Do we have knowledge of the external world? This book discusses the skeptical arguments presented in Descartes' Meditations 1 and 2, as well as how Descartes attempts to refute skepticism by building our knowledge of the external world on the firm footing of clear and distinct ideas. David Hume's "Problem of Induction" shows a serious problem for Foundationalism. Dr. Gina Calderone Table of Contents Descartes' Method of Doubt o I think, therefore I am. o Plato's Allegory of the Cave o Does the mind create reality? Cartesian Foundationalism Our Epistemic Situation The Edifice of Knowledge The Cartesian Circle Problem of Induction o Hume's Argument o How Might Any Claim Be Justified? o Are Causal Laws Justified? o Is PUN Justified? o Problem of Induction Chart Descartes' Method of Doubt Read Descartes' Meditations 1 and 2. For an online copy of the entire text of Meditations on First Philosophy, click here. Seeing something like a table is generally taken to be good reason for thinking that the table exists. Moreover, the information contained in perception is generally taken to be good reason for asserting truths about the table itself that it is flat, rectangular, and brown, for example. But how can we be sure that the table is actually there in the first place? Descartes observes that we know ourselves capable of dreams, some of which seem to be quite real, peopled with familiar objects like tables and chairs and acquaintances and strangers and all kinds of action. In short, in dreams it seems like we are doing things and seeing things but we re not. Since, at least while you re dreaming, you can t always tell it s a dream, how can you know you aren t dreaming now?
An analogous problem arises when we think about claims we feel very certain about--those that, as Descartes puts it, are "clear and distinct": that triangles have 3 sides, for example, and 2+2=4. Is the kind of subjective certainty we feel any indication of the truth of these propositions? One would think so, but Descartes considers the possibility that he was created by an evil genius bent on deceiving him about absolutely everything. If so, one might be utterly unable to imagine a triangle that does not have 3 sides but the power of this evil genius is such that we are nevertheless deceived that is, in spite of the impossibility of imagining one, there are triangles with, say, no sides. Since there's no way to tell that we weren't created by this evil genius, we have no guarantee that, in spite of the impossibility of imagining otherwise, 2+2 is actually 4. These two arguments seem to cast doubt on all of our a posteriori beliefs (those we know on the basis of experience are called into question by the dream argument ), and even our a priori beliefs (those we know on the basis of reflection alone are called into question by the evil deceiver argument ). Now, if knowledge requires that we cannot be mistaken about what we believe, and these arguments show that we can always be mistaken about what we believe, then we have no knowledge whatsoever. The idea that we have no knowledge whatsoever is known as philosophical skepticism. I think, therefore I am. In Meditation 2, Descartes ponders the possibility that he can now doubt even his own existence. Perhaps he is dreaming or under the influence of an evil deceiver, in which case. Right. In order to be dreaming or under the influence of an evil deceiver, you must exist. Hence, Descartes concludes, cogito ergo sum (I think, therefore I exist). That is, as long as one is thinking, the thinking is itself proof of your own existence; for there is no story you can tell such that you come to believe (or doubt) the proposition I exist, where according to the story the proposition is false. Of course, knowing that you exist is a far cry from the knowledge we think we have about all kinds of other things, likes tables and triangles and other people. Descartes argument so far gets one no further than solipsism the possibility that you are the only thing that exists. Worse, it gets you no further than the existence of your mind alone, since the existence of your hand indeed, your entire body can be called into question by the evil deceiver argument; so this knowledge is not yet secure. Is there anything that could secure knowledge of the "external world" knowledge, that is, of anything beyond your own thoughts? (For Descartes' answer to this question, proceed to "Cartesian Foundationalism".) Plato's Allegory of the Cave
In Plato's Republic, Socrates imagines this problem another way. And puts it to a different use; but it is still the problem of knowing when one has grasped "reality", where reality is always filtered through, and perhaps distorted by, the senses. Have a look at this video (click here) of Plato's Allegory of the Cave. You're the philosopher: How can one ever know when one has "transcended" the shadows of the cave (or, as we might put it, the limits and distortions of the senses)? Does the mind create reality? Not convinced that the senses are limited and distorting? Click on this link to a great interactive website full of optical illusions. Cartesian Foundationalism Read Meditation 3 of Descartes' Meditations on First Philosophy. For an online copy of the entire text, click here. In the quest for objective certainty, one can be sure of this: I exist (at least as long as one is thinking). In light of the dream argument and the evil deceiver argument, one might still wonder, what am I? That is, what is it about the I I have just proved exists that remains certain in light of the possibility that I am indeed under the influence of an evil deceiver? Descartes argues that all that remains is the content of one s mind: no matter how deceived I may be about how things actually are, it is beyond doubt that, as I sit here now, it seems that I am at this table in this room reading this text. If we have genuine knowledge of the external world, somehow we must be able to certify the link we think there is between the content of the mind and the outside world.
Our Epistemic Situation The Edifice of Knowledge Descartes' strategy is to refute the evil deceiver argument--that is, prove that God exists and is not a deceiver--in which case, since God would not allow one to be mistaken about absolutely everything, one can be sure that when one is in possession of a "clear and distinct" idea (or perception), this idea maps onto the way the world truly is. Let s look more carefully at how this is supposed to go. Consider the following argument: 1. I seem to see a tree. 2. My perception is clear and distinct. 3. All clear and distinct ideas are true. 4. Therefore, there is a tree. Cartesian Foundationalism refers to the view that when, and only when, claims about the external world (such as, There is a tree) rest upon that is, are deduced from objectively certain premises, this is an instance of knowledge. Objectively certain premises are foundational beliefs those that, like I exist, are immune to doubt. All claims about the content of one s mind are foundational in this sense.
Premises 1 and 2 are supposed to be foundational, but 3 says something about the link between clear and distinct perception and the way the world is. So premise 3 is not foundational. Is it justified? If it is, it must be deduced from other foundational (or justified) beliefs. This is why Descartes needs to refute the evil deceiver argument by way of proving that God exists (and is not a deceiver) to establish with certainty that all clear and distinct ideas are true. It s difficult to say exactly how one arrives at a clear and distinct idea, but Descartes argues that this criterion is the mark of accurate perception. The Cartesian Circle Note that since no claim is justified unless it can be deduced from foundational (or other justified) beliefs, the claim that God exists must be proved on this basis. Thus, Descartes gives a version of the Ontological Argument in which he claims that the cause of his idea of God a perfect being must be God. Since this belief isn t itself foundational again, it refers to the link between what s in the mind and what isn t it too must have a proof. Does it? Descartes defense of the claim that his idea of God must be caused by God appeals to its clarity and distinctness, which assures him of its veracity. But this is problem. In order to prove that there is a tree, we need to prove that all clear and distinct ideas are true, but in order to prove that all clear and distinct ideas are true, we need to prove that God exists, but in order to prove that God exists, we need to prove that the cause of the idea of God is God, but in order to prove this we must appeal to the clarity and distinctness criterion, which has not yet been established as guaranteeing truth. This problem has come to be known as The Cartesian Circle. Hume's Problem of Induction Read the excerpt from David Hume's Enquiry Concerning Human Understanding. Whereas knowledge seems to require that the believed proposition is true, which, according to the skeptic, can never be known due to our fallible and limited senses, justification is a weaker notion consisting in the strength of one's evidence and/or reasons for believing something. Thus, you may be justified in believing that, for example, there is a computer on the desk, even if you don't know (since you might be dreaming) that there is a computer on the desk. Of course, the justification of immediate sense perception is more important when one considers that the claims of science depend on it. For example, scientists think that there is black hole at the center of our galaxy; that all life on Earth is related; that the next emerald discovered will be green; and that the laws of nature hold throughout the universe. Induction is the means we have of arriving at most of these claims. Induction--the sort of interest to us here--is a type of inference that moves from evidence to either a prediction about what will happen, or a generalization about what is always the case. Hume's Argument
David Hume was interested in the justification of these predictions and generalizations-- specifically, to what degree are we justified in believing that, for example, the sun will rise tomorrow? No doubt we do believe that the sun will rise tomorrow, and we certainly feel justified in believing this since the sun has always risen in the past. But Hume points out that we would have no justification for this prediction if it weren't for the assumption that what connects the past with future events is the presence of causal laws. (In this case, since gravity holds the Earth in orbit around the sun and inertia keeps the Earth spinning on its axis, we can be reasonably assured that the sun will rise tomorrow.) Thus, our inference from the past to what will happen in the future is only justified if these causal laws are. Causal laws are generalizations. On what basis are we justified in believing in a causal law? How Might Any Claim Be Justified? Here we should consider how any claim might be justified. Every claim, according to Hume, is either a relation of ideas or a matter of fact. Relations of ideas are always justified since they follow from definitions. Matters of fact on the hand are of two sorts: observed and unobserved. Observed matters of fact are justified, since all knowledge stems from this sort of sensory input. Unobserved matters of fact, however, must be justified by an argument, of which there are two types: deductive and inductive. Deduction is a form of inference that is truth-preserving; thus, any conclusion that follows deductively from true premises is justified. What about induction? Since induction is a type of inference that moves from evidence to a prediction or generalization, inductive arguments must presuppose that nature is uniform, in the sense that causal laws hold everywhere and for all time. Thus, inductive arguments are only justified if the principle of the uniformity of nature is justified. So now we might ask: how are causal laws justified? Are Causal Laws Justified? Hume argues that causal laws are not relations of ideas, since they are certainly not true by definition. Nor are they observed matters of fact--since they are supposed to be "laws", they go well beyond the present and past testimony of the senses and say something about how things are generally. Can one give a deductively valid argument for a causal law? No. Our reason for believing that, for example, F=ma (a generalization), is that all experimentation thus far confirms this hypothesis. It is always possible, however, that we will discover it to be false. Thus, this law is supported by inductive reasoning. Now, as noted, all induction presupposes that nature is uniform, so causal laws, since they are all generalizations of this sort, are only as justified as the principle of the uniformity of nature (PUN). Is PUN Justified?
Here we have this claim that seems to be at the root of all claims of science: that nature is uniform. We are justified in believing that the sun will rise tomorrow only if we are justified in believing causal laws which connect past observations with predictions about what will happen in the future, and we are justified in believing in causal laws only if we are justified in believing that nature is uniform. Hume argues that we are NOT justified in believing that they sun will rise tomorrow because we are not justified in believing that nature is uniform. To show this, you need only run the claim, "Nature is uniform" through all the possibilities there are for justifying such a claim. Ultimately, PUN is justified by induction, which presupposes the very thing we are trying to justify. You're the philosopher: Use the chart below to demonstrate that PUN is not justified, and therefore, neither is the claim that the sun will rise tomorrow.