Descartes' Second-Order Judgments in Meditations on First-Order Beliefs, Lecture notes of Reasoning

Download Descartes' Second-Order Judgments in Meditations on First-Order Beliefs and more Lecture notes Reasoning in PDF only on Docsity! Locating “I think, therefore I am” in the Meditations Sam Pensler A thesis submitted for the degree of Master of Arts University of Otago Dunedin, New Zealand June 2017 ii Abstract “I think, therefore I am” (Cogito, ergo sum) suggests a “naïve” interpretation whereby anyone who argues as follows is certain of their existence. I think. Therefore, I am. Curiously, the famous line doesn’t appear in the Meditations, while it does in Descartes’ other works. Does the naïve interpretation, while a plausible reading of the other works, misread the Meditations? In this thesis, I claim that the Meditations should be naïvely interpreted by defending this position against three central objections. Objection 1: Nowhere in the Meditations does the meditator assert that cogito is certain. I respond that the meditator does assert the certainty of cogito in the first meditation as he doubts his beliefs. This happens when he makes judgments about what he is thinking such as: “I have no answer to these [skeptical] arguments” and “my habitual opinions keep coming back.” Objection 2: Even if the meditator claims cogito in the Meditations, he never accounts for why cogito is certain, which he must do if he uses it as a premise. I show that an argument for the certainty of cogito can be reconstructed by examining how the meditator doubts his beliefs. The idea behind the argument is that for the meditator to doubt his belief system it’s necessary that he is certain that he thinks, in particular, that he is certain about what his beliefs are and their amenability to doubt. In short, the certainty of cogito is built into the method of doubt. Objection 3: The naïve interpretation of the Meditations is false since Descartes says that the cogito is not an argument. For, he says that the cogito is a “simple intuition of the mind”, not a “deduction by means of syllogism.” I respond that Descartes is not denying that the cogito is an argument. He is specifying the type of reasoning process one must use to work through the argument from cogito to sum—sum is discovered by “intuition” rather than syllogistic reasoning. v A Note on Translation and Citation In this thesis, all citations of Descartes will be doubly cited. The first citation is from Collected Works, a compendium of Descartes’ oeuvre assembled by Charles Adam and Paul Tannery (“AT”). This is followed by a citation from The Philosophical Writings of Descartes, an English translation by John Cottingham, Robert Stoothoff, Dugald Murdoch, and Anthony Kenny (“CSMK”). I will almost exclusively quote the CSMK translation of Descartes’s writings. In several places in which the textual analysis is more exacting, I will quote the Latin from AT. For brevity’s sake, in-text references to secondary literature will be limited to a page number closed by parenthesis. If I cite multiple works by one commentator, the year of publication is included in the in-text citation. Full citations are found in the bibliography, formatted in MLA 7. 1 Introduction Famous quotes, as they become famous, sometimes get dislodged from where they were first said. This is true in the work of Descartes. “I think, therefore I am” is not stated in the Meditations, what is now Descartes’ most widely read work. It does show up, though, in the Discourse on Method and the Principles of Philosophy. So, many who read the Meditations for the first time, often familiar with Descartes’ most famous saying, are as surprised by its absence as scholars are eager to point out the aberration. But what are we to make of the gap? At several junctures in the Meditations, the meditator makes claims about his thoughts, his existence, and their relations, most notably at the beginning of the second meditation (AT VII 24-25; CSMK 2:16-17). In the Meditations, did Descartes refine, or even omit entirely, the philosophical insight expressed by “I think, therefore I am”? The overarching claim of my thesis is “no.” Descartes neither omitted nor significantly refined the philosophical insight expressed by “I think, therefore I am” in the Meditations. Shortly, I will state three central objections I will respond to in order to support my overarching claim that there is no “cogito gap” between the Discourse and the Principles, on the one hand, and the Meditations, on the other. However, first I must explain what philosophical insight Descartes is communicating with his famous saying. To draw out the philosophical insight, I will follow Margaret Wilson, who offers a “naïve” interpretation of the famous saying (50-71). Although Wilson doesn’t mention it explicitly, the naïve interpretation, I take it, earns that innocent title because it’s suggested 2 merely by considering the saying “I think, therefore I am” in isolation of the contexts in the Discourse and the Principles where Descartes asserts it. On the naïve reading, the meditator of the Meditations is presenting an argument. This is suggested by the phrase “therefore” (ergo). The argument establishes that the meditator is certain that he exists. The argument is this. I think. Therefore, I am. Suppose, for expository purposes, that I’m deploying this argument. The premise of the argument—I think—is something I’m certain of. I’m also certain that my existence follows from the fact that I think. From these two pieces, I infer the conclusion that I exist. And when I make this argument I become certain that I exist, since the conclusion is established by a certain premise and a certain logical entailment between the premise and the conclusion. Anyone else can make this argument, and so be certain that she too exists. A few clarifications should be made about the naïve argument to avoid confusion. First, a note about the meaning of the premise “I think” (cogito). As I will use the term, when someone judges “I think”, or uses “I think” as a premise in an argument, she is saying that there is some thought currently going on in her mind or, as is now common parlance, she is in some mental state. So, when someone judges “I think” or uses “I think” as a premise, I will say that she is making a “second-order judgment.” A second-order judgment is a judgment about one’s own mental state rather than a “first-order judgment”, which is only about some worldly content. “It will rain” is a first-order judgment whereas “I believe that it will rain” is a second-order judgment. Crucially, on my usage, when someone judges “I think” they could be referring to any first-order mental state that they are in at the time of that judgment. For example, they could be referring to their (first-order) 5 A second clarification is that on the naïve interpretation, the meditator concludes that he exists by doing more than judging himself to be in some first-order mental state. The meditator is also judging that his existence is entailed by the fact that he is thinking. Many questions arise here. For example, in meditator’s argument, is it necessary that he marks out the fact that he thinks entails that he exists as a separate premise from cogito in order to become certain that he exists? Or, is concluding sum on the basis of cogito enough? Does the meditator need to recognize that the fact that he thinks entails that he exists is an instance of a more general principle such as “Whatever thinks exists”? Why is it true that my thinking entails my existence? I will put off these questions until the third chapter of my thesis where I will examine how Descartes begins to answer these questions. The naïve interpretation is neutral about the answers to these questions. Third, note that in the naïve argument, the meditator doesn’t need to judge “I think is certain” to become certain of his existence. In other words, the premise doesn’t need to be: “It is certain that I think.” Relatedly, the conclusion doesn’t have to include certainty. The reason why is that, as we will soon see, the meditator accepts, as a tenet of his method of doubt, that he should only make judgments that are certain. Thus, it’s implied that the premise and conclusion are things of which the meditator believes to be certain, in addition to being true.4 Fourth, the naïve argument doesn’t establish that the reasoner persists, or exists for any period except for the instant at which she is certain that she thinks. So, a completely accurate version of the naïve argument would have the premise and conclusion indexed to the exact same time. 4 For a discussion of what Descartes means by “certainty”, see chapter two, page 68-70. 6 My thesis argues for the naïve interpretation of the Meditations by responding to three central objections. In the first chapter, I take up the objection that the naïve interpretation is incorrect since nowhere in the text of the Meditations does the meditator judge cogito. That is, nowhere does the meditator make a certain categorical second-order judgment. I argue that the meditator does make categorical second-order judgments that he regards as certain in the first meditation when he makes remarks such as “I have no answer to these arguments [the skeptical arguments]” and “my habitual opinions keep coming back, and despite my wishes, they capture my belief” (AT VII 21-22; CSMK 2:14-15). I go on to argue that in the first meditation, by engaging in the process of doubting his first-order beliefs, the meditator is committed to the Certainty of Mind Thesis (“CM”), the view that all his second-order judgments are certain. Surprisingly, then, the meditator has some certainty even in the first meditation. I close by arguing that when the meditator begins the second meditation, he uses these second-order judgments as a premise in his argument for sum. In the second chapter, I respond to the objection that if the meditator were to derive sum from certain second-order judgements (which is what the naïve interpretation claims), then the mediator should provide an account of why his second-order judgments are certain, but he lacks such an account. I argue that an argument for the Certainty of Mind Thesis can be reconstructed in the first meditation that bears resemblance to an argument developed by Tyler Burge. The general idea behind the argument is that the meditator’s process of doubt requires certain second-order judgments. The meditator must be able to be certain about the structure and substance of his (first-order) belief system. He also must have certain second- order judgments to monitor whether his beliefs have changed in the face of the skeptical 7 arguments and determine when he has completed the project of doubting his beliefs. In short, certain higher-order judgments are built into the method of doubt. The connecting theme of chapters one and two is that a close look at the first meditation, particularly the role of reflection in the method of doubt, provides novel responses to two central problems for the naïve interpretation. It turns out that cogito, the premise of the meditator’s argument for his existence, is claimed in the first meditation. Also, we will see that in the first meditation the meditator has an argument available to him that establishes the certainty of cogito and more generally the certainty of the access he has over his own mind. In the third chapter, I examine the objection that the cogito is not an argument at all. This objection emerges not from the text of the Meditations but from passages in the Objections and Replies to the Meditations as well as Descartes’ Conversation with Burman. For example, in one of the passages Descartes says: “When someone says ‘I am thinking, therefore I am, or I exist’, 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” (AT VII 140-141; CSMK 2:100). I argue that Descartes here is not denying that the cogito is an argument. Rather, he is specifying the type of process by which someone must reason through the argument from cogito to sum—sum is discovered by what he calls “intuition” rather than syllogistic reasoning. To defuse the tension, I examine Descartes’ account of various types of reasoning—intuition, deduction, and syllogism—present in one of his earliest written works, the Regulae. 10 his beliefs to doubt. Furthermore, Descartes intends for the skeptical arguments, the levers of his project of doubt, to be the principal focus of the first meditation. Indeed, in the Second Replies, he says that he intends readers to “devote several months, or at least weeks” to the skeptical arguments before going on to the rest of the book (AT VII 130; CSMK 2:94). The meditator also seems to be very explicit that the second meditation is where he will begin to uncover beliefs that are certain. Midway through the first paragraph of the second meditation he says: Anything which admits of the slightest doubt I will set aside just as if I had found it to be wholly false; and I will proceed in this way until I recognize something certain, or, if nothing else, until I at least recognize for certain that this is no certainty. (AT VII 24; CSMK 2:16) An implication of this appears be that he has not at this point in the text, or previously, found anything certain. Shortly, I will explain why such a conclusion would be a mistake— the meditator does find certain beliefs in the first meditation. The traditional reading holds that it’s not until the third paragraph of the second meditation, that the meditator discovers a certain belief of some sort. Because of the way the meditator, in what was just quoted, appears to mark out the third paragraph as the place in which he will discover the first thing he is certain of, the “Archimedean point” (ibid), many think that the third paragraph is the authoritative passage on Descartes’ views about the relation of cogito and sum. This is the passage: Yet apart from everything I have just listed, how do I know that there is nothing else which does not allow even the slightest occasion for doubt? Is there not a God, or whatever I may call him, who puts into me the thoughts I am now having? But why do I think this, since I myself may perhaps be the author of my thoughts. In that case am not I, at least, something? But I have just said that I have no senses and no body. This is the sticking point: what follows from this? Am I not so bound up with a body and with senses that I cannot exist without them? But I have convinced myself that 11 there is absolutely nothing in the world, no sky, no earth, no minds, no bodies. Does it now follow that I too do not exist? No: if I convinced myself of something then I certainly existed. But there is a deceiver of supreme power and cunning who is deliberately and constantly deceiving me. In that case I too undoubtedly exist, if he is deceiving me; and let him deceive me as much as he can, he will never bring it about that I am nothing so long as I think that I am something. So after considering everything very thoroughly, I must finally conclude that this proposition, I am, I exist, is necessarily true whenever it is put forward by me or conceived in my mind. (AT VII 24-25; CSMK 2:16-17. Italics are Descartes) Of the passage, Frankfurt writes: The statement, I think, therefore I am, simply does not occur in the passage at all; and neither does any exactly equivalent statement. In fact, the cogito as such does not appear anywhere in the Meditations. I propose to take Descartes’s text on its own terms and to approach it without preconceptions based on the speculation that cogito ergo sum adequately formulates its meaning. (Frankfurt, 92, italics are Frankfurt’s). In the end, I think there is something correct in Frankfurt’s remark. In the passage, there is no point at which Descartes asserts verbatim “I think, therefore I am.” However, at the least, the meditator is asserting that several of his thoughts entail that he exists. I will point out four of these entailments. First, the line, “No: if I convinced myself of something then I certainly existed.”6 seems equivalent to: (1) If I convinced myself of something, then I certainly existed. The next two sentences are a little trickier to interpret, both because the first sentence starts with “but” (sed) and because the first half of the second sentence, the clause that ends with the semicolon, seems more germane to the first sentence—showing something that follows from what is being supposed in the first—rather than the second sentence, which expresses a slightly different claim. But there is a deceiver of supreme power and cunning who is deliberately and constantly deceiving me. In that case I too undoubtedly exist, if he is deceiving me; 6 “Imo certe ego eram, si quid mihi persuasi.” 12 and let him deceive me as much as he case, he will never bring it about that I am nothing so long as I think that I am something. (AT VII 25; CSMK 2:17)7 Note that the “but” that begins these sentences plays the role of introducing considerations—a supposed evil demon—that is claimed to bear on the truth of (1). Thus, the first sentence and the second sentence up until the semicolon don’t express a new entailment of the form: “If I am deceived by a powerful and cunning deceiver, then I undoubtedly exist.” What they express is a slightly modified version of (1): (2) If I convince myself of something, then I certainly exist, even if I am being deceived.8 Moreover, the remainder of the second sentence can be rendered: (3) If I think that I am something, then indubitably I exist. And the final sentence, the “conclusion” Descartes reaches after “considering everything very thoroughly”9 might be read as (4) If I conceive of or put forward the proposition “I exist” in my mind, then I exist. Before I move on, I wish to say why the traditional reading makes the naïve interpretation looks so unpromising. Recall that on the naïve interpretation the meditator establishes that he exists with certainty by inferring this from the premise “I think”, which he regards as certain, and, as I understand Descartes’ usage of “I think”, it’s equivalent to 7 “Sed est deceptor nescio quis, summe potens, summe callidus, qui de industriâ me semper fallit. Haud dubie igitur ego etiam sum, si me fallit; & fallat quantum potest, nunquam tamen efficiet, ut nihil sim quamdiu me aliquid esse cogitabo.” 8A subtlety about the past/present tense: if my reading of (2) as being a modification of (1) in the way described is correct, it seems plausible that Descartes would have allowed (1) to be rewritten in the present tense, since the two sentences that begin with “But” are themselves in the present tense. This point about tense is significant because “I think, therefore I am” is in the present tense. 9 “Adeo ut, omnibus satis superque pensitatis, denique statuendum sit hoc pronuntiatum, Ego sum, ego existo, quoties a me profertur, vel mente concipitur, necessario esse verum.” 15 Since Descartes is explicit that he doesn’t intend for things established at later stages to be parts of arguments for claims made earlier on, it would be illicit for the possible assertions within AT VII 28-29; CSMK 2:19 to be counted as components of an argument for the certainty of his existence. Notice: AT VII 28-29; CSMK 2:19 occurs after the meditator concludes his argument for the certainty of his existence. The change from arguing for his existence to examining the nature of his existence is implied by the transition sentence that comes directly after the third paragraph of the second meditation: “But I do not have a sufficient understanding of what this ‘I’ is that now necessarily exists” (AT VII 25; CSMK 2:17). That these two matters—establishing his existence and inquiring into the nature of his existence—are different issues and are taken up sequentially is further confirmed in Descartes’ gloss of the second meditation in the Synopsis (AT VII 12-13; CSMK 2:9). A somewhat more compelling defense of the naïve interpretation is to highlight passages from the Objections and Replies in which Descartes attempts to clarify the argument for sum made in the Meditations. The most promising of these passages is a response that Descartes makes to one of Gassendi’s objections. There, I read Descartes as claiming that any second-order judgment about one’s first-order thoughts is completely certain and so it can be used as a premise to infer the certainty of sum. Gassendi’s objection is that the meditator didn’t need “all this apparatus” of putting forward, or conceiving of, the proposition “I exist” to “conclude” that he exists—the meditator “could have made the same inference from any one of [his] other actions, since it is known by the natural light that whatever acts exists” (AT VII 258-259; CSMK 2:180). Below is Descartes’ response. I draw the reader’s attention to two points. (1) Descartes’ response accepts the naïve interpretation. (2) Descartes’ comments suggest that 16 any second-order order judgment will serve a premise in an argument for the certainty of sum since our second-order judgments are completely certain. Again, what reason have you for saying that I 'did not need all this apparatus' to prove I existed? These very words of yours surely show that I have the best reason to think that I have not used enough apparatus, since I have not yet managed to make you understand the matter correctly. When you say that I 'could have made the same inference from any one of my other actions' you are far from the truth, since I am not wholly certain of any of my actions, with the sole exception of thought (in using the word 'certain' I am referring to metaphysical certainty, which is the sole issue at this point). I may not, for example, make the inference 'I am walking, therefore I exist', except in so far as the awareness of walking is a thought. The inference is certain only if applied to this awareness, and not to the movement of the body which sometimes - in the case of dreams - is not occurring at all, despite the fact that I seem to myself to be walking. Hence from the fact that I think I am walking I can very well infer the existence of a mind which has this thought, but not the existence of a body that walks. And the same applies in other cases. (AT VII 352; CSMK 2:243-244) Here, Descartes does plainly assert that a second-order judgment such as “I think I am walking” is certain.10 Still, one might find this way of defending the naïve interpretation objectionable. According to Frankfurt (10), to emphasize that Descartes accepts the naïve reading in his response to Gassendi is to admit that there is a “serious gap” in the meditator’s discussion of sum in the third paragraph of the second meditation, at AT VII 24-25; CSMK 2:16-17. There, the meditator doesn’t make certain second-order judgments; he doesn’t assert cogito. Frankfurt’s objection seems to be this. Defenders of the naïve interpretation wish to read Descartes’ response to Gassendi, in which Descartes argues for the certainty of sum just as the naïve interpretation says, as Descartes’ real account of the cogito. But 10 Of course, “second-order judgment” is not Descartes’ terminology. In the introduction, I explained how and why this terminology is used. Descartes here is credibly read as implying that “I think I am walking” is a second-order judgment since he refers to “I think I am walking” as a “fact” and suggests it may serve as a premise in the argument for sum (which plausibly implies that the reasoner has judged the premise to be true). 17 this account of the cogito differs from the meditator’s account in the second meditation, where an assertion of cogito is not to be found. Thus, defenders of the naïve interpretation who emphasize Descartes’ response to Gassendi must portray the meditator as launching an invalid argument for the certainty of sum in the second meditation. Interpreting the meditator as making an invalid argument is unacceptable. If this counter objection to an emphasis on Descartes’ response to Gassendi is merited, then the naïve interpretation still lacks a cogent response to Frankfurt’s initial textual objection that the meditator doesn’t assert cogito in the text of the Meditations. Where are we? The traditional reading holds that the meditator commences his inquiry into what is certain in the third paragraph of the second meditation, at AT VII 25- 25; CSMK 2:16-17. There, the meditator mounts some argument concerning the certainty of sum and its relation to cogito. The traditional reading is very hard to reconcile with the central conviction of the naïve interpretation that the meditator establishes the certainty of his existence by deriving that conclusion from a certain judgment he makes about his own mind. This is because, as we have seen, the evidence presented by adherents of the naïve interpretation that Descartes makes certain second-order judgments elsewhere—either at some later part of the Meditations or in the Objections and Replies—is unsatisfactory. What is left in the Meditations when read traditionally are scraps of the naïve argument: the if- then statements I extracted from AT VII 24-25; CSMK 2:16-17, which are in some sense more particular versions of the proposition “If I think, then I exist,” and the conclusion, as Descartes seems to put it, that “this ‘I’ now necessarily exists” (ibid). The missing piece is some assertion made by the meditator that he thinks, or at least of the thoughts that occur as the antecedents in the entailments at AT VII 24-25; CSMK 2:16-17. 20 (11) It is a good plan to deceive myself into pretending my former opinions are false and imaginary. And the assertions continue. At the very end of the first meditation, as he is reacting to the refined evil demon argument, the meditator also makes claims about other first-order mental states of his, such as being in a state of “dread” for subjecting old comfortable opinions to doubt and “fear” of having to laboriously use his reason to find indubitable beliefs, possibly without success. The issue we have been concerned with over the last several pages is whether the meditator makes any second-order judgment about his first-order mental life. If the meditator makes a second-order judgment that he regards as certain, this would obviate Frankfurt’s objection that the meditator should not be read as concluding the certainty of sum from the certainty of cogito because the meditator never asserts cogito with certainty. That is, he never makes a certain second-order judgment. Now, we have seen that the meditator does makes several second-order judgments during his procedure of doubting his first-order beliefs. Soon, I will explain why the meditator also regards these second-order judgments as certain. My response to Frankfurt’s objection will also show that the traditional reading is mistaken: the meditator’s argument for his existence begins in the first meditations, at AT VII 21-22; CSMK 2:14-15. A virtue of my reading is that it bolsters the plausibility of the naïve interpretation while understanding the meditator as arguing for his existence in an orderly sequence. The sequence starts with various thoughts being asserted in the first meditation, then various thoughts are asserted to entail sum at the third paragraph of the second meditation, and finally sum is concluded at the very beginning of the fourth paragraph of the second meditation. As we have seen, those like Bernard Williams and Margaret Wilson, who agree 21 with me that the naïve interpretation is correct, rely on textual evidence that the meditator asserts cogito after he concludes sum. This is weak evidence because it portrays the meditator as arguing in a disorderly, disorganized fashion, drawing conclusions from premises that are only stated ex post facto. Such an interpretation would have left Descartes himself quite unsatisfied, given his remarks about the order of discovery we have seen from the Second Objections and Replies. I turn now to three objections to reading the meditator as commencing his argument for the certainty of his existence in the first meditation. The first objection is that for the meditator to draw his conclusion that he is certain of his existence, he must assert that his thoughts are certain. However, goes the objection, in the passages I have cited the meditator makes assertions about several of his thoughts, but doesn’t regard these assertions as certain. He merely states, “my habitual opinions are coming back” (AT VII 22; CSMK 2:15) not “Certainly, my habitual opinions are coming back.” Perhaps the meditator is making these higher-order assertions as uncertain, reasonable conjectures about what is going on in his mind. There are two pieces of evidence that show that the meditator regards his second- order judgments about his thoughts as certain. The first can be gleaned from a line that reveals the central tenet of the method of doubt in the Meditations. At the very beginning of the first meditation, the meditator says, “Reason now leads me to think that I should hold back my assent from opinions which are not completely certain and indubitable just as carefully as I do from those which are patently false” (AT VII 18; CSMK 2:12). What this comes down to is a norm or constraint regarding what should be asserted (and by that token believed, on the assumption that asserting that p entails believing that p) by any rational 22 individual who wishes to engage in the method of doubt. The constraint is a negative one: do not assert to what can be doubted, even ever so slightly. Yet, since the meditator also champions a constructive side of his project, one of establishing in the sciences what is “stable” and “likely to last” (ibid) and what it certain, it’s plausible that the meditator accepts the positive version of the rule of doubt: assert what is certain, that which cannot be doubted at all. Thus, we should expect that whenever the meditator, always didactically obedient to the method of doubt, makes an assertion, he regards that assertion as certain. While it’s true that the meditator doesn’t always explicitly attach the word “certain” to every assertion he makes, the deviation is a matter of style, not substance. This can be seen at AT VII 21-22; CSMK 2:15-16, where the meditator sometimes expresses the entailments so that their consequents sometimes contain the word “certain” and sometimes don’t. His intermittent sprinkling of “certain” is immaterial, and included most likely to remind the reader that the meditator is obeying the rule of doubt, unlike in ordinary life, in which we sometimes assert things that we take to be true but uncertain. The second piece of text that shows the certainty in the meditator’s judgments during the first meditation is a passage we have already seen, Descartes’ response to Gassendi in the Fifth Set of Replies. There, Descartes says that “I am not certain of any of my actions, with the sole exception of thought” (AT VII 352; CSMK 2:244). He goes on to suggest that he can be certain of any of his thoughts, such as the thought or awareness that he is walking (ibid). Here, Descartes appears to commit himself to the thesis that all one’s second-order judgments are certain. Since Descartes is committed to this thesis about the certainty of mind, it would be inconsistent for him to hold that the meditator’s assertions about his mind are uncertain. 25 (1) My habitual opinions are coming back even after I doubted them. (2) If my habitual opinions keep coming back, then I exist. (3) Therefore, I exist. What might be true is that the meditator didn’t recognize or discover that claims like “I have no answer to these argument” are things he asserted with certainty in the first meditation—or could assert with certainty—until the second meditation, which begins with a summary of what skeptical intellectual activities previously have occurred. This point is bolstered by how the objective of the second meditation is described: “Anything which admits the slightest doubt I will set aside…I will proceed until I recognize something certain” (AT VII 24; CSMK 2:16, my italics). “Recognize” leaves open the possibility that the meditator makes certain second-order judgments in the first meditation, but didn’t recognize this and consider its philosophical import at that earlier time. Still, even if the meditator didn’t recognize until the second meditation that he made several certain second- order judgments in the first meditation, this is not a compelling reason to hold that the meditator didn’t intend any of the judgments to be a premise in the argument for sum. He still intends any of them to serve as a premise, he just doesn’t form this intention until the second meditation. The third objection concedes that the meditator makes certain second-order assertions concerning several of his thoughts in the first meditation and that these assertions may be paired together with an if-then statement made in the second meditation to make up an argument for the certainty of sum. The objection maintains that if the meditator argued in this way, he wouldn’t establish the certainty of sum because he must rely on memory, which is open to doubt. The meditator must rely on memory because my reading suggests 26 that at the time the meditator concludes sum in the second meditation, he must recall an assertion he made back in the first meditation, which took place at an earlier time. This objection is correct in saying that if the naïve interpretation reads the meditator as using his memory to argue for the certainty of sum, then he would be unable to show the certainty of sum. Indeed, the meditator acknowledges that memory is open to doubt in the second paragraph of the second meditation. The meditator says that he will suppose that his memory tells him lies (AT VII 24; CSMK 2:16), presumably because his memorial beliefs are open to doubt. The objection is also correct that the first meditation occurs at an earlier time than the second meditation. In the first line of the second meditation, the meditator refers to the first meditation as “yesterday’s meditation” (ibid). In response to the objection, a further look at the beginning of the second meditation shows that the meditator doesn’t rely on memory because he continues to engage in the method of doubt through the first two paragraphs of the second meditation. The meditator likens his persisting state of doubt to a swimmer caught in a whirlpool and vows to “once more attempt the same path which I started on yesterday” (AT VII 24; CSMK 2:16). Since the meditator continues with his enterprise of doubt at the opening of the second meditation, he continues be in a reflective state of mind. He keeps on thinking about what is happening to his mind as he engages in doubt. At the beginning of the second meditation, the meditator has many of the same thoughts that he did at the end of the first meditation. For example, in the first paragraph of the second meditation, the meditator is found again asserting that he should only assent to what cannot be doubted (ibid). He also reasserts that 27 he is convinced by the skeptical arguments in the third paragraph.13 In addition, in the third paragraph the meditator makes new judgments about his first-order mental life: he wonders whether God could have authored his first-order thoughts. Thus, while the meditator begins to make judgments about what is going on in his mind at the end of the first meditation, he continues to do so through the beginning of the second meditation. Strictly speaking, it’s the meditator’s second-order judgments from the very beginning of the second meditation that my reading holds are used in an argument for sum, since they occur at the same time as the meditator’s judgments about how various of these thoughts entail the certainty of sum. By noting that the meditator’s process of doubt and concomitant reflections on his belief system endures into the second meditation, we can see that the meditator continues to make second-order judgments about some of the same first-order thoughts as he did back in the first meditation. For this reason, the meditator doesn’t rely on memory when he concludes sum in the first sentence of the fourth paragraph. The naïve interpretation has it that the premise and conclusion of the argument for sum are indexed to the same time.14 13 Indeed, there are some thoughts that the meditator has in the first meditation, at AT VII 22; CSMK 2:15, like “my habitual opinions keep coming back”, that he no longer has when he begins the second meditation at AT VII 23-24; CSMK 2:16. The reason he doesn’t continue to think that his former beliefs keep coming back by the start of the second meditation is that the final formulation of the demon argument presented in the last paragraph of the first meditation presumably was successful, and he completely stopped assenting to his former beliefs. 14 One might further object that on my reading the two premises, “cogito” and “if cogito, then sum”, and the conclusion “sum” still cannot occur at the same time because the meditator asserts each premise and the conclusion in different sentences within the beginning of the second meditation. So, the different sentences suggest the inference is diachronic. As we will see in chapter three, Descartes understands the cogito to be inferred using a psychological process called “intuition” where the premises and conclusion of an argument can all be asserted and inferred in one mental grasp or a single “movement of thought” as Descartes says (AT X 368-370; CSMK 1:14-15). Thus, it’s beside the point 30 While the view that we have “privileged access” to our minds was probably thought to be a Cartesian view before Ryle, it was Ryle who coined the now popularly used term in Descartes scholarship as well as within contemporary philosophy of mind and epistemology. Ryle also distilled the Privileged Access Thesis into two further theses, what many have called the “Infallibility Thesis” and “Omniscience Thesis” represented below.17 Following Ryle, there has been significant debate among scholars whether Descartes really commits himself to these claims. Infallibility: If S makes the second-order judgment that she is in mental state M, S is in M. Omniscience: If S is in M, S believes that she is in M. According to the Infallibility Thesis, if you judge that you are in some mental state, that judgment is “exempt from error” as Ryle says (137). That is, it will always turn out to be true; you will always in fact be in the mental state that you self-ascribe. Since Ryle, many have attributed the Infallibility Thesis to Descartes, but some do not.18 According to the Omniscience Thesis, if you are in some occurrent mental state, then you will believe that you are in it. As Ryle colorfully puts it, “a mind cannot help 17 Ryle doesn’t name them as such. See 136-138. 18 Sometimes the Infallibility Thesis is called the Incorrigibility Thesis. For those who attribute infallibility to Descartes, see Ryle (140) Audi (91), Kenny (70-72), Vinci (10), McRae (57), Williams (70). Broughton (2008, 179) and Wilson (150-165) partly attribute the thesis to Descartes. Curley (170-193) and Newman (2) do not attribute the view to Descartes. I’m disregarding slight differences in formulations of the view between authors. Also, it’s usually specified in a more exact formulation of the infallibility claim that one’s self-attributions are infallible only if they are arrived at through an introspective process. This specification is made to preclude cases of false self-attributions of mental states arrived at via sense perception (i.e. forming a false belief about yourself by listening to a false conjecture from your therapist) from being counterexamples to the infallibility thesis. 31 being constantly aware of all the supposed occupants of its private stage” (136). Again, there are a range of views on whether Descartes held the Omniscience Thesis.19 Now that we have seen how the meditator makes certain second-order judgments about his mind and belief system as he engages in the method of doubt, we are in a better position to consider Descartes’ commitments to the Privileged Access Thesis in the Meditations. The meditator commits himself to the strong claim that he can make certain judgments about his first-order mental states, or at least those first-order mental states that are part of his enterprise of doubt. To be certain about X implies, among other things, that X is true. 20 So, the Certainty of Mind Thesis implies the Infallibility Thesis. Thus, there is good evidence that the meditator relies on the Infallibility Thesis in the first meditation. Scholars who believe that Descartes commits himself to some Privileged Access Thesis in the Meditations frequently cite passages in the second or third meditations. For example, some scholars think that the meditator implicitly draws on infallibility, omniscience, or both in the second meditation, in his reasoning surrounding the cogito,21 for sum res cogitans,22 and even in his discussion of the wax.23 Other commentators think that the meditator relies on Privileged Access in the third meditation when claims to know 19 Sometimes the Omniscience Thesis is called the Self-Intimation or Self-Presentation Thesis. Ryle (140) Audi (91), Vinci (10), McRae (57), Kenny (70-72), and Williams (70) attribute the view to Descartes. Broughton (2008, 179) and Wilson (150-165) partly attribute the view. Curley (170-193), Newman (2), and Rozemond (57-66) do not. The Omniscience Thesis is limited to occurent states rather than latent or standing mental states. Descartes likely thought that there are latent mental states—in particular, innate ideas—that we are not always aware of, or at the least not clearly and distinctly aware of. See Nelson (163-178) for a discussion of innate ideas in Descartes’ philosophy. 20 See my discussion of certainty in chapter two, page 68-70. 21 Kenny (47-48), McRae (57), Williams (71), and Wilson (59-60). 22 Kenny (70-72), McRae (57), Williams (71), and Wilson (150-165). 23 Broughton (2008, 193). 32 that he has clear and distinct perceptions,24 or in his taxonomy of ideas and thoughts.25 In this chapter, I have argued that from the very beginning of the Meditations, the meditator relies on his privileged access to his own mind while he engages in the enterprise of doubting his beliefs. In the next chapter, I will consider how the meditator might defend the certainty present in his first meditation doubts. 24 Broughton (2008, 193). 25 Wilson (150-165). 35 true in the Meditations. I will argue that, whether he knew it or not, in the first meditation Descartes makes available to the meditator a compelling argument for CM, or rather a slightly more restricted version the thesis. This argument begins with the assumption that the meditator can doubt his first-order beliefs. The argument then shows that the certainty of the meditator’s second-order judgments that are integral to his enterprise of doubt follow from the assumption that he can doubt his beliefs. Tyler Burge makes a similarly structured argument for a somewhat weaker privileged access thesis than CM. So, midway through this chapter we will consider Burge’s argument to assist us in our reconstruction of Descartes. I. Does Descartes Already Have an Argument for CM? Reconstructing an argument for CM would be ill-advised if Descartes makes a strong case for it elsewhere. Let us consider whether Descartes already has a satisfactory argument. To look for an argument, we might return to Descartes’ response to Gassendi’s objection, perhaps the most explicit endorsement of CM across Descartes’ works. Gassendi’s objection is that the cogito argument is too restrictive; he contends that the meditator could have inferred sum from a premise about any of his actions, such as “I walk” rather than only “I think.” Descartes responds by saying, “the inference is certain only if applied to this awareness [of walking], and not to the movement of the body which sometimes—in the case of dreams—is not occurring at all” (AT VII 352; CSMK 2:244). One might read a trace of an argument for CM in Descartes’ response: if one is aware of a mental state M, then one is certain that she is in M. However, I don’t think that Descartes means for this to be an argument for CM. Descartes is only articulating what is 36 certain. We can be certain that we are aware that our body is in some state, but we cannot be certain that our body is in some state. That is, Descartes is only specifying what we can be certain about. If this passage is to offer an argument for CM, Descartes must tell us why awareness or consciousness provides certainty. Here, he does not do this.27 Some commentators have interpreted Descartes as defending CM and other theses concerning privileged access such as infallibility and omniscience by an appeal to his stipulated definitions of “thought” and “idea” from the Second Replies (AT VII 160-161; CSMK 2:113).28 I think this reading is undesirable. If Descartes did in fact rely purely on stipulated definitions in support of CM, a substantial thesis about privileged access, we should seek an alternative reconstruction if possible. Arguments that resort only to stipulated definitions are unsuccessful. Now, if Descartes didn’t rely purely on stipulated definitions to establish CM (in my view, McRae and Kenny don’t adduce forceful evidence that Descartes means for CM to follow from his definitions of “thought” and “idea”), then the defense-by-definition reading is plainly uncharitable. We ought to keep looking for an argument for CM. Another area to examine for a defense of CM may be Descartes’ doctrine of clear and distinct perception. At the beginning of the third meditation, the meditator seems to suggest that a perception being completely clear and distinct is a mark of what is known and certain (AT VII 34-35; CSMK 2:24-25). Perhaps Descartes thought that we can have clear and distinct perceptions of our current first-order mental states, and, in this way, be certain of cogito. 27 Broughton also makes this point (2008, 188-189). 28 For this position, see Kenny (48-50) and McRae (57). 37 While it would take me too far afield to examine Descartes’ doctrine of clear and distinct perceptions, and whether Descartes might have relied on it in support of CM, I wish only to say that there are interpretations of the doctrine on which it doesn’t support CM. For example, Alan Nelson argues for an interpretation of the doctrine on which only innate ideas can be clearly and distinctly perceived (163-167). Further, on Nelson’s view, we only have innate ideas of God, ourselves, corporeal bodies, eternal truths, and possibly also pains and colors. I understand Nelson’s reading as implying that we cannot have clear and distinct perceptions of our current first-order mental states, such as my belief that there is a computer screen in front of me, because these mental states are not inborn in us and latent, as are innate ideas. Rather, my current first-order mental states come and go. CM, on the other hand, does assert the certainty of judgments about our current first-order mental states. So, if Nelson’s interpretation is correct, then the certainty of our judgments about our current first-order mental states must come from another source. And even if the doctrine of clear and distinct perception can successfully be extended to CM, I think it would be insightful to find that the Meditations contains an alternative argument for CM unsupported by the doctrine of clear and distinct perception. That is what I will turn to next. II. Clues for Reconstructing Descartes’ Argument for the Certainty of Mind Thesis Our attempt to locate Descartes’ own defense of the Certainty of Mind Thesis has proved unsuccessful. In the places in which Descartes appeals to CM in his response to Gassendi, he does so without an account of why, when we judge ourselves to be in a particular mental state, those self-ascriptions carry certainty. Other potential lines of defense for CM that 40 today. In that case, my second-order judgment is contingently connected to my current first-order belief. I could have thought “I get to see my dog today” without reflectively judging “I’m currently believing that I get to see my dog today.” Unlike the dog case, as the meditator engages in the method of doubt, the changes in his first-order beliefs require him to make second-order judgments about his first-order beliefs. It’s inconceivable that the meditator (or anyone else) could doubt all his first-order perceptual beliefs, and withhold belief from them, all without having any second-order judgments about how his outlook on the world has changed. Of course, we frequently cast doubt on our beliefs without having second-order judgments. For example, as I raise what looks like veggie pizza to my mouth, my friend Jim interposes: “I think that has some meat on it.” After that, I doubt that it’s veggie. Doubting it’s veggie pizza doesn’t require having second-order judgments. But some doubts do require having second-order judgments. The more pervasive one’s doubts, and the deeper the beliefs that are made suspect in the structure of one’s belief system, the more common it is to reflect on and review one’s doubts. Furthermore, the meditator’s doubts, which are extensive (the doubt strikes many beliefs in the system) and extreme (the doubt strikes beliefs that confer justification to many other beliefs) require second-order judgments. They are an essential part of meditator’s process of doubt. My second point is an extension of the point just made, that the meditator’s second- order judgments are essential to doubting his first-order beliefs. For Descartes, to doubt a belief is to use one’s faculty of reason, among other things. Furthermore, to successfully doubt an entire system of beliefs is to exercise one’s reason to its fullest. Thus, having 41 second-order judgments, which we observed is essential to doubting an entire belief system, is also essential to using reason to its fullest. I will say a little more about how Descartes understands doubt to be an activity performed by reason. Descartes believes that as we grow up, by habit we come to rely mostly on our senses for knowledge. This reliance conceals us from the perfect knowledge (scientia) in metaphysics and the sciences, causing us to have many false beliefs in these areas. Massive doubt is a process in which reason frees itself from the senses, allowing reason to discover true and certain metaphysical and scientific conclusions. This view of doubt as a process whereby reason frees itself can be found in more than several places in Descartes’ corpus, but it’s presented quite clearly in the introductory materials to the Meditations. In the Synopsis to the Meditations, Descartes explains that the “usefulness of such extensive doubt” in the first meditation “lies in freeing us from all our preconceived opinions, and providing the easiest route by which the mind may be lead away from the senses” (AT VII 12; CSMK 2:9). In the Preface to the Reader, we are told that once the mind is freed (which is later identified with reason: “I am a mind, or intelligence, or intellect, or reason” (AT VII 27; CSMK 2:18)), it can “arrive at certain and evident knowledge of the truth.” (AT VII 10; CSMK 2:8). Previously we saw that the meditator’s second-order judgments are essential to his enterprise of doubt. Now, considering how Descartes holds the view that reason entails the power to doubt, it follows that having second-order judgments is essential to reason exercising its full authority. The final two points regard the way in which second-order judgments enter and assist in the meditator’s procedure of doubt. 42 The third point is that the meditator’s second-order judgments are about first-order mental states that are relevant to the process of extensive doubt. Of course, some of the meditator’s second-order judgments, like “My habitual opinions keep coming back”, are just about his first-order beliefs. These second-order states allow him to assess and monitor whether he still holds various beliefs after meditating on the skeptical arguments. Other second-order judgments concern the connection between his former beliefs and his reasons for holding them. For example, the meditator states how he is “compelled to admit” that all his former beliefs can be properly doubted “based on powerful and well-thought out reasons” (AT VII 21-22; CSMK 2:14-15). Here, the meditator is acknowledging that the skeptical arguments are reasons his former beliefs could be false. The meditator also has second-order judgments about attitudes other than beliefs, like his intention to follow through with the enterprise of doubt—“I shall stubbornly and firmly persist in this meditation…I shall…resolutely guard against assenting to any falsehoods” (ibid). So, the meditator has second-order judgments about his first-order beliefs, their reasons, and even non-belief states, such as his intentions to follow through with his enterprise of doubt. The fourth and final point I wish to make is that the meditator has another kind of second-order judgment about what he should believe. After the meditator thinks about how the skeptical arguments are reasons to think his former beliefs are false, he has the following thought: “I must withhold my assent from former beliefs just as carefully as I would from obvious falsehoods” (AT VII 21-22; CSMK 2:14-15, my emphasis).29 It’s appropriate to read “must” as expressing an intellectual standard that the meditator is 29 “ideoque etiam ab iisdem, non minùs quàm ab aperte falsis, accurate deinceps assensionem esse cohibendam, si quid certi velim invenire.” 45 Burge’s arguments is to return to the Meditations in the section to follow, where I will show how similar arguments to Burge’s can be reconstructed in support of the Certainty of Mind thesis. According to Burge, we are rational beings. Rational beings are unique in that they have attitudes. Examples of attitudes include the belief that there will be a snowy Christmas, the desire to travel, or the intention to return a borrowed backpack. Now, rational beings do not just have attitudes on a whim. Their attitudes are based on reasons. I believe that there will be a snowy Christmas because the weatherman said so. I desire to travel because I value seeing new places. I intend to return a borrowed backpack because I agreed to return it. All these reasons can be said to support my attitudes. The reasons make my attitudes rational. But this is not always the case. Consider a little boy who believes there will be a snowy Christmas only because it snowed the Christmas before. He believes that it will snow this Christmas for a bad reason. The fact that it snowed last year on one particular day does not justify someone in thinking that it will snow again this year on the same date. All rational beings are open to evaluation, unlike non-rational individuals such as rocks and thermometers. For example, the belief of the little boy who believes there will be a snowy Christmas this year is incorrect, even if there really will be a white Christmas. That is, even if his belief is true, he still does not believe as he should. The rock, or even the thermometer, which represents the weather, are closed to this type of evaluation since they do not have attitudes. Burge believes that mature humans, young children, and some non- human animals are rational in the sense that (1) they have attitudes (2) their attitudes are 46 usually based on reasons and (3) their attitudes might be rational or irrational depending on their reasons for their attitudes. Burge does not think that just being rational—having attitudes that are based on reasons—entails having second-order judgments that are epistemically supported. For Burge, this requires a heightened rational capacity, which mature humans, but not children and animals, sometimes enjoy. Burge calls this “critical rationality.” His thesis is that “our epistemic warrant for our judgments about our thoughts…derives from the nature of the thinker as a critical reasoner” (91). Before explaining Burge’s arguments for his thesis, it will be helpful to understand what Burge means by “critical rationality.” He defines the term of art as follows. As a critical reasoner, one not only reasons. One recognizes reasons as reasons. One evaluates, checks, weighs, criticizes, supplements one's reasons and reasoning. Clearly, this requires a second-order ability to think about thought contents or propositions, and rational relations among them. (98) While Burge thinks that only mature adults are capable of critical reasoning, and that they sometimes merely reason non-critically (that is, “blindly” without “appreciating reasons as reasons”) he thinks critical reasoning is common practice among adults (99). He does not strictly mark out where blind reasoning turns into critical reasoning. He does say that any individual who uses the term “therefore” is critically reasoning since such use requires “some conception of validity”, which in turn “requires an ability to think of the propositions in a proof as constituting reasons for what follows from them” (ibid). To illustrate the contrast, an example of blind reasoning might be when you get up from the couch after watching a weather report that calls for snow and begin to salt the driveway. Your behavior shows that you formed a belief that it will snow from watching the weather report—you reasoned to your belief. And your snow belief is rational. The 47 weather report gives you good reason to believe it. But, in a way, you are blind to your reasoning. You do not, at the moment that you rise from the couch, appreciate that the testimony of the weather report serves as your reason for the snow belief. You are thinking about other things, perhaps about where the salt is in the garage and whether you need to put on an extra coat. So, you don’t appreciate the testimony as your reason; you do not appreciate your reason as a reason. Your blind reasoning might escalate into critical reasoning without you having an explicit second-order judgment like “the weather reports testimony gives me reason to think it will snow.” Suppose you have a toddler who is at the charming stage of development where she endlessly asks “why” questions. She waddles out to you as you salt the driveway. In responding to her questioning, you begin to critically reason. “Why are you salting?” she asks. “It’s going to snow tomorrow.” “Why?” “That’s what weather report said, so I think it will. They usually get it right.” When you respond to the second question, you are critically reasoning. You are thinking about—and evaluating—whether your snow belief is justified. Burge explains how critical reasoning need not always involve an explicit focus on attitudes: [Critical reasoning] also involves an ability to assess the truth and reasonability of reasoning-hence attitudes. This is not to say that critical reasoning must focus on attitudes, as opposed to their subject matter. Normally we reason not about ourselves but about the world or about practical goods. But to be fully a critical reasoner, one must be able to—and sometimes actually—identify, distinguish, evaluate propositions as asserted, denied, hypothesized or merely considered. (99-100) I read Burge as thinking that critical reasoning is incremental. Once a reasoner appreciates her reasons as reasons, she might appreciate her reasons as reasons in varying degrees in the way that I might appreciate a friend’s generosity more or less. 50 For Burge, critical reasoners, who can control their attitudes by critically reviewing whether they meet certain rational standards, are epistemically responsible for their attitudes in a way that merely rational creatures are not. We are epistemically responsible only because we are capable of reviewing our reasons and reasoning. And we are paradigmatically responsible for our reasons when we check and review them in the course of critical reasoning. (111) We usually imply that various people are responsible for the rationality of their attitudes around the same time that someone refers in conversation to a standard of rationality. Recall the protester who yells at a climate change denier, “you have no respect for scientific evidence!” Besides evaluating the rationality of the climate denier’s attitudes, the protester is blaming the climate denier for not doing so herself. And it is not just moral blame. The denier is in part being blamed epistemically for failing to control and critically review her climate beliefs. On the other hand, merely rational individuals cannot be blamed for their attitudes. It would be a category error to fault the little boy for believing in a white Christmas because there was one last year. He is too young to appreciate his belief about Christmas last year as his reason. He is not psychologically developed enough to be able to critically review whether his reason is good evidence. To take stock, we just seen what, on Burge’s view, critical reasoning is, which he thinks is a fairly common practice among adults. The meditator is a prime example of a critical reasoner. He reviews his beliefs and the reasons on which he holds them. He controls his beliefs through the application of the maxim of doubt, and this leads him to consider skeptical reasons for thinking his beliefs may be false. This procedure of doubt requires the meditator to make second-order judgments to survey his belief system and fully apply the maxim of doubt. 51 There is more in Burge, however, that will be useful to us. The aim of this chapter is to reconstruct an argument in the Meditations for CM. In articulating the concept of critical reasoning, which requires second-order judgments, Burge has not yet provided any argument that the second-order judgments of critical reasoners are themselves rational or certain. I now turn to Burge’s arguments. In the next section, I will show how the meditator can be reconstructed as making similar arguments. First, a word about the type of argument Burge employs. Although Burge does not characterize his arguments as transcendental arguments, several philosophers have interpreted them as such.34 They might be thought to be transcendental in the following sense. First, Burge makes the assumption that we are capable of critical reasoning, and no argument is given for this assumption: “All of us, even sceptics among us, recognize a practice of critical reasoning” (98). Then, without relying on the assumption that we are capable of critical reasoning, Burge reasons to the intermediate conclusion that a necessary condition for being a critical reasoner is that our second-order judgments are rational. Finally, with the transcendental assumption that we are capable of critical reasoning in place, he concludes that we do have rational second-order judgments. Second, a word on the scope of Burge’s arguments. Burge limits the scope of his arguments only to establishing that we have epistemically supported second-order judgments about our propositional attitudes. He does not apply his transcendental arguments to our second-order judgments about mental states such as “imaging, remembering, or reasoning about sensed inner-goings on” (104, Burge’s emphasis). He thinks our second-order judgments about our daydreams, memory states, pain states, or 34 Gertler (5) and Smithies (81) categorize Burge’s arguments as transcendental. 52 sensations such as the redness of my visual experience when I look at an apple, and the like, are rational. However, the reason for the epistemic support of these kinds of second- order judgments is not our status as critical reasoners. Burge’s arguments apply to second- order judgments about the mental states that we critically review while critical reasoning about what to believe or do. These are judgments about our attitudes—beliefs, desires, intentions, and whatever other stances we might take to propositions. Burge usually speaks of beliefs, and this attitude is most pertinent to my discussion of Descartes, so I will confine my discussion of Burge’s argument to second-order judgments about one’s beliefs. The first thing Burge intends to show is that we are epistemically entitled to our second-order judgments. Burge contrasts epistemic entitlement, a term he introduces, with justification. For Burge, justification and entitlement are distinct species of “epistemic warrant” (the property of a true belief that makes it count as knowledge)—what I have been calling “rational” or “epistemically supported” (versus irrational/unsupported) belief.35 When someone’s belief is justified, if pressed they are usually able to access and articulate the reason on which they hold their belief. On the other hand, the kinds of beliefs we are entitled to are those that we cannot access or articulate reasons for without “extreme philosophical difficulty” (93). Moreover, “this articulation need not be part of the repertoire of the individual that has the entitlement” (ibid). So, the contrast concerns whether a believer can access and articulate the reason she holds her belief.36 35 Justification and entitlement are also non-overlapping properties. None of the sorts of beliefs that are entitled (most second-order beliefs, e.g. “My reason is that I saw the weather report” and non-inferred perceptual beliefs, e.g. “I’m looking at the weather report on TV”) can be justified. And conversely, none of the sorts of beliefs that are justified (e.g. “It will snow tomorrow”) can be entitled. 36 Assuming the believer can reason critically. Some individuals, like young children, will have justified beliefs whose reasons they cannot access. 55 Burge also wishes to demonstrate that the rationality and truth of second-order judgments in a critical review are inseparable. To establish that these second-order judgments are always true—that is, they cannot be in error (false)—Burge uses a thought experiment to reduce to absurdity a contrary view that says a critically rational agent could during a critical review have a false second-order judgment about her first- order beliefs. He asks us to consider what he calls a “simple observational view of self- knowledge.”38 On this view, we know about our first-order mental states by way of a contingent causal mechanism, like a mental scanner. The scanner produces second-order judgments that give us access to our mental life. But, being a contingent process, malfunction is possible. And in the event that the scanner malfunctions, we would have false second-order judgments about our lower-level mental life. Burge says: Not all one's knowledge of one's propositional attitudes can fit the simple observational model. For general application of the model is incompatible with the function of knowledge of one's own attitudes in critical reasoning. The main idea is that such application would entail a dissociation between cognitive review and the thoughts reviewed that is incompatible with norms of epistemic reasonability that are basic to all critical inquiry, including empirical, mathematical, philosophical, and practical inquiry. (108) Burge says this incompatibility makes the simple observational model “nonsense” and concludes that the second-order judgments of a critically rational agent are always true. Here is my best description of Burge’s thought experiment designed to bring out the absurdity of the simple observational model.39 The thought experiment will involve 38 By “simple observational view of self-knowledge”, Burge is referring to the inner sense theory of self-knowledge, according to which, we know our mind by way of a quasi- perceptual mechanism, like a mental scanner, which is liable to malfunction. The inner sense theory/simple observational view was defended by David Armstrong (323-338) and William Lycan (ch 4). 39 This is a simplified version of Burge’s example of the guilty suspect. See 110. 56 the case described on page 46, where you blindly reason to the snow belief and then are moved to critically reflect on the snow belief by the toddler. So, in the thought experiment it is assumed for sake of argument that you do have the first-order belief that it will snow tomorrow and you reason is that the weatherman said it would snow tomorrow. Suppose the simple observational model is true. Also suppose that you are a perfectly rational critical and blind reasoner.40 Part 1. Consider when you reason blindly to the conclusion that it will snow tomorrow on the basis that the weather report said it would snow. Everyone should agree that you are justified in holding the snow belief, for it satisfies the relevant norm of rationality: one’s beliefs should fit one’s evidence. Thus, you should hold the snow belief (as you do in the case, by hypothesis). Part 2. Consider when the toddler moves you to critically reflect on your snow belief. However, in this case your mental scanner is broken. So, while you form the true second-order belief “I believe it will snow”, you also form the false second-order belief “My reason for my snow belief is that I wish it will snow.” Based strictly on the information you have in this critical review, everyone should agree that you are irrational in holding the snow belief. Thus, you have not satisfied the relevant norm of rationality—you should dump the snow belief from your belief system.41 40 I.e. in blind reasoning, you always satisfy the norms of reason. In critical reasoning, you always apply the norms of reason with perfect accuracy to what you (second-order) believe are your first-order attitudes. 41 We are supposing that you have no other relevant evidence regarding tomorrow’s forecast besides the weather report. 57 But it would be absurd to say that it is true of you both that you should and should not hold the snow belief. So, the simple observational model is false. Thus, the second-order judgments of a critically rational agent can never be in error. So, Burge has given a transcendental argument that critically rational agents are epistemically entitled to the second-order judgments that function in a critical review of their beliefs and reasons. Burge has also purportedly shown through a thought experiment that a critically rational agent’s second-order judgments must be true. IV. The Meditator as a Critical Reasoner Having introduced Burge’s concept of critical rationality, we can view the meditator as a critically rational agent. His second-order judgments make up a critical review in which he articulates his beliefs and reasons as his beliefs and reasons. He evaluates them against a norm of reasoning: believe nothing that can be doubted. All this second-order mental activity is done with the aim of controlling and massively overhauling his belief system. In this section, I will show that we can reconstruct the meditator as making an argument for CM much like Burge’s transcendental argument for the claim that we are entitled to our second-order judgments. The argument can be gleaned primarily from the two opening paragraphs of the first meditation. As a first step towards this reconstruction, the meditator assumes without argument that he is critically rational. Take the first two paragraphs of the Meditations. I quote the passage at length here because I refer to it at various points in this section. Some years ago I was struck by the large number of falsehoods that I had accepted as true in my childhood, and by the highly doubtful nature of the whole edifice that I had subsequently based on them. I realized that it was necessary, once in the course of my life, to demolish everything completely and start again right from the 60 endowed with reason comes with a capacity to reflectively criticize one’s opinions. After Descartes discusses the opinions he acquired in early life, he writes: But after I had spent some years pursuing these studies in the book of the world and trying to gain some experience, I resolved one day to undertake studies within myself too and to use all the powers of my mind in choosing the paths I should follow. (AT VI 10; CSMK 1:116) On the next page, Descartes illustrates what it means to “use all the powers of my mind in choosing the paths I should follow”: regarding the opinions to which I had hitherto given credence, I thought that I could not do better than undertake to get rid of them, all at one go, in order to replace them afterwards with better ones, or with the same ones once I had squared them with the standards of reason. (AT VI 13-14; CSMK 1:117) Descartes seems to think that possessing “reason” comes with the capacity to critically reason. As I read him, Descartes equates the “right path” a mind can take with a norm of reasoning. In the last quoted passage, he suggests that someone with reason can apprehend the norms of reason and apply them to their opinions during a critical review. In this way, the paths offer the critical reviewer guidance about what to think. As is clear in the first Discourse passage, Descartes also thinks that these paths or norms determine whether an opinion is reasonable or not. The assumption also comes through in the Principles. There, Descartes says the first principle of human knowledge is that “the seeker after truth must, once in the course of his life, doubt everything, as far as is possible”: Since we began life as infants, and made various judgements concerning the things that can be perceived by the senses before we had the full use of our reason, there are many preconceived opinions that keep us from knowledge of the truth. It seems that the only way of freeing ourselves from these opinions is to make the effort, once in the course of our life, to doubt everything which we find to contain even the smallest suspicion of uncertainty. (AT VIII 5; CSMK 1:193) 61 Here, Descartes once again is in agreement with Burge that adults are capable of critically reasoning. Adults can recognize their beliefs as their own and judge their epistemic credentials. These skills are used to their full extent during the enterprise of doubt. Further evidence that Descartes shares Burge’s assumption that adults are critical reasoners is revealed by similarities in their views about the reasoning capacities of children. For, Descartes states that babies make judgments and thus come to have “preconceived opinions.” He also thinks that infants have reasons for their beliefs, though very dubitable ones that they cannot reflect on. At the end of book one of the Principles, Descartes provides an elaborate theory of the developmental changes in the mind-body interactions of very young children and infants. Descartes hypothesizes that since the minds of very young children are “so closely tied to the body”, they are disposed to falsely judge that sensations are properties of external world objects on the basis of dubious metaphysical reasons (AT VIII 35-36; CSMK 1:218-219).42 Like Burge, Descartes holds that children are rational—but not critically rational—agents since they hold beliefs for reasons but cannot critically reflect on those reasons. We have seen that that the meditator assumes that he is critically rational. This is the first premise in Burge’s transcendental argument for the claim that the second-order 42 Descartes provides an earlier version of this view that infancy is the origin of false beliefs in a letter responding to an interlocutor under the alias “Hyperaspistes” (see Letter to Hyperaspistes, August 1641. AT III 3:424). Descartes says that “the human soul, wherever it be, even in the mother’s womb is always thinking.” He goes on to say that every baby has the innate ideas of God, himself, and self-evident truths in his mind although a baby either perceives them in a “confused” way or attends primarily to things “presented by the senses” rather than the innate ideas. Yet, if a baby were “released from the prison of the body, it would find [the innate ideas] within itself.” 62 judgments that feature in critical reasoning are epistemically supported and entitled. Recall Burge’s argument: (1) We are critical reasoners. (Assumption for Trans. Argument) (2) Thus, we are responsible for critically reviewing our attitudes. (1, by def.) (3) If someone is responsible for critically reviewing their attitudes, they are required to form second-order judgments about their attitudes. (Premise) (4) If S has a responsibility to X and S is required to Y to X, then S is entitled to Y. (Res. Implies Entl. Principle) (5) Therefore, we are entitled to our second-order judgments. (2,3,4) Having presented evidence that the meditator accepts premise (1), I wish to examine more closely the opening paragraphs of the first meditation, which are quoted on pages 57-58. I think a close reading of the second paragraph of the first meditation shows that meditator accepts the second and third premises of Burge’s argument. Earlier, we saw that Burge understands critical reasoners not only to have the real psychological capacity to critically review their beliefs against norms of reason but also to be epistemically responsible for their beliefs. In other words, critical reasoners are worthy of praise or blame depending on how they use their critical reasoning abilities to control their belief system. Burge builds this responsibility into the definition of a critical reasoner. Now, one possible position is to acknowledge that we are critical reasoners, those capable of subjecting our beliefs to critical review, but to reject that we are responsible for our beliefs in the way that Burge suggests. Like Burge, the meditator—and Descartes—works epistemic responsibility into the definition of a critical reasoner. However, the meditator goes further, holding that we have an epistemic responsibility and obligation to apply the stringent norm of doubt to our belief system at some time during our life. The meditator says that is “necessary” for him to doubt 65 his claim that we are responsible for avoiding error implies a responsibility to doubt our beliefs given his view that doubt rids ourselves from holding erroneously formed beliefs. Outside of the Meditations, Descartes repeatedly suggests that we have an obligation to seek the truth and that this is accomplished through extensive critical doubt. To Gassendi, who in the Fifth Set of Objections says he approves of Descartes’ project of freeing the mind from all preconceived opinions, Descartes responds: “Indeed, no one can pretend that such a project should not be approved of” (AT VII 348; CSMK 2:241-242). A few lines later, Descartes writes that the enterprise of doubt is a project that “everyone agrees should be performed.” Later, in the preface to the French edition of the Principles, Descartes bids newcomers to philosophy, which in his view starts with doubt, a necessary first step in seeking wisdom. He motivates critical reason-based philosophy by arguing that our status as minded beings carries an obligation to philosophize: The brute beasts, who have only their bodies to preserve, are continually occupied in looking for food to nourish them; but human beings, whose most important part is the mind, should devote their main efforts to the search for wisdom, which is the true food of the mind. (AT IX 4; CSMK 1:180) And a few lines later: Now this supreme good, considered by natural reason without the light of faith, is nothing other than the knowledge of the truth through its first causes, that is to say wisdom, of which philosophy is the study. (AT IX 4; CSMK 1:180-181) For Descartes, the best thing that we can do is discover philosophical knowledge, scientia, the true principles that serve as the structure of other non-basic knowledge. Since this wisdom is the supreme good, we have an obligation and responsibility to search for it. It’s plausible that Descartes thinks our responsibilities to avoid error and doubt our beliefs are 66 derived from our responsibility to search for wisdom, our ultimate responsibility. Perhaps, Descartes might argue for our responsibility to critically review our attitudes as follows:43 (1) It is our responsibility to search for wisdom. (AT IX 4; CSMK 1:180-181) (2) Thus, it is our responsibility to avoid all error and seek the truth. (from 1; AT VII 58; CSMK 2:40-41) (3) Our belief system is founded on false beliefs from childhood. (AT VIII 35-36; CSMK 1:218-219) (4) Thus, it is our responsibility to critically doubt our beliefs (from 2,3; AT VII 17-18; CSMK 2:12) Having seen that Descartes thinks that we are responsible for critically reviewing our belief system, let us consider whether he accepts the third premise in Burge’s argument, that forming second-order judgments about one’s belief system is required to carry out one’s obligation to critically doubt one’s belief system. In the second paragraph of the first meditation, the meditator suggests that doubt requires inspection of on one’s opinions. For the purpose of rejecting all my opinions, it will be enough if I find in each of them at least some reason for doubt. And to do this I will not need to run through them all individually, which would be an endless task. Once the foundations of a building are undermined, anything built on them collapses of its own accord (AT VII 18; CSMK 2:12. My italics) While the explicit mention of second-order thinking is absent here, the meditator regards reflecting on his beliefs—his basic, foundational beliefs—as a key part of tracking down the beliefs that are to be subjected to doubt. In the Seventh Objections and Replies, Descartes again appears to acknowledge the role of second-order judgments in the method of doubt. To elucidate the position of the 43 I cite the passages to show where Descartes makes each responsibility claim. Of course, the citations in conclusions (2) and (4) are in no way meant to show that Descartes intended the reader to see that a responsibility claim he made in one work followed from a responsibility claim one of his other works. 67 meditator at the opening of the Meditations, Descartes draws an analogy to a person who has a basket of apples and, having learned that some of his apples are rotten, is afraid that the rot may spread to the remaining good juicy apples. Descartes says that it would be prudent for him to tip over the basket, and then “cast his eye over each apple in turn, and pick up and put back in the basket only those he saw to be sound, leaving the others” (AT VII 481; CSMK 2:324. My emphasis). Of course, Descartes means for the apples to represent the meditator’s beliefs.44 Crucially, the language Descartes uses to convey how the man must sort out the apples—by “casting his eye over them” and putting back only “those he saw to be sound” suggests that the method of doubt requires second-order judgments at various stages in the process. Second-order judgments are necessary to first identify one’s foundational beliefs and then make determinations about their epistemic credentials. Thus, from the very beginning of the Meditations, the meditator accepts the premises of an argument that establishes that he is epistemically entitled to second-order judgments. While he neither explicitly links these premises into an argument for that conclusion or comments on the Responsibility-Implies-Rights Principle, we should reconstruct the meditator as making such an argument to defend the reasonableness of his reflective judgments that are part of his process of doubt. Now, it would be highly anachronistic to reconstruct the meditator as concluding, on the basis of this argument, that his second-order judgments are epistemically entitled, 44 One relevant difference between Descartes’ apple sorting example and the process of doubt is that, with the apples, all the apples are inspected. By contrast, in the process of doubt, only basic beliefs, the ones on which the rest of one’s belief system is derived, are reflectively inspected and doubted. 70 psychological certainty. Perhaps one passage in which Descartes speaks more univocally in terms of epistemic certainty arises in the Second Set of Replies: if this conviction is so firm that it is impossible for us ever to have any reason for doubting what we are convinced of, then there are no further questions for us to ask: we have everything that we could reasonably want… such a conviction is clearly the same as the most perfect certainty. (AT VII 144-145; CSMK 2:103. My emphasis.) It what follows, epistemic certainty will be our focus. I will make the case that the meditator has available to him a modified version of Burge’s transcendental argument with conclusion that his second-order judgments are as reasonable as any belief or judgment can be. They are so reasonable that they are true. This modified argument trades on the unusual rigor of the meditator’s critical reasoning required for the method of doubt compared to that discussed by Burge. I will now explain how the meditator’s critical reasoning is especially rigorous. Compare the responsibilities of a mature adult to those of the meditator. Burge tells us that the mature adult, as someone capable of critically reasoning, is epistemically responsible for making his first-order beliefs respect “norms of epistemic reasonability basic to all critical inquiry, including empirical, mathematical, philosophical, and practical inquiry” (108). These norms of reasonability are norms such as “your beliefs should fit your evidence.” That is, on Burge’s view, we are responsible for making our attitudes only reasonable. To fulfill our epistemic responsibility, we must critically review and control our first-order attitudes by inspecting the reasons they are based on. By contrast, the meditator is called on to do so much more. On Descartes’ view, he is responsible for making his first-order beliefs respect the norm of doubt—“believe only between psychological and epistemic certainty in Descartes’ corpus, and how Descartes may even conflate the two notions at times. 71 what is epistemically certain.” In other words, the meditator must make his first-order beliefs maximally reasonable. This requires using the robust norm of doubt rather than a norm such as evidential fit. We can see that the evidential fit norm is too weak for the meditator’s purposes since many of his former opinions—for example, believing that there is a fire before him and believing that the fire is warm—satisfy the norm of evidential fit, but, as the skeptical arguments purportedly show, are still uncertain. Descartes seems to register the stringency of the meditator’s epistemic obligation at least twice in the Meditations. First, in the preface to the Meditations he says that “the route which I follow” in acquiring certainty “is so untrodden and so remote from the normal way” (AT VII 7; CSMK 2:6). The stringency is again acknowledged at AT VII 22; CSMK 2:15, a passage that has come up twice already, where the meditator says that his former, doubtable beliefs are still reasonable. So, the meditator concedes that his former first-order beliefs are epistemically supported since they stand up to the norms of reason used in ordinary critical review. The meditator, however, has ratcheted up the bar of acceptability. We saw earlier that the transcendental argument hinges on a link between one’s responsibility to critically reason and the reasonability of the second-order beliefs that are essential to the review. The idea was that someone couldn’t be responsible for making their first-order beliefs reasonable unless their second-order beliefs about the reasonableness of their first-order beliefs are also reasonable. This inference relies on a principle, the Responsibility Implies Entitlement Principle as I have called it, that links responsibility to reasonability (or entitlement, on Burge’s view). The principle (as applied to specifically to epistemic responsibility rather than moral or political responsibility, and also framed in terms of reasonableness of belief rather than entitlement to belief) is that if someone is 72 epistemically responsible for having one’s belief system satisfy some norm X and satisfying X requires believing Y, then believing Y is reasonable. The same train of thought can be used to argue for the certainty of the meditator’s second-order judgments, given the rigor of his epistemic responsibility. In a compressed form, the idea is that the meditator cannot be epistemically responsible for ensuring that his belief system has only certain first-order beliefs unless his second-order judgments about the lower beliefs are certain. Here is the same point drawn out slightly more. The meditator is responsible for making his belief system adhere to the norm of doubt, the norm that one should believe only what is maximally reasonable. Now, for the meditator to satisfy the norm of doubt, he must have certain second-order judgments as part of a critical review of the epistemic credentials of his first-order beliefs, particularly whether those first-order beliefs are maximally reasonable. So, the meditator’s second-order beliefs that are part of the critical review are also maximally reasonable—certain. So, assuming that the meditator would accept some principle linking epistemic responsibility with reasonability, the meditator has available to him the following transcendental argument in defense of the Certainty of Mind Thesis. (1) We are critical reasoners. (Assumption for Transcendental Argument) (2) Thus, we are responsible for critically reviewing our belief system so that it satisfies the norm of doubt. (From 1, by def.) (3) If someone is responsible for critically reviewing their belief system so that it satisfies the norm of doubt, then they are required to form certain second-order judgments about their attitudes. (Premise) (4) If S has a responsibility to X and fulfilling X requires S to have a certain belief that Y, then S’s belief in Y is certain.47 (Particular version of Res. Implies Ent. 47 This is a specific version of the Responsibility Implies Entitlement Principle formulated in terms of maximal reasonable belief instead of entitlement. On pages 53-54 I discussed the principle and on pages 67-70 I explained how Descartes makes use of the epistemic concept of reasonable belief (and maximally reasonable, i.e. epistemically certain belief). 75 In response to this objection, we will see that Descartes understands the meditator as someone who is more proficient at critical reasoning than most mature adults. In fact, the meditator is so proficient at critical reasoning that he cannot become distracted or mentally fatigued; he lacks the cognitive weaknesses that open him up to second-order error. Now, the transcendental argument only attributes second-order certainty to those who have reached the meditator’s ability to hyper-critically reason. So, the objection is misguided when it claims that the transcendental argument attributes second-order certainty to any critical reasoner who attempts to doubt their beliefs. Let us see where Descartes describes the meditator as an elite critical reasoner. One begins to see Descartes making this point in the Preface to the French Edition. There, Descartes is very clear that the Meditations is for a highly select audience: I am also going to deal with the foundations of First Philosophy in its entirety. But I do not expect any popular approval, or indeed any wide audience. On the contrary I would not urge anyone to read this book except those who are able and willing to meditate seriously with me, and to withdraw their minds from the senses and from all preconceived opinions. Such readers, as I well know, are few and far between. (AT VII 9; CSMK 2:8) Descartes even begins the Preface by telling the reader of the Meditations that he chose not to publish the first edition in French because he didn’t want masses of weak minded people to get ahold of it, only hardened academics conversant in Latin. The issues seemed to me of such great importance that I considered they ought to be dealt with more than once; and the route which I follow in explaining them is so untrodden and so remote from the normal way, that I thought it would not be helpful to give a full account of it in a book written in French and designed to be read by all and sundry, in case weaker intellects might believe that they ought to set out on the same path. (AT VII 7; CSMK 2:6-7) Previously I argued that Descartes thinks that the masses are in fact critical reasoners, but who usually apply (merely) norms of reasonable belief such as the norm of evidential fit, but not the norm of doubt (see page 59-61). If that argument is correct, then Descartes isn’t 76 here saying that many humans are incapable of critically reflecting on their first-order beliefs. So, Descartes means something else when he calls many intellects “weak.” For Descartes, the mind can gain certain knowledge not by improving its capacity to reason or critically reason, which are already divinely constructed perfect capacities. Rather, the human mind acquires certain knowledge by carefully attending to itself and focusing solely on using its rational powers to the fullest. Descartes articulates this point in the Second Set of Replies, where he says that the Meditations are written in what he calls the “analytic” style of exposition: Analysis shows the true way by means of which the thing in question was discovered methodically and as it were a priori, so that if the reader is willing to follow it and give sufficient attention to all points, he will make the thing his own and understand it just as perfectly as if he had discovered it for himself. But this method contains nothing to compel belief in an argumentative or inattentive reader; for if he fails to attend even to the smallest point, he will not see the necessity of the conclusion. (AT VII 155-156; CSMK 2:110) In his view, this is the style of thinking used by ancient mathematicians when they made great discoveries in geometry (AT VII 156; CSMK 2:111). As a practitioner of the analytic method, the meditator enters the project of doubt by giving his first-order beliefs his undivided attention and critically reviewing them as meticulously as is humanly possible. Descartes even says that anyone engaging in the method of doubt should “devote several months, or at least weeks” to the skeptical arguments before going on to seek positive knowledge (AT VII 130; CSMK 2:94). So, a true imitation of the meditator’s method of doubt requires more than being a critical reasoner or even critically reasoning with the rigorous norm of doubt. One must also think about one’s mental life with complete attention and uninterrupted thought. While Descartes thinks that everyone can do this with enough practice and time, and everyone is 77 even obligated to since we are all responsible for attaining certainty and wisdom, only those few who really do follow the method make certain second-order judgements about their mind. 80 in any way. Let us see how these passages might contradict one another. Below is Article 10. I have italicized the line Burman is most plausibly referring to. For some context, in the Principles Descartes first asserts “I think, therefore I am” in Article 7 of Part I; Article 10 might be read as clarifying Article 7. 10. Matters which are very simple and self-evident are only rendered more obscure by logical definitions, and should not be counted as items of knowledge which it takes effort to acquire. I shall not here explain many of the other terms which I have already used or will use in what follows, because they seem to me to be sufficiently self-evident. I have often noticed that philosophers make the mistake of employing logical definitions in an attempt to explain what was already very simple and self-evident; the result is that they only make matters more obscure. And when I said that the proposition I am thinking, therefore I exist is the first and most certain of all to occur to anyone who philosophizes in an orderly way, I did not in saying that deny that one must first know what thought, existence and certainty are, and that it is impossible that that which thinks should not exist, and so forth. But because these are very simple notions, and ones which on their own provide us with no knowledge of anything that exists, I did not think they needed to be listed. (AT VIII 7; CSMK 1:194-195) This seems to suggest that, to establish sum, one must know that “it is impossible that that which thinks should not exist.” Assuming that the italicized claim “it is impossible that that which thinks should not exist” is semantically equivalent to “whatever thinks exists,” it’s straightforward how Burman took Article 10 to be in tension with the denial Descartes makes in the Second Replies. Namely, Article 10 asserts that the cogito argument requires the premise “Whatever thinks exists”, while the denial from the Second Replies denies that the cogito argument requires that premise. In his response to Burman, Descartes seems to regard Burman as articulating the tension between the Second Replies and the Principles in this way (which is entirely resolvable, in Descartes’ view).49 49 See Cottingham (xx-xxii) for a longer argument in support of this way of framing the tension that Burman poses to Descartes. 81 We have seen two ways in which Descartes’ denial that sum is established by means of syllogism is in tension with an understanding of Descartes as establishing his existence by making an argument. One way is that the denial rejects an understanding of the cogito, like that suggested by the naïve interpretation, where it is an argument. The denial should be taken this way if Descartes uses the term “syllogism” to mean an argument. A subtler conflict is that the denial is in tension with an understanding of the way in which Descartes elsewhere seems to suggest that “I think” must be combined with the supporting premise “Whatever thinks exists” to successfully establish “I exist.” In the remainder of this section, I will argue that these two tensions dissolve on a close analysis of Descartes’ remarks, the first from the Second Replies, the second from the Appendix to the Fifth Replies, and the third from his Conversation with Burman. More specifically, my analysis will argue for the following three theses. First, against the first tension, Descartes doesn’t deny that one can deploy an argument for sum anywhere in the remarks. Rather, the denial in the Second Replies specifies the type of psychological procedure—the kind of reasoning—one must use to successfully argue from cogito to sum. Namely, to discover sum by arguing from cogito, one must reason from cogito to sum by “intuition” rather than by “deduction” or “syllogism.” Descartes distinguishes these three modes of reasoning in the Regulae, one of his earliest works. He is channeling the three-part distinction in the three remarks. Second, the chronological order of the remarks represents a shift in Descartes’ thinking about whether sum can be discovered through syllogistic reasoning. By the time of his Conversation with Burman, Descartes becomes neutral about whether it is possible to discover sum syllogistically. So, Descartes recants the denial he made in the Second 82 Replies. While, Descartes becomes neutral regarding whether sum can be discovered by syllogism, he still holds, first, that discovering sum through syllogistic reasoning is liable to reasoning errors of which a discovery by intuition is immune, and, second, if sum could be discovered via syllogism, this would only occur in exceptional cases involving a reasoner who is highly attentive and undistracted. Third, against the second tension, Descartes’ response to Burman clarifies that the passage from the Second Replies doesn’t deny that the cogito argument relies on the general premise “Whatever thinks exists” and so is not in conflict with Article 10 of the first section of the Principles. In fact, the Descartes’ Conversation with Burman shows that the most accurate representation of the cogito is an argument of the form: Whatever thinks exists. I think. Therefore, I exist. Furthermore, that the cogito argument is best represented as this argument, which exemplifies the Barbara form, one of the four forms documented in Aristotle’s theory of syllogism, doesn’t imply, for Descartes, that sum must be reasoned to syllogistically. Let us now examine in full the three remarks in which Descartes discusses his views about syllogistic reasoning and their connection to the cogito. The three remarks can be chronologically ordered with the remark beginning with the Second Replies, followed by the remark from the Appendix to the Fifth Objections, and finishing with the Conversation with Burman.50 50 The Second Objection and Replies was published along with the first Latin edition of the Meditations in 1641. The Appendix to the Fifth Set of Objections came into circulation when it was affixed to the Objections and Replies for the publication of the first French edition in 1647, but could have been written as early as 1644 (see the first footnote on AT VII 198; CSMK 2:268). Descartes’ Conversation with Burman is dated 1648. 85 that will serve as the basis for a method to make us better thinkers. What he is not doing is creating three different logical vocabularies in which certain chains of vocabulary make up valid arguments and others do not, in the way that modal logic features different logical vocabulary and valid arguments than propositional logic. Rather, Descartes is interested in what the human mind does when it reasons. More specifically, first, he is interested in the psychological methods that a thinker can use to move from the premise(s) to the conclusion of an argument. Second, he is interested particularly in different ways to reason to conclusions where we discover something new. That is, reasoning where we draw a conclusion about some matter we haven’t decided on before. Third, he is interested in what is psychologically responsible for reasoning that leads to very good outcomes: new conclusions that are true and certain (AT X 366-370; CSMK 1:13-15). Descartes introduces intuition and deduction as two reasoning processes that “arrive at a knowledge of things with no fear of being mistaken” (AT X 368; CSMK 1:14). Reasoning by intuition leaves “no room for doubt about what we are understanding” (ibid). Reasoning by intuition involves having a clear and distinct perception that “the original proposition follows necessarily from the [premises].” In addition, whatever we intuit is certain (if something is uncertain, it either has not or cannot be intuited). One example Descartes gives of a conclusion we can discover via intuitive reasoning is “3+1=2+2.” By considering the two propositions “3+1=4” and “2+2=4”, one can intuit the proposition “3+1=2+2.” While Descartes uses his concept of intuition to explain what happens in the mind when we draw logical consequences, this is an explanation of logical consequences we draw in one mental grasp or a single “movement of thought” as Descartes says (AT X 368- 86 370; CSMK 1:14-15). So, intuition is a way of reasoning through arguments that don’t have many premises (whether an argument can be intuited that has many premises—say more than two or three—might vary from one person to the next: compare the child learning arithmetic to the professional mathematician). For arguments that have more premises— arguments with “remote” conclusions, as Descartes calls them—we use deduction. Deduction is a psychological process that is defined partly in terms of intuition. When we deduce a conclusion from premises, we make a series of intuitions between premises to intermediary conclusions, and so on, in “a continuous and uninterrupted movement of thought” until we draw the main conclusion (ibid). As with intuition, in deduction we form certain beliefs in the conclusion of our reasoning. Yet, the certainty we attain through deduction requires remembering the sequence of intuitions that you have completed. To illustrate how it takes more for the deduced conclusions to be certain than for the intuited ones to be certain, consider two chain link fences: one ten feet long and a second fence stretching out of sight. It’s reasonable to be very confident that all the links in the ten- foot fence are connected because you can see all its links in one glance. But to reasonably be very confident that the extended fence is linked up, more is required. You must walk its length, stop at multiple places to make note of the links, and make a point to remember each stop. So, while intuition and deduction both lead to certainty and cannot be mistaken, intuition is instantaneous and requires nothing more than the clear and distinct perception of very few premises and the conclusion. In deduction, which involves many premises, more is required for certainty. An argument must be reasoned though over time in a continuous series of intuitions, and each step of the argument must be preserved in memory. 87 To best appreciate the significance of Descartes’ denial that the cogito is a syllogism, it will be helpful to further explore Descartes’ contrast between syllogistic reasoning with intuition (and deduction, since deduction, on Descartes’ theory of reasoning, requires a series of intuitions). In the Regulae, Descartes appears to contrast syllogistic reasoning and intuition as two styles of reasoning associated with different accounts of logical consequence. He says that the dialecticians, who teach Aristotle’s theory of syllogism to their students: prescribe certain forms of reasoning in which the conclusions follow with such irresistible necessity that if our reason relies on them, even though it takes, as it were, a rest from considering a particular inference clearly and attentively, it can nevertheless draw a conclusion which is certain simply in virtue of the form. (AT X 405-406; CSMK 1:36) Descartes goes on to critique the dialecticians.51 But, before doing so, in this passage he is plausibly read as attributing to the theory of syllogistic reasoning an account of logical consequence. Perhaps Descartes thought along the following lines. According to the dialecticians, an argument or inference is valid (or, less technically, “good”) when reason identifies it as having one of the twenty-four accepted forms of argument set down in the 51 When Descartes uses the term “syllogism” or “syllogism in Dialectic”, he is referring to the study of logic that was prominent in medieval scholastic schools (the “logic of the Schools” as he calls it in the preface of the Principles). This late medieval logic largely accepted Aristotle’s theory of syllogism as detailed in the Prior Analytics. While medieval logicians expanded on some of Aristotle’s syllogisms for modal sentences, those having to do with possibility, Aristotle’s syllogisms for non-modal sentences were with little exception treated as a definitive science of valid reasoning. Aristotle’s theory of syllogism developed a system to identify twenty-four valid forms of three-line arguments or “syllogisms” out of two-hundred fifty-six possible arguments forms. The theory of syllogism was not just seen as a scientific accomplishment, but as something that could be taught to make people think better. Like the critical thinking courses that are taught to undergraduates today, in the late middle ages there was a pedagogy called “dialectic” which centered on the idea that students could ameliorate their reasoning skills by learning to use the theory of syllogism. 90 reason why sum is not inferred by syllogism is that, to have done so, he would have to previously known the major premise “Everything that thinks exists.” But to have previously known that and not yet have known sum is incompatible: it would be impossible for there to be a time that Descartes, on the one hand, knew that everything that thinks exists, but, on the other hand, didn’t know that he exists. Descartes thinks that the dialecticians who think that sum can be discovered syllogistically fall into the error of endorsing something that is impossible—that an individual can know the general premise and at the same time not know “I exist.” Descartes, here, is recasting a more general objection leveled towards syllogistic reasoning that he states in the Regulae. The objection is that it’s impossible to use syllogistic reasoning of the form “All X are A. S is an X. Therefore, S is A” to discover new knowledge: Dialecticians are unable to formulate a syllogism with a true conclusion unless they are already in possession of the substance of the conclusion, i.e. unless they have previous knowledge of the very truth deduced in the syllogism. It is obvious therefore that they themselves can learn nothing new from such forms of reasoning, and hence that ordinary dialectic is of no use whatever to those who wish to investigate the truth of things. Its sole advantage is that it sometimes enables us to explain to others arguments which are already known. It should therefore be transferred from philosophy to rhetoric. (AT X 406; CSMK 1:36-37). The objection applies only to certain arguments whose form exemplifies the Barbara form of syllogism, like the classic: All men are mortal. Socrates is a man. Therefore, Socrates is mortal. The apparent problem is that, if, say, I’m not sure whether Socrates is mortal, it would be impossible for me to discover this fact about Socrates by reasoning via syllogism from two other things I know—that all men are mortal, and that Socrates is a man. In arguing so, I 91 wouldn’t have discovered anything new. Since I already knew that all men are mortal, I also knew, at that same time—by that “All men are mortal” premise alone!—that Socrates is mortal. Socrates, after all, is a member of the class of all men. The same objection applies to a formulation of the cogito as: Whatever thinks exists. I think. Therefore, I exist. The objection is not that the argument is invalid. It’s that someone who issues such an argument and knows that the general premise is true could never genuinely discover the conclusion as a fact not previously known. The cogency of Descartes’ objection depends on the view that to know general claims like the major premises “Whatever thinks exists” or “All men are mortal”, one must have extensive knowledge of particulars. Consider some potential requirements for knowing general claims. First, to know a general claim, one must know of (or be acquainted with) all the particulars that are members of the class described by the subject of the general claim (for example, thinkers and men, respectively). Second, of each of these particulars that one knows of, or is acquainted with, one must know that the particular has the thing which the general claim attributes (existence and mortality, respectively). With these two requirements on knowledge of general claims in place, Descartes’ objection appears to carry weight. For, from these requirements, when I know “whatever thinks exists,” I’m acquainted with myself, since I’m one of the thinkers, and, moreover, I know that I’m one of the thinkers that exists since existence is what is being attributed in the general claim. In other words, with these requirements, I already know that I exist. The Cartesian objector then says that sum can never be genuinely discovered via syllogistic 92 reasoning since such reasoning requires one to know “Whatever thinks exists,” and whenever this is known by the individual doing the syllogistic reasoning, that individual already knows that she exists. In fact, in the first and second remarks Descartes looks like he is making a requirement on knowledge of general claims similar to those mentioned in the above paragraph. In the first he says, “It is in the nature of our mind to construct general propositions on the basis of our knowledge of particular ones.” He comes close to restating the point in the second remark with the example of the boy learning geometry. The boy cannot “learn” the general claims that make up the axioms of Euclid’s geometry “unless we show him examples in particular cases.” Given Descartes’ views that we “learn” and “construct” general claims only by first examining particulars, it’s plausible that Descartes, at the time he made the first and second remarks, accepted the requirement that, to know a general claim, one must have substantial knowledge of all the particulars that are in the class discussed in the subject of the general claim. The presence of such a requirement in the first two remarks, which justifies the strong objection against syllogistic reasoning we have been discussing from the Regulae, lends further coherence to the view that Descartes, at the time of writing the first and second remarks, did endorse the strong objection. In the second remark, Descartes qualifies his strong objection against syllogistic reasoning. His parenthetical comment that “we may also reverse the order and deduce other particular truths once we have discovered general ones” allows that sum can be soundly derived from “Whatever thinks exists” and “I think”, just not discovered. So, Descartes’ remark clarifies that he is only rejecting the idea that syllogistic reasoning can lead to discoveries—he does sanction syllogistic reasoning for purposes other than discovering