Radical Doubt: From Pyrrho to the Brain in a Vat

Scepticism is not denial, and that is the first correction this article has to make. Our research file states the position directly: the sceptic does not claim that knowledge is impossible, because that would itself be a positive claim requiring justification. The Pyrrhonist suspends judgement, neither affirming nor denying, and the resulting condition is one of inquiry rather than of settled conviction. The file makes the etymological point in the same breath, and it is the point to hold onto for the next sixteen sections: skepsis means investigation. Everything the tradition is accused of, paralysis, nihilism, a shrug dressed as a philosophy, follows from reading it as an answer instead of as a refusal to give one.

Two of those five are the ones this article argues, and they pull in opposite directions. Our own file's table already separates them: Pyrrhonian scepticism is listed as therapeutic, its scope all theoretical matters, its claim that suspension of judgement leads to tranquillity; methodological scepticism is listed as heuristic, doubt as a tool for discovering what can withstand scrutiny. The same file states flatly that Descartes is not a sceptic. The Stanford Encyclopedia of Philosophy's entry on Descartes' Epistemology, consulted independently for this article, says the same thing from the other side: it describes his aim as being "to reach certainty", and it quotes him, in its own sentence, distinguishing himself from "the sceptics, who doubt only for the sake of doubting". The entry continues that phrase about certainty with an image of clearing ground, "to cast aside the loose earth and sand so as to come upon rock or clay"; the two fragments are joined in the source by a dash, which house style does not reproduce, so they are quoted here as two fragments and not silently run together. Hold the difference in one line: Sextus doubts in order to stop, and Descartes doubts in order to start.

What Pyrrho is reported to have said comes to us at third hand, through a summary by Aristocles, and our file gives it in three parts: that things are equally indifferent, unmeasurable and undecidable; that our sensations and opinions are therefore neither true nor false; and that we should be without opinion, uncommitted and unwavering, which will produce ataraxia. The three Greek epithets are adiaphora, astathmeta and anepikrita, transliterated here without the circumflexes the encyclopedia prints on the e in the second and third. Now the hedge, which is absent from our file and which changes what the passage can be used for. The Stanford Encyclopedia's entry on Pyrrho records that those three epithets are subject to competing interpretations, a metaphysical reading on which things really are indeterminate and an epistemological reading on which we simply cannot determine how things are, and it treats the choice between them as open. That is not a footnote. If the claim is metaphysical, then the founder of the tradition is making exactly the kind of positive assertion the tradition exists to refuse, and the disagreement is about him. No full English translation of the Aristocles passage is quoted here, because none was read for this article off a primary source.

The most repeated story about Pyrrho is also the least secure, and it is told here once with its hedges intact. Our file files it under the word Legend: that Pyrrho travelled with Alexander's army to India, that he was influenced by Indian ascetics, the gymnosophists, and that a connection to Buddhist or Jain scepticism has been debated but remains speculative. The Stanford Encyclopedia's entry is more guarded again. It hedges the expedition as something he allegedly joined, hedges the encounter with the naked wise men, the gumnosophistai, as something we are told, and says in terms: "Diogenes Laertius (9.61) claims that his philosophy developed as a result of this meeting, but it is not clear what basis, if any, he has for this assertion." It does allow that the Indians may have influenced his physical impassivity. The story may be told. It may not be used to explain where Pyrrhonism came from, and it is not used that way here.

What survives is less than a reader might assume. Our file names two works of Sextus, Outlines of Pyrrhonism and Against the Professors, and describes the first as the most complete surviving account of ancient scepticism. Both identifiers in its bibliography resolve exactly to the R. G. Bury translations in the Loeb Classical Library, published by Harvard University Press in 1933 and 1949, and both are cited below. One naming note, because the literature is inconsistent and this article has to pick: the Loeb English title is Outlines of Pyrrhonism, the Julia Annas and Jonathan Barnes translation calls the same work Outlines of Scepticism, and the standard abbreviation in the scholarship is PH. This article uses Outlines of Pyrrhonism throughout and quotes from Annas and Barnes, naming them each time. So here is the shape of the evidence base for a movement that ran for roughly five centuries: one author whose dates are a guess, plus scattered testimony.

The whole machine is in that one sentence, and the word to build on is the noun in it. Scepticism is an ability, a capacity to do something, not a set of beliefs about what can be known. The sentence then runs the ability through four stages in order: oppositions are set out, the opposed accounts turn out to be equal in weight, equipollence, which our file names in Greek as isostheneia; that equality produces suspension of judgement, epoche; and tranquillity follows. Our file gives the same chain in its own words and names both Greek terms. Its own opening section makes the corresponding point, that scepticism is an activity rather than a settled conviction, which is the same claim the definition makes and the same claim that answers the self-refutation objection in section 16. The spelling above is the source's, with two l's in tranquillity and no e in judgment, and the next paragraph spells the same word with one l, also as the source has it.

The strangest claim in the tradition is in the last stage of that sentence, and it needs the simile that our file quotes without a citation. Tranquillity follows suspension of judgement "as a shadow follows a body", six words which the Stanford Encyclopedia's entry on Sextus locates at Outlines of Pyrrhonism I 29, in the Annas and Barnes translation, noting the same image at Diogenes Laertius IX 107. Our file prints the phrase with no locus and no translator, which is why both are supplied here. The entry explains what the simile is doing: the tranquillity arrives unexpectedly, not through deliberate pursuit. Hold what that costs. On this account judgement cannot be suspended in order to become tranquil, because pursuing an end is itself a commitment. The peace is a side effect of giving up the search for a verdict, and it does not arrive for anyone who gives up the search in order to get it. The encyclopedia's own sentence carrying those six words spells tranquility with one l, where the passage at I 8 spells it with two. That inconsistency is in the source, not introduced here, and it is exactly why the two passages are quoted separately rather than merged.

The obvious objection arrives immediately: if judgement is suspended about everything, how does anyone get out of bed. Sextus has an answer and it is a specific passage, which our file summarises accurately without citing. It says the sceptic lives by appearances, phainomena, following custom, natural inclination, bodily needs and craft expertise, without committing to theoretical beliefs about the underlying nature of things. The passage behind that is Outlines of Pyrrhonism I 24, where four guides are named, and it is worth quoting rather than summarising, because the summary sounds like an evasion and the passage sounds like a life. In the Annas and Barnes translation, as the Stanford Encyclopedia prints it: "By nature's guidance we are naturally capable of perceiving and thinking. By the necessitation of feelings, hunger conducts us to food and thirst to drink. By the handing down of customs and laws, we accept, from an everyday point of view, that piety is good and impiety bad. By teaching of kinds of expertise we are not inactive in those expertises which we accept." Nothing there is a belief about how things really are. All four are reports about how things appear and what a body does about them.

Our file has one worked example of the method and it is the cleanest illustration in the subject, so the idea is used here and the wording is not. Take honey. The sceptic does not assert that the honey is sweet, which would be a claim about what honey is; the sceptic reports that it appears sweet, which is a claim about an appearance and cannot be wrong in the same way. That distinction, between reporting an appearance and asserting a fact about the world, is what makes the whole position livable, and it is what the fourfold observance above is built on. Our file prints a sentence to that effect inside quotation marks, with capital letters for emphasis, as though Sextus had said it. He is not recorded as having said it in those words, the emphasis capitals are our file's own house style, and the underlying passage, at Outlines of Pyrrhonism I 20, was not read off a primary text for this article. So the point is made here in this article's own words and no version of that sentence is presented as a quotation.

While the Pyrrhonists were suspending judgement, Plato's own Academy was doing something related and different. Arcesilaus, head of the Academy, whose dates our file gives as about 316 to 241 BCE and the Stanford Encyclopedia's entry on Ancient Skepticism gives more precisely as 316/5 to 241/0 BCE, attacked the Stoic criterion of truth. The Stoics held that some impressions are cataleptic: so clear and distinct that they compel assent and cannot be false. Arcesilaus argued that for any such impression an indistinguishable false one is possible. The encyclopedia supplies the case our file leaves out, and it is what turns an abstract claim into an argument a reader can hold: eggs. No impression of one egg could be distinguished from impressions of other eggs. The Stoic side of this belongs to Stoicism, which carries katalepsis in its own words as the Stoic answer to how anyone knows anything at all; this article carries the attack and does not re-run the doctrine. Our file also prints a sentence inside quotation marks, attributed to Arcesilaus, to the effect that nothing can be known, not even that. It could not be verified as his sentence from any edition, translation or scholarly citation for this article, and it does not appear here. The scholarly anchor for this section is Harald Thorsrud's chapter on Arcesilaus and Carneades in The Cambridge Companion to Ancient Scepticism.

Carneades, about 214 to 129 BCE in our file and 214 to 129/8 BCE in the encyclopedia, is where the Academic road gets practical, because he answers the question the Pyrrhonists answer with appearances: how does anyone act without knowing. His answer is to pithanon, the persuasive, and our file lists three grades of it: probable; probable and uncontradicted; probable, uncontradicted and thoroughly examined. The Stanford Encyclopedia gives the same three grades with a component our file drops, and the component is the whole point. Each grade is indexed to how much is at stake. In its words: "a three-stage criterion: (1) In matters of importance, skeptics adhere to the persuasive. (2) In matters of greater importance, they adhere to the persuasive and undiverted. (3) In matters that contribute to happiness, skeptics adhere to persuasive, undiverted, and thoroughly explored impressions." Without the stakes index the three grades read as an arbitrary ladder. With it they are a rule about how much checking a decision deserves, which is both what Carneades meant and something a modern reader already does.

Now the difference between the two schools, and one honest gap in what could be established for this article. Our file draws the line like this: the Academics made a positive claim, that nothing can be known, and offered probability as a guide, while the Pyrrhonists refused even that claim and suspended judgement about whether knowledge is possible. That is the standard textbook framing and it is worth having. What has to travel with it is that the Stanford Encyclopedia's entry on Ancient Skepticism, asked directly for a sentence drawing that comparison, does not draw it there. What it says is: "When comparing Pyrrhonian and Academic skepticism, two topics stand out: Pyrrhonism aims at tranquility; and it assigns pride of place to appearances." So the reference literature this article consulted puts the line somewhere else, and better evidenced. Both are carried here, and whether the Academics were really dogmatists about the impossibility of knowledge is itself a live scholarly question rather than a settled fact. Note in passing that the second of those two topics is what section 05 was about.

Ancient scepticism has a toolkit, and it comes in two lists. The older one is the ten modes of Aenesidemus, first century BCE, which the Stanford Encyclopedia's entry on Sextus records as ascribed to older sceptics and identified with Aenesidemus. Our file enumerates all ten, and the useful thing is not the enumeration but the fact that they are one argument repeated. Different animals perceive the same thing differently. So do different people. So do different senses in the same person. So does the same person healthy or ill, waking or asleep, young or old. Position, distance and place change how a thing appears; nothing is ever perceived in isolation but always mixed with other things; quantity and composition alter properties; frequency alters judgement, which is why a comet arrests attention and the sun does not; and customs, laws and beliefs conflict. Underneath all ten is a single move: every appearance is an appearance to someone, from somewhere, under conditions. Three of the modes point straight at the bent oar and the optical illusion. Those belong to a philosophy of perception article that has not been built yet, so they are named here as modes and not developed as illusions.

The second list is shorter, later and much harder to answer. Agrippa's five modes, as our file gives them, are disagreement, infinite regress, relativity, hypothesis and circularity. Stefan Sienkiewicz's Oxford monograph, devoted to exactly these five as they appear in Sextus, names the same set in slightly different words, as the modes of disagreement, hypothesis, infinite regression, reciprocity and relativity. About the man himself almost nothing is settled, and our file's question mark understates the problem. It writes Agrippa with a bracketed first century CE and a question mark. Two entries of the Stanford Encyclopedia disagree with each other. The entry on Sextus Empiricus gives "Agrippa (fl. first century BCE; see Schofield 2007; Hankinson 2010; Polito 2014), who probably lived at the end of the first century BCE." The entry on Ancient Skepticism dates him instead to the "1st to 2nd century CE". Those are two different centuries and two different eras, printed by the same reference work. This article carries both and picks neither.

One loose end, named so that it does not get tidied wrongly. Our file writes the Agrippan trilemma and a second name for it side by side, as though the two names were interchangeable and equally old. They are not. The Agrippan modes are ancient; the other name is a twentieth-century coinage, and our file gives no source for it. No source was resolved for this article either, and neither who coined it nor when could be established. So the ancient name is used throughout this page, and all that is said about the other is that the same argument circulates under a second and much later name. Nobody is credited with inventing it here and no year is attached to it, because this article cannot support either.

Three of Agrippa's five modes combine into the hardest argument in this article, and it is the reason ancient scepticism is not a historical curiosity. Take any belief and ask what justifies it. Whatever the answer, the same question can be asked of that. Our file states the result: the attempt leads to infinite regress, or to an arbitrary stopping point, or to circularity, none of which is satisfactory, therefore justification fails, therefore suspend judgement. The file also makes a point worth crediting, because the popular versions lose it: the famous trilemma is drawn out of a pentad, three horns extracted from five modes. Sienkiewicz's monograph is devoted to the five as Sextus presents them and names them in the same terms. Here is why this is a live argument rather than an antique one. Modern epistemology has not disposed of the trilemma; it has distributed itself across the horns. Foundationalism accepts the second horn and argues that some beliefs are basic and need no further justification. Coherentism accepts the third and argues that circles of the right size and shape are not vicious. Infinitism accepts the first and argues that an endless chain of reasons is exactly what justification is. Each is a school with holders, each pays a price it acknowledges, and the argument they are all answering is about two thousand years old. What justified true belief does with this, and where the analysis of knowledge breaks, belongs to an article on knowledge that has not been built yet, so it is not run here.

The doubt in the First Meditation is built as a ladder, and each rung takes back something the rung below left standing. Our file gives three waves: the senses sometimes deceive; I cannot tell dreaming from waking; and an all-powerful deceiver could make even my beliefs about mathematics and logic false. The Stanford Encyclopedia's entry on Descartes' Epistemology splits the middle wave into two, which is sharper and is what this article uses. The now dreaming doubt undermines the judgement that I am presently awake. The always dreaming doubt undermines the claim that waking experience comes from external objects at all. Then comes the last rung, in the Cottingham translation as that entry prints it: "I will suppose therefore that not God, who is supremely good and the source of truth, but rather some malicious demon [mauvais genie] of the utmost power and cunning has employed all his energies in order to deceive me." The bracketed French is the source's own interpolation and is reproduced rather than quietly removed. The escalation is the argument. The demon comes last because it reaches the truths that survive dreaming, which is to say arithmetic, and nothing in the First Meditation reaches further than that.

Descartes is not a sceptic, and our file says so plainly, which is the one place it is sharper than the popular literature. It states that he uses scepticism as a tool to overcome scepticism by establishing certain foundations, and that the cogito is the Archimedean point from which knowledge gets rebuilt. Section 01 has the external side of the same point, the encyclopedia quoting him against sceptics who doubt only for the sake of doubting and describing his aim as reaching certainty. The corollary is the thing not to skip. Descartes matters to a history of scepticism not because he was one, but because he built the strongest sceptical hypothesis anyone had built in order to knock it down, and the hypothesis outlived the knocking down. Section 12 is what happened to it.

Our own research file attaches the formulation I think, therefore I am to the Meditations of 1641. It does this twice, once in its summary and once under a heading about that book, and it is a document whose header declares itself Tier 1. The Stanford Encyclopedia's entry on Descartes' Epistemology states in terms that the canonical formulation is "a formulation that does not expressly appear in the Meditations". What is in the Meditations, in the Second Meditation, is a different and stranger sentence, which that entry prints in Cottingham's translation: "I am, I exist, is necessarily true whenever it is put forward by me or conceived in my mind." Read what that actually claims. It is not a standing truth that sits there being true; it is true whenever it is put forward, which makes it a claim about an event. The famous form lives in two other books, and both were read for this article in John Veitch's translations at Project Gutenberg. In the Discourse on the Method, Part IV: "as I observed that this truth, I think, therefore I am (COGITO ERGO SUM), was so certain and of such evidence that no ground of doubt, however extravagant, could be alleged by the sceptics capable of shaking it, I concluded that I might, without scruple, accept it as the first principle of the philosophy of which I was in search." And in the Principles of Philosophy, Part I article VII: "Accordingly, the knowledge, I THINK, THEREFORE I AM, is the first and most certain that occurs to one who philosophizes orderly." The capitals in both are Veitch's typography and are reproduced rather than normalised. Two honest limits belong on this page. Veitch's are nineteenth-century English translations hosted by a primary-source archive, not publisher editions and not facsimiles of the French of 1637 or the Latin of 1644, and the quotations are offered on that basis with his name attached each time. And no Latin text was read for this article, so the Latin formulations are reported as formulations here and never printed as quotations.

It is tempting to file that as a pedantic catch, so here is why it is not. The argument and the slogan are in different books four years apart, and they do different work. The Discourse offers the formulation as a first principle to build on, which is the form that circulates. The Meditations offers something with a time index in it, true whenever it is put forward, which is a claim about the act of thinking rather than about a standing fact, and which is the version that actually survives the demon of section 09. A reader who only knows the slogan has the conclusion without the strange part. Our file's own error is the ordinary one, and it is repeated in a great deal of published writing about Descartes; the reason to correct it on the page rather than quietly is that this article's whole subject is what happens to a claim nobody checks.

The next argument is not about deception at all. It is about tomorrow. David Hume's problem of induction, as our file states it, is that the inference from what has been observed to what has not cannot be rationally justified: the sun has risen every morning, which does not prove it will rise again, and the assumption that nature is uniform cannot itself be justified without circularity, because the only evidence for it is past experience. The Stanford Encyclopedia's entry on The Problem of Induction locates the argument in A Treatise of Human Nature of 1739, Book 1 part iii section 6, and in An Enquiry concerning Human Understanding of 1748, Section iv, and quotes Hume's own statement of the circle. Such reasoning, he says, must "proceed upon the supposition, that the future will be conformable to the past", and any probable argument for that supposition would involve "taking that for granted, which is the very point in question". At Treatise 1.3.6.4 the entry prints the principle in his own eighteenth-century spelling: "if Reason determin'd us, it would proceed upon that principle that instances, of which we have had no experience, must resemble those, of which we have had experience". Our file names only the Enquiry, and the Treatise passage is the one the literature cites, so both are given here. Notice what has just happened structurally. That is the Agrippan circularity mode, arriving again seventeen centuries later, in a form nobody has answered.

Hume's account of what we do instead is not a justification and he does not offer it as one. Our file puts it as custom and habit: inductive reasoning rests on those rather than on rational demonstration, we cannot help believing the future will resemble the past, and that is a psychological fact about human beings rather than a reason. The encyclopedia supplies his own passage at Enquiry 5.1.8, where the mechanism is named with two examples: "having found, in many instances, that any two kinds of objects, flame and heat, snow and cold, have always been conjoined together; if flame or snow be presented anew to the senses, the mind is carried by custom to expect heat or cold, and to believe". The source emphasises the words custom and believe with bolding and italics; the words are reproduced here and the emphasis is not. Hold the distinction, because it is the shape of this whole article in miniature. Hume is not saying induction is fine because we are built to do it. He is saying the question of whether it is rational was answered no, and that we go on doing it anyway.

That leaves the only livable position this article is going to reach, and our file names it. It distinguishes destructive or excessive scepticism, which paralyses belief and action, from mitigated scepticism: a modesty about the limits of reason, and a willingness to proportion belief to evidence. The file adds that Hume is a practical person who thinks and acts, but does so recognising that certainty is unavailable. One phrase in the file's framing is not carried here. It glosses mitigated scepticism as including an awareness of cognitive biases, which is a modern category and not Hume's, and it is not attributed to him on this page. What can be said is that mitigated scepticism is the position this article can honestly recommend, and also that it is not an answer to the trilemma in section 08 and does not pretend to be. It is a way of living with the trilemma unanswered. The obvious next move after Hume, that the world in itself is permanently sealed off behind the mind's own machinery, is Kant's, and it belongs to an article in this wing that has not been built. This article stops here.

Hume's other scepticisms belong to other articles and one of them is already live. Our file notes his scepticism about causation, that we never perceive a necessary connection between cause and effect but only constant conjunction and habitual expectation, which is the induction argument wearing a different hat and gets no more than this clause here. It also notes his scepticism about personal identity, and that is Personal Identity's subject in this same wing, where the bundle theory of the self is argued properly. Our file prints a short phrase from Hume inside quotation marks in that passage with no source attached; it was not verified for this article and it is therefore not quoted here, even though the neighbouring article may carry it on its own evidence. A quotation licence is not transferable between articles, which is a small rule doing a lot of work on a subject like this one.

Our file makes a claim about what happened to Descartes's demon after Descartes, and it is the bridge from the seventeenth century to the twentieth: the evil demon hypothesis became the template for modern sceptical scenarios. That is a claim about influence, no independent corroboration of it was obtained for this article, and it is carried here as our file's assessment rather than as established intellectual history. What can be stated flatly is narrower and checkable. On another wing of this Athenaeum, Are We Living in a Simulation? reaches for exactly that comparison in its own words, calling the simulation hypothesis effectively unfalsifiable in the way that Descartes's demon is. So the template claim is at least visibly true of one adjacent argument, which is less than our file asserts and more than nothing.

That table is reasoning rather than a claim from any research file, and it is marked as such. Our file's phrase, the modern evil demon, is a good headline and a slightly misleading equation: three hundred and forty years separate the two, and they do not claim the same thing. One further neighbour has to be named and then left alone. Nick Bostrom's simulation argument is a probabilistic argument concluding a disjunction about ancestor simulations, which is a different kind of argument again from either of these, and it belongs to Are We Living in a Simulation? on another wing, where it is carried in full. Nothing from it is imported here.

Putnam's brain-in-a-vat argument is an attempt to refute the scenario, not a statement of it, and a great deal of retelling has this exactly backwards. Our own file gets it right, which deserves crediting: it labels the passage Putnam's refutation and explains the mechanism correctly. The mechanism runs on a theory about what words refer to. If a speaker has always been a brain in a vat, then that speaker's words brain and vat do not refer to real brains and real vats, because those are not the things the speaker has ever been in contact with; they refer to simulated brains and simulated vats. So when a vat-dweller thinks the thought I am a brain in a vat, the thought is not about brains and vats at all, and on Putnam's account it comes out false. The Stanford Encyclopedia's entry on Content Externalism and Skepticism, whose author it returns as Gary Ebbs, prints his own words at page 15 of the book. In its rendering, with the ellipsis the source itself carries and with the inner quotation marks set as single here so that the nesting reads, where the source has them double: "But part of the hypothesis that we are brains in a vat is that we aren't brains in a vat in the image ... So, if we are brains in a vat, then the sentence, 'We are brains in a vat' says something false". The encyclopedia's own word for the conclusion is self-refuting. Never write that Putnam proposed or invented the scenario, and never write that he argued we might be brains in vats. He argued the reverse, from a theory of reference, and part of the argument's fame is how many people remember it the wrong way round.

Whether the argument works is another matter, and here this article says less than our file does. Our file writes that the refutation is controversial and that many philosophers find it unconvincing. The evidence gathered for this article supports the first half and not the second. The Stanford Encyclopedia entry on Content Externalism and Skepticism, asked directly whether the argument is widely rejected, does not say so; what it records is substantial ongoing philosophical work on how to reconstruct and evaluate it, which is engagement rather than dismissal. Independently, an entire Cambridge volume exists on the question, and Tony Brueckner's chapter Putnam on brains in a vat is cited below. So the honest statement is that the argument is contested and that philosophers continue to disagree about how to reconstruct it. No headcount is offered here, because none was taken. The difference between debated and refuted is the difference this whole article is trying to keep visible.

On its own the vat is a curiosity. Bolted to one plausible principle it becomes the central sceptical argument of the last fifty years, and the principle is closure. Our file gives the argument in its knowledge form: if a person knows P, and knows that P implies Q, then that person knows Q. Apply it: a person claims to know they have hands; having hands implies they are not a handless brain in a vat; they do not know they are not a brain in a vat; therefore, by contraposition, they do not know they have hands. The Stanford Encyclopedia's entry on Skepticism gives the same shape formally, but about justification rather than knowledge, writing the principle as "For all propositions p and q, if p entails q, and Jp, then Jq" and the argument as "CP1. If I am justified in believing that h, then I am justified in believing that ~SH. CP2. I am not justified in believing that ~SH. Therefore, I am not justified in believing that h." Knowledge and justification are not interchangeable, and sliding between them mid-argument is the commonest way this gets mangled, so this article runs the knowledge version and names the other rather than blending them.

Our file names three standard responses, and each costs something. Robert Nozick's tracking theory denies closure: knowledge is a matter of a belief tracking the truth, and tracking does not automatically pass along entailment, so a person can know they have hands without knowing they are not a vat-brain. The cost is closure itself, which is about as intuitive as principles get. Contextualism says the standards for the word know shift with context, so in an ordinary conversation a person knows they have hands and in a philosophy seminar they do not, and both are true; Keith DeRose's Solving the Skeptical Problem, in The Philosophical Review in 1995, is cited below as the anchor for it. The cost is a stable meaning for the word know. The third response our file names is pragmatic rejection of the scenario, for which it names no source and none was sought here. One citation note, kept in the open: DeRose's paper resolved exactly and is linked below, and no usable identifier for Nozick's Philosophical Explanations, Harvard University Press, 1981, was obtained, so it is cited by author, title, publisher and year with no link beside it. None of the three has settled the matter, and none of them is presented here as having done so.

One further contemporary line has to be named and then left, and the reason for leaving it is worth stating. Our file's summary lists three contemporary sceptical arguments: the brain in a vat, closure, and the underdetermination of theory by evidence. The first two get their own sections in it. The third gets nothing at all: no section, no explanation, no source. Its own cross-reference index points at a philosophy of science document for that material, and the article built from that document does not exist yet. So underdetermination is named here as a third line of sceptical argument, is placed where it belongs, in the philosophy of science, and is not developed, because this article has nothing to develop it from.

The same argument shape reappears in twenty-first-century meta-ethics, and one detail of how it reappears is worth stopping on. Our file defines moral scepticism as doubt about the existence of moral knowledge, moral facts or moral truth, distinguishes it from amoralism, which is simply not caring about morality, and lists three varieties: error theory, non-cognitivism, and moral epistemic scepticism. The Stanford Encyclopedia's entry on Moral Skepticism, by Walter Sinnott-Armstrong, confirms the characterisation of error theory, on which all substantive moral assertions and beliefs are false, or neither true nor false, "if they claim (or semantically presuppose) that moral facts or properties exist". It cites the book behind that position as Mackie, J. L., 1977, Ethics: Inventing Right and Wrong, New York: Penguin; no usable identifier for it was obtained for this article, so it is cited exactly as that entry cites it and the entry is named as the source of the citation. Now the structural point. The entry splits each variety in two: "Dogmatic skepticism about moral knowledge = nobody ever knows that any substantive moral belief is true. Dogmatic skepticism about justified moral belief = nobody is ever justified in holding any substantive moral belief. Pyrrhonian skepticism about moral knowledge withholds assent from both dogmatic skepticism about moral knowledge and its denial. Pyrrhonian skepticism about justified moral belief withholds assent from both dogmatic skepticism about justified moral belief and its denial." That is the distinction from section 01 and section 06, the one between denying and suspending, arriving in contemporary ethics with the ancient names still attached. What follows from any of this for how anyone should actually behave is a different subject and is not argued here.

The first objection is that the position eats itself. Our file states it fairly: scepticism looks self-refuting, because to say I know that I know nothing is a knowledge claim, and to say no belief is justified is itself a belief requiring justification. And it gives the Pyrrhonist reply, which is sound and which the article has already assembled the parts for. The Pyrrhonist does not assert that nothing can be known. He reports his own state of mind, that he has been unable to decide, and his scepticism is an activity, investigating and suspending, rather than a doctrine. That is the same point as the definition at Outlines of Pyrrhonism I 8: an ability, not a thesis. Our file closes that section by printing a sentence inside quotation marks, attributed to Sextus, to the effect that the sceptic does not dogmatise, not even about the impossibility of dogmatism. No locus is given for it, and it could not be located in any translation consulted for this article, so it is not quoted here and it is not paraphrased as his wording either. The licensed passage at I 8 makes the same point better and comes with its locus attached.

The harder objection is practical and it has a name, the apraxia charge. If judgement is suspended about everything, our file states, action becomes impossible: there is no deciding to eat, and no deciding to step back from a cliff. The standard reply is the fourfold observance of section 05: the Pyrrhonist follows appearances, custom, natural drives and craft expertise, living practically while suspending judgement on theoretical matters, so the charge confuses theoretical commitment with practical engagement. Our file leaves it there, as though the reply closed the question. It does not. This is one of the liveliest questions in the field and it has a canonical title: the chapter Can the sceptic live his scepticism?, in the Cambridge volume Explorations in Ancient and Modern Philosophy of 2012, is cited below, and so is Casey Perin's chapter Scepticism and belief. The chapter record carries no author of its own; the containing volume resolves to M. F. Burnyeat, so it is described here as a chapter in his volume rather than credited to him as a byline this article did not read. Whether the reply works is exactly what the literature is still arguing about, and this page does not settle it.

Our file makes a case for scepticism's productive role, and the case is good even where the wording needs handling. It says scepticism forced the development of better epistemological theories, naming foundationalism, coherentism, reliabilism and contextualism; that it motivated the scientific method, as empirical testing in response to theoretical uncertainty; and that it remains, in the file's own words, the most powerful tool for testing the foundations of knowledge. That last phrase is our file's superlative, no external source was found for it, and it is attributed here rather than adopted. The stronger version of the same point does not need a superlative at all, and section 08 already built it. Three of the four theories our file names are the three horns of the Agrippan trilemma with modern arguments attached: foundationalism takes the arbitrary stop and defends it, coherentism takes the circle and defends it, contextualism changes what the word know is asking for. The theories exist in the shape they do because of the argument they are answering.

The most honest sentence in our file sits right beside that one, and it pulls the other way. Scepticism, it says, has never been a popular position, and few philosophers are consistent sceptics. Independent support for the substance, though not for any count, comes from the Stanford Encyclopedia's entry on Skepticism, which notes that contemporary philosophers find Pyrrhonian scepticism interesting "not because they take seriously the possibility of its truth, but rather because there are interesting arguments in its favor". That adds a reason to the observation: the arguments are studied by people who do not hold the position. What is not said here, in either direction, is a number. Nobody in the research behind this article counted philosophers, so this page says that scepticism has few adherents and that its arguments are worked on by people who reject its conclusion, and it does not say almost none, or nobody today, or give a figure.

No, this is not a settled field with no active dispute, and our own research file says that it is. Its closing block asserts that no significant counter-arguments exist in the scholarly literature for the core claims in the document, and that the material represents established philosophical consensus with no active scholarly dispute over the fundamental claims presented. Section 8 of that same document is titled Counter-Arguments and Critical Assessment and runs three of them. Beyond its own internal contradiction, this article has documented live disputes on five separate points: whether Pyrrho's position was metaphysical or epistemological, when Sextus lived, when Agrippa lived, on which two entries of the same encyclopedia disagree with each other, whether Putnam's argument works, and whether the sceptic can live his scepticism. That closing block is boilerplate, it is false on the document's own evidence, and nothing in this article's verdict rests on it.