism. The explanations in each case are different, and I won’t attempt to rehearse them here. What I would like to point out is that it should be clear that, as I have defined and explained Mooreanism here, it is fully compatible with conservatism. Neither CT nor TT include or entail a rejection of conservatism. TT might look to be incompatible with conservatism as it says that it is in virtue of CT that we can use our knowledge or justified belief in Moorean truths as a rational basis to reject skeptical arguments which conflict with them. This might make it seem as though the Moorean takes the liberal view and denies that we need independent justification for ~ S to have justification for believing that we have hands. In reality, though, Mooreanism is neutral with respect to liberalism and conservatism. A liberal Moorean will say that you can use your knowledge of or justified belief that H to rule out skeptical arguments to the contrary (and so affirm ~ S), and also that having justification for H doesn’t depend on having independent justification for ~ S. A conservative Moorean, by contrast, will say that you can use your knowledge of or justified belief that H to rule out skeptical arguments to the contrary (and so affirm ~ S), and also that your perceptual experience of having hands justifies H only if you also have independent justification for ~ S.³⁷ We can think of it this way: conservatism answers the question, “When will a perceptual seeming that P give one justification to believe that P?” The conservative answer is: “When one also has some independent justification to reject skeptical arguments that ~ P.” Different conservatives will say different things about what this independent justification must be. Mooreanism answers the non-identical question, “Am I in a position to reject skeptical alternatives to P?” The Moorean answer is: “Yes, for P is more rational for me, has more going for it, than does the skeptic’s premises.”

Based on the foregoing, let me now point to an interpretation of Mooreanism with which I take issue. For Neta (2010), Mooreanism would be the claim that the justifier which gives me justification for H also gives me justification for ~ S; let “J1” stand for this justifier, and we’ll say that J1 is my perceptual seeming of being handed. Neta’s view is, on my view, deficient, for two reasons.³⁸ One is that the Moorean need not say that J1 is also the very same thing which gives me justification for ~ S. A Moorean is free to say that my justification for ~ S is independent of my justification for H. A Moorean might say that what gives me justification for ~ S is my total evidence, or a track record argument, or whatever. The other difficulty with Neta’s conception of Mooreanism is that it fits less well with the commonsensist, Moorean tradition. Mooreanism is really, at heart, a metaphilosophical claim about the epistemic superiority of commonsense propositions – as I have been at pains to argue in this paper – and not the view that my perceptual seeming for the existence of my hands is also what serves as the justification for the view that there is an external world. Mooreanism is fully compatible with that first-order epistemology claim, but it is not that claim.

Before drawing this subsection to a close one final clarification of the Moorean view is in order. Wright (2008) argues that neo-Mooreanism is the view that Moore’s proof doesn’t suffer from transmission failure. In a case of transmission failure, the premises of an argument, though themselves justified, would fail to pass along this justification to the conclusion they entail. Wright’s position presents us with an occasion to further clarify the Moorean position. A Moorean need not deny that Moore’s proof suffers from transmission failure (so, interestingly, a Moorean, on my view, could reject what Wright refers to as neo-Mooreanism). We can imagine a Moorean who endorses (i) the conservative claim that Moore has justification for H only if he has independent justification for ~ S and (ii) a view of transmission according to which, if one’s justification for a premise depends on one’s justification for the conclusion, then the premise cannot transmit justification to the conclusion (whether this is the correct way to think about transmission is not relevant at the moment).³⁹ Someone who endorses (i) and (ii) can also endorse CT and TT. This Moorean will say that it is more rational for Moore to believe that he has hands and that he is not in a radically deceptive skeptical scenario than it is for Moore to believe the conjunction of the skeptic’s premises, and also that in virtue of this Moore may reject the skeptic’s argument.⁴⁰

4 Moore for Almost Everyone

Having explicated Mooreanism I’d like to now argue that a great many epistemologists are free to be Mooreans. Epistemologists debate many things, most of which are orthogonal to debates between Mooreanism and its competitors. For starters, consider debates about epistemic justification, such as those between internalists and externalists, or foundationalists and coherentists. Partisans of any of these views can be Mooreans: since Mooreanism is not a theory of epistemic justification it is compatible with any theory of justification that is compatible with CT and TT. Start with internalists and externalists: both are free to be Moorean in the sense that both are free to affirm CT and TT. And you do find internalists, such as Huemer (2001), and externalists, such as Greco (2002), both affirming Mooreanism. Mooreanism is not a theory of epistemic justification, so folks who subscribe to different theories of epistemic justification are free to embrace Mooreanism provided that those theories don’t entail the denial of CT or TT.

Coherentists and foundationalists can also be Mooreans. I am thinking of foundationalism as the view that non-inferential justification is possible and actual. There are different versions of coherentism, but the core idea of the view is that coherence is necessary for justification: a belief is justified only if it is part of a coherent web of beliefs. Advocates of both views can be Mooreans because they can both embrace CT and TT. The foundationalist Moorean will likely say that Moorean truths are properly basic for us. And the coherentist can say that the negations of Moorean truths don’t cohere as well with our standard systems of belief as do the Moorean truths. So, for example, Poston’s (2014) explanatory coherentism, which holds that justification for believing a proposition P depends on whether P belongs to an explanatory system that is more virtuous than any other competing system, is compatible with CT and TT.⁴¹ The Moorean who adopts explanatory coherentism can say that we have more justification for believing the Moorean truths because they belong to an explanatory system that is more virtuous than any competing system which includes their skeptical negations.

We see the same dynamic at play when we move from debates about justification to debates about knowledge. Consider, for example, debates about the Gettier condition and knowledge-first epistemology. Regarding the former, safety theorists and sensitivity theorists are both free to be Mooreans. Again, Mooreanism doesn’t take a position on the fourth condition on knowledge. Safety theorists say that S knows P only if it’s the case that, if S believes P, it’s likely that P is true. Sensitivity theorists say that S knows that P only if it’s the case that, were P false, S wouldn’t believe P. These are the two main theories about what’s required for knowledge in addition to justification, truth, and belief. And it should be clear that both theories are compatible with CT and TT. And so it should not be surprising to find safety theorists like Sosa (1999) and sensitivity theorists like Black (2008) both endorsing Mooreanism.

The same holds for knowledge-first epistemologists and their detractors. The two key ideas of Williamson’s (2000) knowledge-first epistemology are that knowledge is unanalyzable (i.e. it cannot be broken down into component parts) and that a cognizer’s evidence is identical to that which she knows, or E = K. The knowledge-firster can argue that the skeptic’s argument rests on the dubious assumption that our evidence in the skeptical scenario is the same as our evidence in the non-skeptical scenario.⁴² Our knowledge that we have hands, say, can be used to reject the premises of the skeptic’s argument that we don’t. Though knowledge-firsters like to talk about knowledge rather than rationality or justification, they can surely endorse CT and TT: nothing forbids them from saying that it’s more rational to affirm commonsense truths than it is to affirm skeptical premises. Their other claim, that knowledge is unanalyzable, also fails to conflict with CT and TT. So, knowledge-firsters are free to be Mooreans. And so, obviously, can those who reject it, provided they don’t embrace any philosophical theses which entail the negation of CT and TT.

A final example should suffice. Pragmatic encroachers say that whether you know that P depends on, as Fantl and McGrath (2014:88) put it, “what’s at stake in whether the proposition is true.” When a lot is at stake, you need extra-strong evidence for P to have enough justification to know that P. So, if a lot was at stake in whether Moorean truths were true, the normal amount of evidence we have for those truths may not be enough for us to know them. Pragmatic encroachment isn’t really a response to skepticism, but considered in relation to skepticism it initially looks like a partially-concessive, and thus non-Moorean, “response” to skepticism: we can know Moorean truths, and maybe even use them to rule out the skeptic’s premises, but only if there isn’t too much at stake in whether the Moorean truths are actually true. Recall, however, that Moore thinks he has conclusive evidence and certain knowledge that he has hands. If this is right, then it’s quite reasonable to think that our beliefs in Moorean facts do meet any extra-strong evidence requirements that might be operative in situations where the stakes are high. Thus, pragmatic encroachers are free to be Mooreans. Given the foregoing, it would seem to be the case that most epistemologists are free to be Mooreans.

Most of the debates in epistemology are orthogonal to whether Mooreanism is true or not. Is knowledge different from understanding? Is knowledge more valuable than mere justified true belief, and if so, why? What is the relationship between belief and credence? On these and many other issues – indeed most of the debates in epistemology – the answer one gives has little to no bearing on Mooreanism. Let me close by mentioning three fuzzy cases.

I said above that contextualism is a non-Moorean view because it is a partially concessive position, granting knowledge in non-skeptical contexts but revoking it in skeptical contexts. There may be a way, however, for contextualists to be Mooreans. Conee (2014) argues that contextualism is technically compatible with the claim that “knowledge attributions like ‘John knows that he has hands’ are true even in the skeptical context because John’s belief that he has hands meets the more stringent epistemic demands operative in that context.” Now, if a contextualist agreed with Moore that we have conclusive evidence that we have hands, then it’s not hard to see how a contextualist could think that a knowledge ascription like “John knows he has hands” would be true even in the skeptical context. For, what more could anyone want than conclusive evidence? Moore (1925) himself claimed to know the truths of commonsense with certainty; if that’s so, then perhaps Conee’s suggestion is on the mark and I was mistaken to say, as I did above, that contextualism is a non-Moorean view.

One problem with this line of thought is that it doesn’t appear to do justice to contextualism itself. Contextualism is supposed to explain why we are intuitively pulled in two different directions: toward the thought that we know we have hands and the thought that, when faced with a skeptical argument, we don’t after all know that we have hands. That is, contextualism is supposed to do justice to our commonsense intuitions and our pro-skeptical intuitions. It does this by giving something to both sides: our intuitions are right in each case, something explained by the shift in standards that occurs when one moves from a non-skeptical context to a skeptical one. But if we retain mundane hands-knowledge (say) even in the skeptical context, in what sense are we doing justice to our intuition that we don’t, in the face of the skeptical argument, have such knowledge? Thus, it looks like we can make contextualism compatible with Mooreanism only by rendering it incapable of doing the very thing that made it an interesting and attractive position in the first place.⁴³

Another fuzzy case is infallibilism. This thesis is usually understood as the view that knowledge that P requires that one’s grounds for P entail that P.⁴⁴ And one of the most ready-at-hand objections to infallibilism, so construed, is that it leads to too much skepticism; for, after all, we rarely, if ever, have grounds for our beliefs that entail the truth of their contents.⁴⁵ Because of this, infallibilism is almost universally rejected. But this rejection may well be far too quick, at least as far as Mooreanism is concerned. For starters, as Climenhaga (2017) points out, no small number of important non-skeptical philosophers appear to have been infallibilists, including Descartes, Locke, Keynes, Moore himself, and Ross. On Moore’s own account, as we have seen, we do indeed have certain knowledge of many of the truths of commonsense. Climenhaga’s preferred conception of infallibilism says that S knows that P iff P is epistemically certain for S, where P is epistemically certain for S iff S clearly perceives that P. Given that Mooreanism just as such does not take a position on many “first-order” epistemological questions, it certainly seems possible for a Moorean to affirm that we do clearly perceive that Moorean propositions, or at least many of them at any rate, are indeed true, and hence are epistemically certain for us and also to deny that the same holds for the controversial philosophical principles needed by the skeptic.

Climenhaga does confess that infallibilism has moderately skeptical implications but denies that this is problematic. Whether this is so, and whether it is problematic for Mooreanism, will depend on whether Moorean facts, and how many of them, can be clearly perceived. And this, in turn, will depend on one’s other epistemological and metaphysical commitments. Thus, we cannot fully adjudicate the compatibility of Mooreanism and infallibilism here. Perhaps the best we can do here is to say that infallibilism, just by requiring certainty for knowledge, does not thereby rule out certain knowledge of Moorean facts. But whether infallibilism can accommodate the full variety of Moorean facts typically accepted commonsense philosophers is an open question.

5 Conclusion

I have not here attempted to defend Mooreanism. Rather, I have focused on trying to get clear on how we should think about Mooreanism and then argue that, once properly understood, most epistemologists are free to embrace Mooreanism. What has emerged is that Mooreanism is an ecumenical and metaphilosophical, commonsensist response to radical skepticism. On the Moorean view, Moorean truths are epistemically superior to philosophical arguments to the contrary. Mooreanism so construed fits both with Moore’s own anti-skeptical arguments, with the commonsense tradition, and with the work of many contemporary commonsense epistemologists, some of whom work explicitly under the banner of Mooreanism. So understood, Mooreanism should not be identified with one particular position in specialized debates in epistemology, such as the view that the safety condition is the correct fourth condition on knowledge, or liberalism in the epistemology of perceptual belief, or even the view that Moore’s proof doesn’t suffer from transmission failure. All such understandings of Mooreanism miss the essentially metaphilosophical and ecumenical nature of the view. If you’re looking for a response to skepticism that is neither concessive, partially concessive, nor potentially concessive, then the Moorean camp is where you need to be.