Four centuries of objections to the method of doubt and the evil genius — sorted by what each one actually attacks
The chat asked one question: what exactly is wrong, if anything, with hyperbolic doubt, and in how many logically distinct ways has it been said? Its answer is a taxonomy. There are nine types of objection, from the self-refutation charge of 1641 to the New Evil Demon of 1983. Once they are sorted by logical form, they reduce to three master-moves plus one family aimed elsewhere. The doubt borrows what it doubts; the doubt destroys the meanings it needs; the doubt is not real doubt. The family aimed elsewhere, the Cartesian Circle, attacks not the doubt but Descartes’ way out of it.
One distinction organises everything else. An objection to the doubt asks whether hyperbolic doubt is coherent, motivated and survivable. An objection to the answer asks whether the God who ends the doubt is proved without circularity. Arnauld’s Circle presupposes that the doubt is coherent. Bouwsma’s and Putnam’s arguments deny that it is, and say nothing about the Circle. Most textbook confusion about “the problem with Descartes’ doubt” comes from treating these two as one complaint. Huet is the only objector who genuinely makes both.
What this review adds to the chat. The chat’s report was written by a web-research agent. Here, its quotations are checked against the books in your folder. Most survive. Five do not survive as reported, and they are corrected in §14. The most consequential correction: the charge that treating the doubtful as false is “more like adopting a new prejudice than relinquishing an old one” is Gassendi’s (Fifth Objections, CSM 2:180), not Bourdin’s. And Bouwsma never wrote the two sentences the report puts in his mouth.
Descartes’ First Meditation does not contain one argument. It is a nested sequence, with a methodological policy laid over it. An objection is only as good as its aim: it must say which component it attacks.
The question asked: how many logically distinct objections are there, who makes each one, and what is the best reply to each? The brief demanded a taxonomy, not a list, “so the reader sees which critics are making the same underlying move in different vocabularies.”
The question refused: whether Descartes is right to doubt. The chat takes no side on skepticism itself. It classifies objections by logical form and records, for each, the “state of play.” Where it concludes, its verdict is deliberately negative about the refutations: the doubt has not been refuted; it has been contained, relabelled and diagnosed.
The chat’s report is the spine. The primary texts and the objectors held in your folder are listed here because the review quotes them directly. The rest of the report’s roster appears in the sources at the foot of the page.
| Work | What it contributes here | Enters at |
|---|---|---|
| Chat report, “What Is Wrong with Hyperbolic Doubt? A Taxonomy of Objections to Descartes’ Method of Doubt and the Evil-Genius Hypothesis, 1641–2026” (DEscartes project, 22 July 2026). A PDF copy is in the folder. | The nine types, the dossiers, the matrix, and the doubt/answer distinction. | throughout |
| Descartes, Meditations with Objections and Replies, CSM vol. 2 (Cambridge, 1984) | The target in Med. I, and Descartes’ own answers to Types 1, 3, 5 and 6, all given in 1641–42. | §§01, 03–08 |
| Richard Kennington, “The Finitude of Descartes’ Evil Genius,” JHI 32 (1971), 441–446 | Argues the demon is a finite deceiver, distinct from the omnipotent God. This decides which beliefs are in doubt. | §02 |
| O. K. Bouwsma, “Descartes’ Evil Genius,” Philosophical Review 58 (1949), 141–151 | The locus classicus of meaning-parasitism: the two adventures of the evil genius, paper flowers and “cerpicio.” | §04 |
| Wittgenstein, On Certainty · Putnam, Reason, Truth and History (1981), ch. 1 · Brueckner, Essays on Skepticism (2010) | The grammatical and semantic members of the same family, and the anatomist of their limits. | §04 |
| C. S. Peirce, “Some Consequences of Four Incapacities” (1868), in The Essential Peirce, vol. 1 | The founding statement of the pragmatic charge. | §05 |
| Thomas M. Lennon, The Plain Truth: Descartes, Huet, and Skepticism (Brill, 2008) · Richard Popkin, The History of Scepticism (2003) | Huet as the objector who attacks both the doubt and the answer; the seventeenth-century narrative. | §§06–07 |
| Barry Stroud, The Significance of Philosophical Scepticism (1984) · Michael Williams, Unnatural Doubts (1991/1995) | The standing defence of the doubt’s depth, and the leading attempt to dissolve it. | §10 |
| Stewart Cohen, “Justification and Truth,” Philosophical Studies 46 (1984), 279–295 | The New Evil Demon, in which the demon becomes an instrument against reliabilism. | §12 |
The brief (RESEARCH_PROMPT_Objectors_to_Cartesian_Doubt_Review.md) is in the folder. A second copy of the same report, with a different title, is _A Comprehensive Review of the Objections and Objectors…pdf.
Read this once; everything after it is elaboration.
There are nine objection-types and seven possible targets, and they do not line up. Each type aims at a particular component of the First Meditation, and an objection that succeeds against one component can leave the others untouched. The table gives the pairing the chat establishes.
| Type | Component attacked | Canonical objectors | Best reply | State of play |
|---|---|---|---|---|
| 1 · Self-refutationborrows what it doubts | (v) methodic denial; the use of reason while doubting | Gassendi, Hobbes, Bourdin, Huet; Dicker’s puzzle | Doubting a proposition’s deliverance is not abandoning a faculty’s use | Answerable; Dicker’s puzzle lingers |
| 2 · Meaning-parasitismdestroys its meanings | (vi) universality | Gassendi, Ryle, Austin, Bouwsma, Wittgenstein, Putnam, Davidson | Stroud, Brueckner: it yields unassertibility, not falsity | Migrated into content-externalism |
| 3 · Pragmaticnot real doubt | (v), (vii) motive and function | Gassendi, Hume, Peirce | Theoretical, not practical, doubt (Descartes CSM 2:270; Broughton) | Deflected |
| 4 · Not far enoughthe skeptic’s complaint | (vii) where the doubt stops | Huet, Foucher; Popkin’s lineage | The cogito is intuited in one act | Under-appreciated, internally powerful |
| 5 · The cogito failsthe survivor doesn’t survive | the first certainty | Gassendi, Huet, Hobbes, Lichtenberg, Leibniz | “Not a syllogism” (CSM 2:100) | Lichtenberg survives as a constraint |
| 6 · The Circlethe answer fails | the exit, not the doubt | Mersenne, Arnauld; Van Cleve, Della Rocca | The memory response (CSM 2:100); particular vs general | Escapable, at a cost |
| 7 · Diagnosticoptional picture | (vi), (vii) the doubt’s naturalness | Carnap, M. Williams, Rorty, therapeutic Wittgenstein | Stroud: diagnosis is not refutation | Dominant, contested |
| 8 · Metaphysical defusal | (iii)/(iv) the scenario’s being skeptical | Chalmers | Only for lifelong envatment; relabels the loss | Optimistic, revisionary |
| 9 · New Evil Demonthe inversion | not Descartes at all: reliabilism | Lehrer & Cohen, Cohen 1984 | Excuse, not justification (Williamson, Littlejohn) | Unresolved, reoriented |
The dispute over Cartesian doubt is not one dispute. The report’s key sentence is this: “Arnauld’s Circle is an objection to the answer and leaves the doubt untouched; Bouwsma’s and Putnam’s arguments attack the coherence of the scenario and are silent on the Circle.” Whole disputes turn out to be critics talking past each other because they aim at different components.
Once that is visible, the nine types stop looking like a list and start looking like a map of targets. Types 1–3 and 7 aim at the doubt’s coherence and motive. Type 4 aims at where it stops. Type 5 aims at what survives it. Type 6 aims at the exit. Types 8 and 9 no longer aim at Descartes at all: they reuse his scenario for other purposes.
Before any objection is heard, the target has to be fixed. The chat splits the First Meditation into seven separately attackable components.
A good objection hits exactly one component. The most consequential fork is whether the demon is finite or infinite, because that decides whether mathematics is in doubt.
It is easy to read the objectors as quarrelling over one argument. They are not. The First Meditation is a staircase of escalating doubts, plus a policy (treat the doubtful as false) and a purpose (clear the ground for a foundation). Each can be attacked on its own.
The chat’s first move, before any objector is heard, is to number the parts. That numbering is what makes the rest of the page legible.
The chat distinguishes seven components. The first four are the escalating grounds of doubt. The last three belong to the method.
“But from time to time I have found that the senses deceive, and it is prudent never to trust completely those who have deceived us even once.”
AT 7:18 · CSM 2:12 checked
The whole method in miniature. The inference from occasionally to never completely is precisely what Bourdin will call throwing out the whole basket because some apples are bad.
“For whether I am awake or asleep, two and three added together are five, and a square has no more than four sides.”
AT 7:20 · CSM 2:14 checked
Component (ii) has an explicit limit. Any objection to “the dreaming argument” that treats it as reaching mathematics is aimed at the wrong component.
“And yet firmly rooted in my mind is the long-standing opinion that there is an omnipotent God who made me the kind of creature that I am. How do I know that he has not brought it about that … I too go wrong every time I add two and three or count the sides of a square …?”
AT 7:21 · CSM 2:14 checked
“…I think it will be a good plan to turn my will in completely the opposite direction and deceive myself, by pretending for a time that these former opinions are utterly false and imaginary. … I will suppose therefore that not God, who is supremely good and the source of truth, but rather some malicious demon of the utmost power and cunning has employed all his energies in order to deceive me.”
AT 7:22 · CSM 2:15 checked
Components (iv) and (v) sit in the same paragraph. Denial is announced as a plan, and the demon arrives as its instrument.
“…he begins by digging out a set of trenches from which he removes the sand, and anything resting on or mixed in with the sand, so that he can lay his foundations on firm soil. In the same way, I began by taking everything that was doubtful and throwing it out, like sand…”
AT 7:536–7 · CSM 2:366 checked
Component (vii) in Descartes’ own image. The doubt is excavation, not foundation. On this reading the objections of Types 1 and 3 aim at the digging, while Type 6 aims at what is built afterwards.
The chat flags a quarrel it does not resolve: is the demon of T4 simply the deceiving God of T3 under a less offensive name, or a weaker hypothesis? Kennington begins by naming the view he opposes, and his statement of it is exact:
…is replaced by the supposition of an evil genius. Yet God and the evil genius are both omnipotent: this view is unanimously held, so far as we have been able to ascertain, by the distinguished… Kennington, “The Finitude of Descartes’ Evil Genius” p. 441 checked
Against it he argues from the text. The demon is never called omnipotent. The list of what it renders doubtful (sky, air, earth, colours, shapes, sounds, “all external things”) omits the mathematics that the deceiving-God passage had reached. Then comes the step that makes the finitude thesis matter:
Only if the evil genius is of finite power can the doubt or the cogito be any sort of argument whatever. Kennington p. 445 checked
On the other side the chat places Anthony Kenny, for whom the substitution is made “simply because it is less offensive and less patently incoherent,” and “the content of the two hypotheses is the same.” Bernard Williams and Alan Musgrave are placed there too. not held The Kenny wording comes from secondary citation, as the chat itself admits; Kenny’s Descartes (1968) is not in the folder.
That matters for four of the nine types. Dicker’s puzzle (Type 1: why is logic exempt when mathematics is not?) changes shape. Bouwsma’s concession of “four or five” clear and distinct ideas (Type 2) becomes a reading rather than a dodge. Huet’s complaint that the doubt did not go far enough (Type 4) gains force. And what the cogito must survive (Type 5) shrinks. The chat records the dispute and declines to adjudicate it: “it turns on whether the demon-passage’s omission of mathematics is deliberate (Kennington) or incidental (Kenny).”
Three objection-types say the doubt is defective as doubt: self-undermining, meaningless when totalised, or feigned.
All three were raised within months of publication, and Descartes answered each. The answers work against the objections in the form first given. Only the meaning-parasitism family acquired, after 1949, a form Descartes never saw.
Borrowing, parasitism, pretence. These three arguments recur, under different names, from 1641 to 1986. The chat’s central structural claim is that the second one is a single argument in six vocabularies, sharing a single limit.
Each section below gives the canonical formulation, the roster, the best reply and the state of play, the four heads the brief required. The quotations have been checked against the folder.
The doubt presupposes the logic, language and memory it purports to suspend.
The chat’s canonical formulation: “To doubt p one must understand p, retain it, and reason about it — capacities the doubt has allegedly bracketed.” Its roster starts with the Jesuit Pierre Bourdin. On inspection the sharpest early statement of the charge belongs to someone else.
“Why instead did you consider everything as false, which seems more like adopting a new prejudice than relinquishing an old one?”
AT 7:257–8 · CSM 2:180, “On the First Meditation” checked; Descartes quotes it back at CSM 2:242
Correction. The chat attributes this sentence to Bourdin, citing Seventh Objections AT 7:454ff and CSM 2:305ff, and gives it as “acquiring a new prejudice.” In your CSM it is Gassendi’s, in the Fifth Objections, and the word is “adopting.” Bourdin’s attack on methodic denial is real but different in kind: it is the long burlesque answered by the basket of apples (CSM 2:324) and the architect (CSM 2:366).
“Now, what shall we say if it turns out that reasoning is simply the joining together and linking of names or labels by means of the verb ‘is’?”
AT 7:178 · CSM 2:125 checked — the chat’s version (“reasoning will be nothing but the joining together…”) is a paraphrase
If reasoning is only the linking of names, the meditator’s reasoning is at the mercy of conventions that the doubt never examined.
Who doubts that a Frenchman and a German can reason about the same things, despite the fact that the words that they think of are completely different? Descartes, Reply to Hobbes · AT 7:178–9 · CSM 2:126 checked
Two further members complete the roster. Huet gives the charge its most dangerous form, a criterion-circle: reason cannot be certified by reason. Maia Neto’s study lists among Huet’s arguments “the circularity involved in attempting to prove by reason that reason is certain” checked (Maia Neto 2022, ch. on the Censura). Dicker’s puzzle is the modern exegetical version: the Meditations go on using logic while mathematics is under demonic doubt. Why is logic exempt?
Hyperbolic doubt is an attitude taken toward propositions when they are not being attended to. It is not a suspension of the faculties that entertain them. The objection equivocates between doubting logic’s deliverances and abandoning logic’s use. The chat cites Frankfurt (1970) and Newman’s SEP entry for this; Frankfurt’s chapter “Mathematics and the Omnipotent Deceiver” is in the folder (ch. 8, p. 93 checked).
State of play. Answerable in the form Gassendi and Hobbes gave it. Dicker’s puzzle remains a live irritant. The usual resolution is that logical laws fall under the deceiving-God doubt, not the demon doubt, and are restored only with God. That resolution presupposes Kennington’s side of §02.
“False” needs “true”; totalise the contrast term and it goes empty.
This is the chat’s largest family and its best structural result. Gassendi’s remark that the false must be compared with an instance of truth, Ryle’s counterfeit coin, Austin’s trouser-words, Bouwsma’s paper flowers, Wittgenstein’s hinges, Putnam’s vat-brains and Davidson’s charity are, the chat says, “one argument in six vocabularies.”
Bouwsma’s evil genius tells two stories. In the first he turns everything into paper, and Tom sees through it: the illusion is detected because Tom knows the difference between flowers and paper. In the second the demon makes an illusion no sense can detect, and Tom answers that the demon has simply changed the meaning of the words.
“That the paper flowers are illusory is revealed by the recognition that they are paper. As soon as Tom realizes that though they look like flowers but are paper, he is acquainted with, sees through the illusion, and is not deceived.”
Bouwsma, “Descartes’ Evil Genius” p. 145 checked
“…But there still are what we mean by flowers. If your intention was to deceive, you must learn the language of those you are to deceive.”
Bouwsma p. 150 checked
This is the thesis, in Tom’s mouth. An undetectable “illusion” is not an illusion in our sense, so the demon has changed the subject rather than deceived anyone.
The chat quotes Bouwsma as saying “an illusion requires the possibility of discovery” and “if the evil demon fashions an illusion that we can never know about, it is no illusion at all, at least as we use the term.” Neither sentence occurs in the article in your folder. Both are fair summaries of the position, but they are paraphrase. The words on the page are T8 and T9. Nor does the demon make flowers of “paper, paste, and cerpicio.” Cerpicio is the extra sense he invents in the second adventure (“Esse est cerpici,” p. 150).
The chat’s other remark about Bouwsma is borne out word for word. His demon boasts “that apart from a few, four or five, clear and distinct ideas, he could deceive any son of Adam about anything” p. 141 checked. That is the concession critics use against him: he never engages the deceiving-God doubt of mathematics.
If you tried to doubt everything you would not get as far as doubting anything. The game of doubting itself presupposes certainty. Wittgenstein, On Certainty §115 checked
…'I do not exist' is self-refuting if thought by me (for any 'me'). So one can be certain that one oneself exists, if one thinks about it (as Descartes argued). What I shall show is that the supposition that we are brains in a vat has just this property. Putnam, Reason, Truth and History p. 8 checked — the argument is announced as “in a certain way, self-refuting” at p. 7
Putnam’s explicit model is the cogito itself. The vat argument claims for “I am a brain in a vat” the self-refuting structure that “I do not exist” has for Descartes. That makes Putnam a bridge between this page and Chat 4, where Hintikka treats “I do not exist” as existentially inconsistent.
This is Stroud’s line and Brueckner’s. In Brueckner’s “Semantic Answers to Skepticism,” the naive way of running Putnam’s argument “obviously begs the question against the skeptic” Essays on Skepticism, p. 148 checked. The chat’s further quotation of Brueckner, “the most the argument proves is that if we are brains in a vat, then the sentence … is false,” could not be found in the 2010 collection not found. It may be in the 1986 journal text, which is not held.
State of play. The family shows that the first-person expressibility of global doubt is problematic. It has not closed off the doubt, and the debate has moved into content-externalism.
Doubt adopted by fiat is pretence, and nothing follows from a pretence.
Gassendi’s jibes begin it: come down from the pretence, and why walk on the earth if you doubt it exists? Hume’s “incurable” antecedent skepticism continues it. Peirce states it in the form everyone now quotes:
“We cannot begin with complete doubt. We must begin with all the prejudices which we actually have when we enter upon the study of philosophy. … Hence this initial scepticism will be a mere self-deception, and not real doubt … Let us not pretend to doubt in philosophy what we do not doubt in our hearts.”
“Some Consequences of Four Incapacities,” The Essential Peirce vol. 1 pp. 28–29 checked
A small correction. The chat also gives Peirce’s definition of genuine doubt as “an uneasy and dissatisfied state” as if it came from the same 1868 paper. It comes from “The Fixation of Belief” (1877), EP1 p. 114 checked.
…before we can decide to doubt, we need some reason for doubting; and that is why in my First Meditation I put forward the principal reasons for doubt. Appendix to the Fifth Replies (to Clerselier) · AT 9A:204 · CSM 2:270 checked
The Burnyeat–Broughton line makes Cartesian doubt theoretical, not practical. Descartes never claims to believe in the demon or to change his conduct because of it. Broughton’s Descartes’s Method of Doubt is in the folder. State of play: the charge lands hard against a caricature (doubt by fiat) and is deflected by the theoretical reading. Whether theoretical doubt can deliver the psychological indubitability Descartes wants is still contested.
Two types come from the other direction: the doubt was not radical enough, and what it left standing does not stand.
Huet turns Descartes’ own doctrines against him, and the chat calls this the most under-appreciated family. Lichtenberg survives every reply as a limit on what the cogito delivers.
Every objection so far said the doubt goes too far. The Pyrrhonist says it did not go far enough, and that Descartes stopped doubting exactly where consistency demanded he continue.
This part is the hinge between this chat and the three that follow. Type 5 asks whether the cogito survives the doubt, which is the question of Chat 3, Chat 4 and Chat 5.
Lennon’s book opens by calling the Censura “the most comprehensive, unrelenting, and devastating critique of his philosophy ever published” p. viii checked. Huet’s arguments, as the chat assembles them, use Descartes against Descartes.
The famous phrase “he who thinks can both be and not be” is securely documented, the report says, only in the parallel Thomist critique of Antoine Goudin (1724). Huet’s own wording has to be checked against Lennon’s printed translation before it is quoted as his. Against Cartesian Philosophy is not in the folder, so the check cannot be made here not held. Popkin’s gloss (“I think, therefore perhaps I may be”) is likewise reported, not located.
Best reply, and state of play. If the cogito is grasped in a single act, the temporal gap closes (§07). On the created truths, doubt is psychologically impossible while attending, so the worry bites only retrospectively. The chat’s verdict is that Type 4 is “arguably the most powerful internal critique,” because every premise in it is Descartes’ own.
Four objections, and one reply that answers three of them.
| Objection | What it says | Source |
|---|---|---|
| The syllogism | “I think, therefore I am” needs the suppressed major “whatever thinks exists,” an unproved universal. So it either rests on something undoubted or begs the question. | Gassendi, Huet; the Angers inquisitors (Ariew 1999) |
| The temporal gap | Premise and conclusion are separated in time; memory is reintroduced. | Huet; also Hobbes, Gassendi |
| Lichtenberg | “It thinks, we should say, just as one says, it lightnings.” The “I” is smuggled in. | Waste Books K; Campbell 2012 — see Chat 5 |
| The corporeal thinker | From “I am thinking” it does not follow that I am a mind rather than something corporeal. | Hobbes, CSM 2:122–3 checked |
| Incompleteness | The first truths of fact are two: ego cogito and varia a me cogitantur. | Leibniz, Animadversiones on Pr. I.7 not held |
“…he does not deduce existence from thought by means of a syllogism, but recognizes it as something self-evident by a simple intuition of the mind. This is clear from the fact that if he were deducing it by means of a syllogism, he would have to have had previous knowledge of the major premiss ‘Everything which thinks is, or exists’…”
AT 7:140 · CSM 2:100 checked
If this holds, the syllogism charge fails, and so does the temporal gap, since a single intuition has no interval. Whether it holds is the whole subject of Chat 4 (Hintikka) and Chat 6 (Dicker, who reconstructs the cogito as an inference after all).
State of play. The question-begging charge is neutralised if the cogito is intuitive. Lichtenberg survives as a serious constraint on how much the cogito delivers: a thinking, not a thinker-substance. That is why the further step to res cogitans, Hobbes’s target, is more vulnerable than the bare “I exist.”
One objection-type concedes that the doubt is coherent and attacks the way out of it.
The Circle is escapable if particular clear and distinct perceptions certify themselves while attended to. The price is that God’s guarantee accomplishes less, and the general demon doubt may never be lifted.
The Circle is the most famous objection to Descartes and, strictly, no objection to his doubt at all. It grants the doubt everything and asks whether the proof of God that ends it is question-begging.
The chat’s most useful single sentence belongs here: “A reader who treats ‘the Cartesian Circle’ and ‘the doubt is self-refuting’ as the same complaint (a common textbook error) has conflated Type 6 with Type 1.”
“…how the author avoids reasoning in a circle when he says that we are sure that what we clearly and distinctly perceive is true only because God exists. But we can be sure that God exists only because we clearly and distinctly perceive this.”
AT 7:214 · CSM 2:150 checked
“…when I said that we can know nothing for certain until we are aware that God exists, I expressly declared that I was speaking only of knowledge of those conclusions which can be recalled when we are no longer attending to the arguments by means of which we deduced them.”
AT 7:140 · CSM 2:100 — Second Replies checked
Correction. The chat introduces this as “Descartes’ two moves in the Author’s Replies to the Fourth Objections,” while giving AT 7:140. The locator is right and the attribution wrong. The passage is in the reply to Mersenne, a year before Arnauld, and on the same page as T11. In the Fourth Replies Descartes refers Arnauld back to it (CSM 2:171).
The modern reconstructions follow Van Cleve (1979), who opens by observing that “in its generalized form, the Cartesian Circle is none other than the Problem of the Criterion.” Only the general rule that clear and distinct perceptions are true needs God; particular perceptions certify themselves while attended to. Della Rocca, Newman and Nelson, and Loeb debate the memory, psychological and validation-of-reason readings. not held None of these articles is in the folder.
State of play. The dominant modern verdict (Van Cleve, Newman, Della Rocca) is that the Circle can be escaped, at the cost of narrowing what God’s guarantee accomplishes. Critics reply that this leaves Descartes unable to answer the general demon doubt he raised, so that the doubt effectively stands.
Three modern types stop trying to refute the doubt. They diagnose it, defuse it, or turn it against something else.
Diagnosis is not refutation (Stroud). Defusal pays with a revisionary semantics. And after 1983 the demon is a weapon against reliabilism, not against knowledge of the world.
The chat’s most striking historical claim is that the demon never lost; it changed jobs. The target moved while the weapon stayed.
Michael Williams gives the fullest version. Skepticism is not a presuppositionless discovery. It is the product of “epistemological realism,” the assumption that “knowledge of the external world” forms a natural, context-independent kind. His chapter of that name begins at p. 89 checked. Drop the assumption and the doubt gets no grip.
Stroud’s reply is that diagnosis is not refutation. His chapter shows why the doubt is so hard to shake:
With this thought, if he is right, Descartes has lost the whole world. He knows what he is experiencing, he knows how things appear to him, but he does not know whether he is in fact sitting by the fire with a piece of paper in his hand. Stroud p. 12 checked
If that conception can be explained and defended, the sceptical conclusion will remain intact and its scope and negative significance will be undiminished. Stroud p. 38 checked
State of play. Dominant in late twentieth-century epistemology. But it trades one contentious commitment (representationalism) for another (contextualism about epistemic kinds), and Stroud’s worry that it changes the subject is unrefuted.
Chalmers’s “The Matrix as Metaphysics” (2003) grants the scenario in full and denies that it is skeptical. If I am envatted, “I still have hands, there is still a table in front of me”; the world is simply made of bits. not held
First, the defusal works only for lifelong envatment. Chalmers concedes that a recent envatment or an interfering creator yields partial skepticism. Second, it relabels the loss: if “table” refers to a computational structure, the ordinary belief that tables are material, mind-independent and not simulated is still false.
Take your demon-deceived counterpart. Her experiences are subjectively identical to yours, but her beliefs are almost all false. Intuitively she is just as justified as you are. Her processes are wholly unreliable, so reliabilism gives the wrong answer.
…if we have reason to believe that our cognitive processes are reliable, it follows that in the demon world we would have every reason to believe that our cognitive processes were in fact reliable. Cohen, “Justification and Truth” p. 281 checked
…reliabilism avoids the skepticism of the Cartesian conception of the truth connection. For the evil demon hypothesis only demonstrates the possibility that there are belief forming processes which… Cohen p. 281 checked
Cohen’s own words show the inversion. Reliabilism was attractive because it seemed to escape Cartesian skepticism, and the demon is brought back to show what that escape costs. The chat’s state of play relies on Peter Graham’s “The New Evil Demon Problem at 40” (PPR, 2024): “Reliabilists remain undeterred.” not held The replies divide into the heroic (Goldberg, Senor, Lyons), the special-circumstances (Goldman, Sosa, Burge) and the excuse manoeuvre (Williamson, Littlejohn). Andrew Moon’s “Three Forms of Internalism and the New Evil Demon Problem” (2012) is also in the folder.
The matrix, the audit of the chat’s claims against the texts, and the thread into the next chats.
The taxonomy holds. Five citations need correcting; none of the corrections changes the taxonomy.
“The doubt has not been refuted; it has been contained, relabeled, and diagnosed.” That is the chat’s verdict, and nothing in the check against the texts disturbs it.
| Objector | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|---|
| Mersenne | ● | ||||||||
| Hobbes | ● | ● | |||||||
| Arnauld | ● | ||||||||
| Gassendi | ● | ● | ● | ● | (●) | ||||
| Bourdin | ● | (●) | |||||||
| Huet | ● | ● | ● | ● | |||||
| Foucher | ● | ||||||||
| Leibniz | ● | ||||||||
| Pascal | ● | ● | |||||||
| Hume · Peirce | ● | ||||||||
| Reid | ● | ● | |||||||
| Bouwsma · Ryle · Austin | ● | ||||||||
| Wittgenstein | ● | ● | |||||||
| Putnam · Davidson | ● | ||||||||
| M. Williams · Rorty · Carnap | ● | ||||||||
| Nozick · Dretske | (●) | ● | |||||||
| Chalmers | ● | ||||||||
| Lehrer & Cohen · Williamson | ● |
● central · (●) secondary. The matrix follows the chat, with one change: Gassendi is added under Type 1, since T6 is his. The chat places Nozick and Dretske under Type 8 (containment by concession); they reject closure rather than defuse the scenario, so a separate column would be more exact.
Hyperbolic doubt is a coherent theoretical instrument whose first-person expression is paradoxical and whose answer is contested. That is why it has outlived every refutation.
The next chat stays in the First Meditation, with the one reader (Derrida, answering Foucault) who asks what the doubt had to exclude, namely madness, before it could begin. Chat 3 then takes up Type 5 directly.
The chat as one collapsible tree. Click a node to collapse or expand its branch. Drag to pan; hold Ctrl or ⌘ and scroll to zoom, or open it full screen.
The interactive argument map is available only in the web version; it reproduces, as a collapsible tree, the structure of the sections above.