Almost everyone agrees on what Spinoza and Leibniz are trying to accomplish through their projects; they are clearly attempting to demonstrate metaphysical truths from principles that they take as appropriate foundations. Because it is natural to think of Descartes, Spinoza, and Leibniz as a group, I think there is a tendency to apply a conception of a rationalist project abstracted from their projects onto Descartes’s work, resulting in what I call the naïve reading of the Mediations. Under the naïve reading Descartes too is taken to be searching for philosophical truths, either through transcendental arguments or by using a special cognitive faculty. I argue that, despite their naturalness, such readings of the Mediations face significant problems either in the form of inconsistency with textual evidence or by taking Descartes’s project to have significant structural difficulties, such as the infamous Cartesian circle. Following Frankfurt, I develop an interpretation of the Meditations that takes Descartes to be primarily interested in certainty, not truth, and, following Hatfield, which understands the importance of the Meditations in relation to Descartes’s scientific projects. Specifically I claim that in the Mediations Descartes aimed to provide a foundation for his science that, by being certain, was immune from the kind of radical revision that his science constituted for the Aristotelian worldview. This interpretation has the advantages of fitting with the text and giving his project an interpretation under which it is at least structurally sound (i.e. doesn’t involve reasoning in circles), although it does drive a wedge between Descartes and the rest of the rationalists.
1: The Naïve Reading
That may sound like an unusual reading of Descartes, and since its plausibility is justified partially by the problems facing more natural and more philosophical approaches perhaps it is best to start with them. One popular reading of the First Meditation is to see it as portraying Descartes as concerned primarily with avoiding the possibility of error. There he asserts that “I should hold back my assent from opinions that are not completely certain and indubitable just as carefully as I do from those which are patently false” (7:18), and adds “once the foundations of a building are undermined, anything built upon them collapses of its own accord” (7:18). Admittedly, in context, that remark is directed not at the metaphysical system Descartes intends to construct, but rather at the mediator’s pre-existing belief system. But certainly if Descartes takes these remarks about foundations to be true of one system we have every reason to conclude that he would take it to be true of all systems, and so, taken together, they suggest a reason that might have led Descartes to begin his project by introducing skeptical doubts. Descartes, as we know, wants to establish something that is “stable and likely to last” in the sciences, and so he mustn’t allow any false propositions to be part of his system, especially at the beginning, as these false propositions will constitute an unsound foundation and, when used as a basis for further conclusions, will lead to even more false beliefs.
On the basis of such reasoning it is natural to conclude that Descartes wishes to reject absolutely everything in order to avoid beginning with such unsound foundations. Since it doesn’t seem, at least at this point, that we have any fail-safe way of distinguishing the true from the false (as evidenced by the fact that we do make errors, which we sometimes take as true for long periods of time), the only way to avoid error is simply to reject every belief that we have out of hand. Descartes’s later comparison of our set of beliefs to a basket full of apples (7:481), where bad apples can infect good ones and where we must empty the basket out completely so as to be sure of filling it with only the good ones, seems to support this reading of the Meditations. If we understand his project in this way his next step, the cogito, is in effect to produce a true belief from nothing; despite the fact that we have set aside all our preconceived opinions we are still able to demonstrate that the cogito holds, and thus that our existence is guaranteed, without appeal to any of the beliefs set aside. Thus the cogito is a known good apple which we can put back in the basket, and from there further claims can be built until we have recovered much of what we thought we knew, along with a new Cartesian metaphysics.
This reading, while tempting, is naïve, and one interpretive issue that faces it is that the picture of the Mediations as deductions from certain special propositions that need no foundation is difficult to support textually. But such issues are not the focus of this paper. A more significant problem with the naïve reading is that it takes Descartes’ philosophical project itself to be a substantially naïve enterprise. On reflection it is obvious that to get anywhere philosophically significant with the kind of certainty that Descartes is claimed to be after that certain principles of reason, such as logical laws, must be assumed to hold. Consider, for example, the cogito. Although there are many interpretations of the cogito I will simply consider Broughton’s reconstruction (118-9) as an example of this naïve reading:
1. If I have reason to doubt my existence then I must grant that it may be the case that it is something other than my existence that leads me to doubt that I exist.
2. If I believe I exist because I exist, or because of some deceiving cause, then I exist.
3. If I have reason to doubt my existence then I must grant that I exist.
If premises 1 and 2 are to be taken to establish 3 with certainty, and if the cogito is to be independent of all the beliefs that we have supposedly rejected because of the doubts, then it must be the case that we know that premises 1 and 2 are true in virtue of their logical structure alone and we must know that together they guarantee the truth of the conclusion, 3. Similar claims can be made about almost any reconstruction of the cogito along these lines; no matter how it is claimed that we arrive at it to really endorse “I think, I exist” we must accept some principle that informs us of the truth of the cogito as infallible, or at least beyond doubt. But then we must not have honestly emptied our minds of preconceived opinions as Descartes instructed us to, since surely such principles would have been tossed out with the rest. This observation, in conjunction with the naïve reading of the Meditations, implies that either Descartes was simply blind to his reliance on reason, an implausible mistake for a great philosopher to make, or that Descartes thought that reason was simply incapable of error, which also seems unlikely, since in the course of his formal training in logic surely Descartes must have encountered logical fallacies that were intuitively taken were taken as correct and had to be removed by training.
Of course it is possible for great philosophers to be in error or make mistakes just as mere mortals do, and so it isn’t prime facie impossible that Descartes thought reason was immune from error, and could be assumed to be so from the outset, or that he simply didn’t realize the essential role that reason played. But, unfortunately for the naïve reading of the Meditations, there is textual evidence that undermines these suggestions. In the Replies to the Second Objections, Descartes writes: “What is it if someone should perhaps imagine that the very thing whose truth we have been so firmly persuaded appears false to God or to an angel, and that as a consequence it is false, speaking absolutely? … For we are supposing a persuasion so firm that it can in no way be removed – a persuasion, therefore, that is exactly the same as the most perfect certainty” (7:145). The firm persuasion spoken of in this passage can be naturally interpreted, in context, to be the operation of pure reason (reason applied when all beliefs that might be doubted are set aside), which yields things such as the cogito. And yet in this passage Descartes admits that the conclusions reached in this way might be false, although if they were that we would never know them to be false nor have any grounds for doubting them. In light of that comment it seems unlikely that Descartes took reason to be necessarily immune from error (although he might have thought it was immune to detectable error), for, if he did, this passage would make little sense. If he believed in reason’s absolute infallibility he should have asserted instead that there is simply no way for certain beliefs to be in error, from any perspective.
If Descartes doesn’t take reason to be immune from the possibility of error then, if it is to be consistent, the naïve reading would have us believe that he must have overlooked the role of reason, and thought that he could make do without its principles until he had properly grounded them (lest possibly faulty principles of reason undermine the foundation). But this proposal too conflicts with the text: “the term ‘preconceived opinion’”, Descartes writes, “applies not to all the notions which are in our mind (which I admit is impossible for us to get rid of) but only to all the opinions which we have continued to accept as a result of previous judgments that we have made” (9A:204). It was preconceived opinions that Descartes thought we should rid ourselves of, and so here he is explicitly stating that he didn’t intend to empty his mind of everything at the beginning of the Meditations, but only of a certainly class of beliefs. Since the principles of reason do not follow from previous judgments, they would seem to be exactly the sort of thing that Descartes isn’t insisting that we must rid ourselves of. And, as indicated, they may very well be impossible to do without, since we can’t think or philosophize without them. This leaves the naïve reading of the Meditations without a textual foundations, because Descartes explicitly admits that reason might lead us astray and that reason, among other things, is not to be put aside at the outset of the Mediations, meaning that the system Descartes will establish in this way does not appear to be incapable of being false (or, if it is, at least not as the naïve reading would have it, i.e. because the foundation contains only beliefs that must be true).
2: In Search of Certainty
Thus we are in need of a sophisticated reading of the project of the Mediations to replace our naïve one. And the first step towards such a reading is to take a closer look at how Descartes portrays his project in the First Mediation. In light of the issues just discussed, the primary problem with the naïve reading appears to be that it assumes a conception of Descartes’ project that owes more to a familiarity with the rationalists in general than with Descartes specifically. The naïve reading takes Descartes’s aim to be to deduce metaphysical truths from certain foundations, but that is not what Descartes portrays himself as doing. In the very first paragraph of the First Mediation Descartes establishes that his aim is to establish something that is “stable and likely to last” (7:17). Conspicuously he doesn’t assert that he wishes to establish metaphysical or scientific truths, but rather something that won’t be later significantly revised or overturned, which leaves open the possibility that this stable body of claims might be false in some way. Similarly, Descartes’ maxim that we should withhold our assent from any proposition that we have reason to doubt (7:18), again emphasizes a desire to find something indubitable with no concern to whether what is indubitable is true or false. And at 7:22 he further reiterates the claim that certainty is his goal. At least in the First Meditation then, where Descartes is setting up the project of the Meditations as a whole, certainty seems the primary objective.
This reading of the Meditations, as a search for an indubitable metaphysics by setting aside any foundations that could possibly be called into doubt, seems to fit what Descartes says in the first meditation, but it may appear, at first glance, that it is incompatible with the apple basket metaphor to which he appeals in the Seventh Objections. That passage may seem to provide strong textual support for the naïve reading, and we must find another way to interpret it if an alternative reading of the Meditations is to be consistent with the text. In that passage Descartes claims, at least at first glance, that the motivation for emptying out our minds is to prevent falsity from having any foothold, which seemingly commits Descartes to a search for truth since falsity, if unknown, poses no danger to certainty. Let’s take a look at the passage to see if another reading is possible:
Suppose he had a basket full of apples and, being worried that some of the apples were rotten, wanted to take out the rotten ones to prevent the rot from spreading. How would he proceed? Would he not begin by tipping out the whole lot out of the basket? And would not the next step be to 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? In just the same way, those who have never philosophized correctly have various opinions in their minds which they have begun to store up since childhood, and (1) which they therefore have reason to believe may in many cases be false. (2) They then attempt to separate the false beliefs from the others, so as to prevent their contaminating the rest and making the whole lot uncertain. Now the best way that they can accomplish this is to reject all their beliefs in one go, as if they were all uncertain and false. (3) They can then go over each belief in turn and re-adopt only those they recognize to be true and indubitable. (7:481)
In this passage I have identified three sentences that may appear to pose problems for the reading of the Meditations under development. 1 might seem to imply that the problem with beliefs adopted since childhood is that some of them are false. But a closer look reveals that the real problem is that they have reason to think that those beliefs may be false, which means that they are not indubitable, as Descartes would wish, and thus that anything founded on them would be dubitable as well. This is to emphasize the fact that it is not their falsity that Descartes finds troublesome, which would be a problem in a search for truth, but that they might be questioned, that reasons to distrust them might be found, which is a problem in a search for certainty. But that is not the end of the difficulties this passage raises. In sentence 2 Descartes seems to assert that the method of doubt is a means to accomplish the goal of getting rid of false foundational beliefs. But I think it is reasonable to take that as simply poor wording on the part of Descartes or his translators, because in that same sentence he states that the false beliefs are to be eliminated so that they don’t make the whole collection uncertain. That is not what false beliefs would do; they would make the whole lot false. And so it is better to read this sentence as asserting that the possibly false beliefs, the dubitable ones, should be separated so as to remove uncertainty, which also fits the previous sentence better, given that Descartes has just been talking about a set of beliefs which in many cases may be false (not which are necessarily false). Finally, we must make sense of sentence 3, which appears to claim that the beliefs re-adopted will be both true and indubitable, or that truth and indubitability are the same thing (which obviously contradicts what he says at 7:145, as mentioned previously). It is hard to make sense of this sentence, but what Descartes may have had in mind is that the mediator using his method can re-adopt those beliefs that they “recognize to be true”, and that recognizing a belief as true is not to say that the belief is really true, but only that we can’t help but accept its truth. In other words, that it is indubitable. Thus this reading of the Meditations does seem like it can fit with what Descartes actually wrote in this passage, if not the ideas that it is most tempting to project on to him, but it remains to be seen whether it can be made to cohere with the overall project.
Before moving on, I wish to first say a few words about the meaning of reason and indubitability in this context. People can mean many things when they talk about reason. They can, for example, designate by it the term certain logical norms that they consider supremely rational. But under the interpretation I am developing, those are not the kinds of things that we would want to say are part of reason, and thus which are not emptied out of the mind by Descartes (as mentioned previously in the discussion of the passage at 9A:204). Because people can learn and unlearn such rules it is clear that they are dubitable. So when we talk about reason in this context we must mean something less, specifically something like the operations of the mind or pure intellect that make us simply see a connection of entailment between two ideas or propositions, such that we see one of them as necessarily following from the other (this fits with Descartes’s remark at 7:144 that certain intellectual activities give rise to spontaneous convictions in the truth of ideas, which I take to be the activity of reason). Since reason as such is primitive and innate, and not the result of a previous judgment or training, we don’t have to set it aside as dubitable as a result of the dialectic of doubt, again as mentioned in 9A:204. And we can use this understanding of what reason is to say what is dubitable as well: a claim is dubitable if and only if we can rationally entertain a scenario in which it is false. Obviously this means that the things reason compels us to believe, such as connections of entailment, are themselves indubitable, as reason won’t allow us to entertain a scenario in which they are false. Although Descartes never explicitly embraces such a definition of what is dubitable, it coheres with his method of doubt, given that he throws claims into doubt by giving explanations of how they could possibly be false.
3: Interpretive Questions and Difficulties
With this understanding in place, we can now turn our attention to the problems and unanswered questions that this interpretation of the Mediations naturally raises. One of these problems is that, while we may not be able to doubt any particular claim that reason leads us to, we are fully capable of doubting reason as a whole, suspecting that our faculty of reason may be flawed and lead us to errors. Descartes himself seems fully capable of entertaining such doubts. At 7:145 he comments that reason might be false from some absolute perspective, which seems tantamount to admitting that reason as a whole is dubitable. Furthermore, the skeptical scenario Descartes constructs at 7:21, where he entertains the possibility that we have come about by chance and are highly defective beings, seems to cast doubt on the faculty of reason as well, since it is unlikely that the faculties of a defective being are especially reliable. If we accept the validity of such doubts then it would seem that nothing is indubitable, because even if we accept that we can’t help but accept certain beliefs we could still entertain the idea that those beliefs as a class might be false, in whole or in part, and thus we might be led to withhold our assent from them. And if that is the situation we find ourselves in nothing can get us out, because at that point there are no indubitable claims with which we might “prove” the correctness of reason. Overcoming this problem is one thing that we must do to improve the reading of the Meditations under development.
This reading also raises additional questions. We might wonder what makes the faculty of reason especially important for inquiry, given that there are other mental faculties that also produce spontaneous judgments. The senses, for example, seem to put ideas into our minds that we cannot easily deny. Now the obvious answer is that these other faculties are dubitable, but that response is not totally satisfactory given that reason itself also seems somewhat dubitable. Furthermore, the judgments of the senses are only dubitable in the sense that we may withhold our assent from them given appropriate motivation, although they will always present themselves to us unbidden. And it looks like we might say the same thing about reason, that it presents certain beliefs and inferences to us, but that we might still withhold our assent from them. Secondly, we might wonder what the importance of indubitable ideas is. If indubitably might come apart from the truth, so to speak, it would seem better to have true ideas, even if they might be doubted; certainly there is no logical problem with doing accepting dubitable beliefs as true since we already accept many things that Descartes takes to be dubitable. So to complete our interpretive task we must explain why Descartes aims at indubitable propositions, why reason plays a central role in this endeavor, and find a way to put reason as a whole beyond the reach of doubt or show how that problem might be defeated in some other way.
Let us turn then to other interpretations to see whether they can shed any light on these matters. I begin with Frankfurt’s interpretation in Demons, Dreamers, and Madmen. Frankfurt presents a reading of the Meditations similar to the one being developed here; he too thinks that Descartes is concerned primarily with finding something beyond doubt, and he thinks that Descartes is largely unconcerned with whether what he determines to be certain is true in an “absolute” sense (Frankfurt 25-6). Frankfurt also agrees that reason has a special place within Descartes’s system. He claims that Descartes doesn’t take seriously the possibility that he might be mad because madness undermines the mediator’s capacity for reason (Frankfurt 38). Descartes, on Frankfurt’s reading, is only concerned with reasonable grounds for doubt because reason is necessary for the investigation to be conducted at all, which he concludes partly on the basis of Descartes’ dismissal of madness as a serious possibility when considering grounds for doubt in the First Meditation (7:19). This provides an answer to one of the interpretative questions raised previously: reason has a special status because Descartes sees it as a precondition for our ability to conduct the Meditations. Frankfurt also has a suggestion as to why Descartes is after certain beliefs and not those that are “absolutely” true; he suggests that, since certainty as understood by Descartes entails coherence, certainty is all we really want of our beliefs because, if they cohere, we can never be in a situation where they can fail us or where we can be made aware of their absolute falsity (Frankfurt 179). Both suggestions seem like natural additions to the interpretation under development.
But none of these suggestions help deal with the problem created by the dubitability of reason itself. We can’t, for example, resist entertaining such doubts because we deem reason necessary. That justifies, at the most, proceeding as if reason were reliable, it doesn’t justify concluding that it is necessarily reliable. Nor might we appeal to the coherence of reason as sufficient, regardless of whether it leads us astray in other ways, because we might doubt that coherence; we may simply judge that it is coherent as a result of one of its flaws while inconsistencies lie in wait just beyond the bend. And, while we are on the topic of the shortcomings of Frankfurt’s interpretation, we might also question whether coherence is a worthwhile goal given that what our beliefs have to cohere with is allowed to be redefined (Descartes, for example, deemphasizes coherence with the senses). It is easy to construct toy systems that are as useless as they are coherent.
Of course not all the experts would agree either with Frankfurt’s interpretation of Descartes or the variation under development. Hatfield, for example, construes Descartes project differently. He acknowledges that passages such as 7:145 and the wording of the First Meditation may lead us to believe that Descartes is seeking certainty instead of truth, and that such a reading of the Meditations points towards a way of avoiding the infamous Cartesian circle, since Descartes would only be aiming to demonstrate that we can be certain of god’s existence, which requires no questionable appeal to claims that are supported by god’s existence (Hatfield 172). However, Hatfield thinks that the existence of the divine guarantee, found in the Fourth Meditation, rules out that interpretation. It is beyond the scope of this paper to investigate exactly what Descartes is doing in the Fourth Meditation, but, given that Hatfield sees it as posing an interpretive problem, some treatment of it must be provided. I take it that it is not in dispute that going into the Fourth Meditation, before the truth rule is in play, that all we can say about our conclusions is that we are certain of them, which allows that they may be false in some absolute sense (if we could conclude that they were true at this point then there would be no need to establish a truth rule, we would already have such a rule). And in the Fourth Meditation Descartes concludes that clear and distinct perceptions are true (7:62). Putting those two claims together, what we can conclude is that the Fourth Meditation purports to establish that it is certain that clear and distinct perceptions are true. But that doesn’t mean that they are true, because, as has already been noted, things that we are certain of might be false, and so it may be that the conclusion of the Fourth Meditation is false. In other words, clear and distinct perceptions may mislead, even though we are certain that they won’t. What, then, is the point of the truth rule? Well, if something is true, then it is clear that nothing will contradict it, and so we can take Descartes as saying in the Fourth Meditation that reason will not contradict beliefs we take as certain (remembering that we simply find various ideas and inferences to be certain without justification; thus it is conceivable that they might conflict). Thus the Fourth Meditation can be understood as providing a kind of consistency proof for reason (Frankfurt develops a similar interpretation of the truth rule in his chapter 15). This doesn’t disprove Hatfield’s proposed interpretation of the Fourth Mediation, but it does show that what Descartes says about truth there need not contradict the reading of the mediations under development.
Let’s now consider what Hatfield says about the other interpretive problems facing us, namely whether reason as a whole is dubitable and what Descartes’s goal in establishing an indubitable metaphysics was. While Hatfield recognizes the possibility that reason itself may be cast into doubt, he doesn’t see a way for Descartes to escape this problem either. He writes that Descartes, by his own standards, cannot rule out the defective origins hypothesis by appeal to god, since if true it would throw the existence of god into doubt, nor is he justified in simply presupposing that the intellect functions properly (Hatfield 231). Hatfield does, however, propose a solution to our other interpretive problem. He suggests that it is important to read the Meditations in the context of Descartes’ concerns in physics and with overturning Aristotelianism. Descartes’ physics was based on different principles and a different way of looking at the world than the systems that had preceded it, and thus its very foundations would seem questionable to many. Thus, Hatfield suggests, we can read the Meditations as an attempt to guide the readers to use their faculty of reason, the pure intellect, both in order to clear away the confused principles central to Aristotelian physics and to lend credibility to the foundations of the new physics (Hatfield 232-4, ch. 9). This certainly seems to be a plausible idea, and if we can show how indubitability is essential to that aim we will have solved one of our interpretive problems.
4: Problems Solved and Questions Answered
So while neither of the experts have solutions to all of our interpretive problems they had ideas that solved different problems, and so it falls to us to put those pieces together into a cohesive whole. The best place to begin is by addressing what might be thought of as the central question facing any interpretation of Descartes: what is the aim of the Meditations? Under this interpretation Descartes is taken to be aiming at certainty, but why is certainty to important to him? Well for Descartes certainty is a kind of mental state where belief cannot be shaken by any coherent skeptical scenario. The term skeptical scenario is one that I have borrowed from Janet Broughton’s book Descartes’s Method of Doubt. There she uses the term skeptical scenario to refer to a shared structure present in the doubts raised by Descartes in the First Mediation (Broughton 64). Specifically she states that these doubts all consist of conceivable situations in which the beliefs to be thrown into doubt turn out to be false and where some mechanism is posited that is causing us to have the beliefs that we do despite their falsity. Positing that we might be dreaming, for example, is a skeptical scenario because the waking world could be different than the beliefs that our dreams cause us to have. And so this suggests that Descartes takes avoiding skeptical scenarios to be a worthwhile goal.
With that in mind we can fit the pieces together, beginning with the motivation Hatfield proposed for Descartes: attempting to found a new science and to overthrow what passed for science at the time. But Descartes didn’t just want to establish a new science; the opening lines of the First Meditation reveal that he wanted to establish a lasting science, one that would be free from the kind of overthrow he was attempting to affect of Aristotelian science. But how can we free our science from the possibility of substantial future revisions? Given the nature of his own science, I think Descartes realized that scientific revision only occurs when new theories can be proposed that are worthy in their own right and which explain how the scientists working under the old theory came to falsely believe it to be correct, by demonstrating how things may appear to be as the old theory claimed. After all previous scientists must have had some reasons for espousing the theories that they did, either in the form of evidence or certain metaphysical positions. Proposing a substantially different theory thus requires some explanation of how these old reasons were misleading, otherwise they will seem to contradict the new theory, making it much less plausible. Descartes’s science does this, for example, by proposing a mechanism for vision that involves only extension, and which could lead people to falsely believe that colors exist in the objects themselves as the old science maintained. In a way then every new scientific theory constitutes a kind of skeptical scenario for existing theories: it describes a world in which the previous scientific beliefs turn out to be false and it explains how the evidence apparently supporting those theories was really misleading. Therefore Descartes’s science (or at least its basic principles) could be guaranteed to be free from the possibility of future revision if it could be shown that no skeptical scenario could touch it, since revision requires a new scientific theory that acts as a kind of skeptical scenario for the old science. In other words, certainty precludes scientific revision. Thus we can explain Descartes’ search for certain propositions instead of true propositions. It would appeal to Descartes the scientist to uncover the final science, and it would be greatly reassuring to him to uncover that such a final science was possible; Descartes lived in an era where it was widely accepted that to have knowledge was to have come to beliefs via a process much like deduction, making error impossible. The overthrow of Aristotelianism surely shook such beliefs, perhaps suggesting that knowledge of that kind was impossible and that mankind might have little or no knowledge. Such worries could all be put to rest if the new science had special protections against revision that Aristotelian science had lacked.
Furthermore, following Frankfurt, we can take Descartes to think that the ability to reason correctly is a precondition for constructing skeptical scenarios, especially scientific skeptical scenarios, on the basis of his dismissal of the madman. Thus Descartes’ project is free from worries about our ability to reason, because if we can’t reason then no one can properly construct a science that genuinely counts as a skeptical scenario to his (and proposing an alternative science is implicitly to endorse reason). On the other hand, if reason does function properly, then the Mediations can in fact do what Descartes takes them to, namely to establish the basic principles of his science as indubitable and thus immune from rational revision. Our last interpretive difficulty is thus surmounted: Descartes does not falsely presuppose that reason functions properly, given what his aims are. Now this may seem to contradict what was established earlier, namely that the defective nature doubt raised by Descartes himself casts doubt on reason (since, according to this interpretation, Descartes should properly dismiss possibilities that imply that reason is defective, as he does with the possibility of madness). However, if we attend closely to Descartes’s presentation of the defective nature, it becomes apparent that Descartes uses the same language to talk about it as he did to talk about the deceiving god, implying that in this skeptical scenario we are supposed to be deceived about the truth of various propositions, not that we are unable to reason correctly.
The solution to the interpretive difficulties discussed then is, in a nutshell, to reject considering only Descartes the philosopher, and instead to look at what the Meditations might have meant to Descartes the scientist. If we look only at Descartes the philosopher we are naturally led to think that he must have been after epistemological or metaphysical truths, but there is no reason to think that Descartes the scientist understood the Meditations in such terms. Now I admit that such a reading of the Mediations and its importance to Descartes may seem highly counterintuitive, especially given that Descartes writes in his introductions that he aims to demonstrate certain facts about god and about the immortality of the soul. But given that he fails to demonstrate the immortality of the soul, that such an introduction would make it much more likely for his work to be published, and that he comments in a private letter that the Meditations contains the foundations of his physics, I conclude that such intuitions are not conclusive reasons to reject this reading. On the other hand this reading is recommended by its ability to smoothly handle certain persistent interpretive difficulties; given the aims of Descartes the scientist, he is free to assume that certainty constitutes the end of scientific inquiry and that all possible theories are acceptable to reason as part of what might be thought of as a set of preconditions for the possibility of scientific inquiry. Although such premises might be questioned it wouldn’t be productive for Descartes the scientist to worry about them, since he is ultimately interested in particular scientific claims, not whether scientific claims in general are possible. In this way the notorious Cartesian circle and similar troubles are avoided completely. And surely that speaks in favor of it.
Ultimately then this reading is not motivated by anything Descartes says directly, ultimately it is motivated by problems that certain simpler readings of the text raise. Following Frankfurt this reading solves some of those problems by taking Descartes to interested in certainty, and by taking his claims regarding truth to speak only about how we can be certain that some of our beliefs are true, and not as justifying the earlier stages of the project. And, following Hatfield, this reading takes Descartes’s conception of reason, and thus of certainty, to be tied to a conception of the cognitive faculties, such that reason is not a set of external guiding principles, but rather a feature of how we actually think. Thus, as Hatfield suggests, Descartes may be understood as exploring our cognitive faculties in the Meditations to discover what is certain for us. But this reading departs from both these authors by explaining the aim of the Meditations in terms of Descartes’s scientific endeavors, as a search for a scientific foundation that would not be subject to future revision. This explains why Descartes took the human cognitive faculties associated with reason and the intellect and made them both central and foundational; those faculties constitute a precondition for the very ability to do science. And it explains why Descartes’s made certainty his goal; certainty for Descartes essentially amounts to an immunity to revision or correction. So let us suppose then that we accept this reading because it can solve these problems and because it at least isn’t contradicted by the text. What consequences does that have for our understanding of philosophy as a whole, assuming we take cues about how to philosophize from Descartes? The lesson would seem to be that a philosophy which stands by itself, as the naïve reading took the Meditations to stand by itself, is often problematic; it runs into difficulties involving circularity and sometimes we might have trouble understanding what the point of it all is. And the solution to such difficulties, the new reading seems to suggest, is to take philosophy as essentially connected to non-philosophical projects, both in order to give that philosophy direction and to provide a stable starting point for it.
 Hatfield also takes this passage to indicate that Descartes accepted reason from the outset of the Meditations, even though this passage might be interpreted in a very limited sense as asserting only that some things are left in the mind, not necessarily reason, which may be restricted to a set of notions, possibly ideas, such as of god, of body, etc. Hatfield writes “he acknowledges that the First Meditation doubt could never empty the mind totally; that would be to abandon thinking altogether. In effect, he contends that … the principles of reasoning cannot be negated through a process of doubt. Not only are they called upon in evaluating the reasons for doubt but they also for the indelible structure of the mind.” (92) Obviously this isn’t the only possible reading of the passage, but it fits with it and charitably blocks attributing Descartes a rather elementary oversight.
 Broughton too takes this to be what dubitability is for Descartes. See especially chapter 4 and the discussion of skeptical scenarios.
 Of course such a proof of consistency doesn’t actually prove anything, but given that Hilbert once made a similar mistake, and that Descartes was writing significantly before the advent of formal metamathematical reasoning, I think we can forgive him that error.
Broughton, J. (2002). Descartes’s Method of Doubt. Princeton University Press
Cottingham, J., Stoothoff, R., & Murdoch, D. (1984). The Philosophical Writings of Descartes vol. 2. Cambridge University Press
Frankfurt, F. (1970). Demons, Dreamers, and Madmen. The Bobbs-Merrill Company, Inc.
Hatfield, G. (2003). Descartes and the Meditations. Routledge