Recommended Reading

B. Williams (1978) Descartes, pp. 32-71, 72-101. Penguin Books.

Text of Descartes

Meditations I, in Perry, J. and M. Bratman (1999) Introduction to Philosophy (3^rd ed.)

Oxford:Oxford University Press, pp. 116-139.

Biographies

Arcesilaus: Born c. 318BC in Pitane. 6^th scholarch of the Academy. Died 243.

Augustine: Born 354 in Tagaste. Bishop of Hippo. Died 430.

Descartes: Born 1596, la-Haye-Descartes. Minor noble family. Died 1650, Stockholm.

Introduction

The system-building of Aristotle has taken us a long way from Socrates’ way of doing philosophy. You’ll recall that his characteristic method of elenchus was intended to make clear that those who claimed to know things did not, in fact, know them. And it did this – for the most part – without replacing those defeated claims with other claims that Socrates would say that he did know were true. At the end of the classic Socratic dialogue one typically finds the participants in a state of aporia/perplexity, not knowing what they had thought they knew. This is clearly not how one is supposed to feel at the conclusion of the Republic of Plato or Aristotle’s Metaphysics.

Nevertheless, the skeptical attitude did not entirely disappear: there were a number of schools of doubters in the ancient world that rejected the knowledge claims of the system-builders. Curiously enough, the most influential of these was the official teaching of the later Academy. That school, you’ll remember, had been founded by Plato and for a long time had defended and extended his metaphysics; but under the guidance of its sixth leader, Arcesilaus, it returned (says Cicero) to the aporetic style of Socrates and the early Plato.

Academic Skepticism

The Academic skeptical arguments began with a critique of the reliability of sense perceptions. This was directed particularly at the claims of a philosophical school called the Stoics, but the critique of senses affects any claim that knowledge is based on the senses. In Aristotle’s system, for example, the first step of any investigation was to collect observations of the matter to be explained, and the further steps would be taken on the basis of the inductions that could be made from those discoveries.

The problem with the senses, as the Academics saw it was that we could never be sure that what we were perceiving was an accurate representation of the world. Their specific argument was that, even if the sense impression was a perfectly accurate representation of an object in the world, we could not be sure that it was not actually caused by some other object, or is not a representation of any existing thing at all. If we knew one of a pair of identical twins, for example, we might think we recognized them when we met them on the street – but we could be mistaken, for we could really be seeing the twin that we don’t know. And the sense impression would tell us nothing to prevent that error. Or, if we were asleep, we might be sure (while we were asleep) that we could see all sorts of fantastic things, dragons and unicorns and so on, and we would be unable to determine just from the sense impressions (if that is the right word) that there were no actual dragons and unicorns being represented, and that we were misinformed about the world.

The significance of this for the Aristotelian system, say, is that it sees knowledge as a structure of deductions built upon fundamental claims about the world that are themselves derived by induction from observations of the world. If the very basis of this structure is untrustworthy then all the rest of our ‘knowledge’ is a house built on sand[1]. The untrustworthiness of sense impressions means that the general claims that we derive from them by induction can’t be trusted either. We can’t go from ‘all the individual swans I’ve seen are white’ to ‘all swans are white’ because we can’t believe that all the swans we saw were white. We can’t even believe that they were swans, or that there was anything to see at all! And if general statements like that can’t be trusted then how can we have any faith in deductions that use them, such as an argument that ‘All swans are white’, Socrates is a swan’ therefore Socrates is white’? Clearly we cannot; and therefore the entire structure of scientific knowledge as Aristotle sees it is made impossible.

So given these uncertainties, what should the philosopher claim? Apparently, nothing at all! One thing in particular the skeptics could not claim, and that is that ‘it is impossible to know anything.’ If they made that claim, then they would be claiming to know something, which is that nothing can be known. This would be a contradiction – and no contradiction can be true. They are usually described as recommending a suspension of judgement. No claims to knowledge should be made: neither that something is the case nor that it is not the case. In fact the investigations of the school (skepsis means ‘investigation’) were directed at putting arguments to both sides of a disputed question so that suspension of judgement could be better justified.

Problems with Academic skepticism

1.       Skepticism and Action

That’s a pretty extreme sort of a position to take, and it occurred to many that it must make it difficult for such philosophers to go about their necessary daily business and to survive in the world. How would it be possible for a skeptic to feed himself without assuming the truth of a large number of statements about the world. For example, if I want to have toast for breakfast I’ll get out the bread and toaster and put the bread in it and turn on the power and press down the slider and so on. But all of those actions are dependent upon my having certain beliefs. I have to believe that I have bread in the pantry, that there is a toaster in the cupboard, that putting the bread in the toaster will result in toast, and so on. All of those beliefs are unjustifiable. Why would I use beliefs I think are unjustifiable in order to guide my actions?

And note that this isn’t just an ad hominem attack, claiming that the skeptics don’t seem to live by their own philosophy. This is a serious objection, because in the ancient world the fundamental problem of philosophy was how to live a good life. Seeking the solution to that problem was what motivated their addressing all the other problems – including the epistemological problems that we’re considering now. If a style of philosophy made it impossible to say anything intelligible about how we should live, that is a serious problem for it.

The skeptics have various responses to this objection. For example:

Arcesilaus asserts that he who suspends judgment about everything will regulate his inclinations and aversion and his actions in general by the rule of “the reasonable,” and by proceeding in accordance with this criterion he will act rightly; for happiness is attained by means of wisdom, and wisdom consists in right actions, and the right action is that which, when performed, possesses a reasonable justification. He, therefore, who attends to “the reasonable” will act rightly and be happy.[1]

Which, frankly, doesn’t seem very convincing.

2.       Skepticism and Logic

The objection to skepticism as a way of life, however, needn’t deter a truly heroic skeptic. They might simply accept that action is unjustified and live or die with the consequences of their inaction. It is an objection to the conclusion rather than to the argument leading to it. There is, however, a more direct objection to the claim that nothing can be known (if they’d ever made that claim outright) and that is that there are some things that we just do know. At the very end of the ancient world, as it was going down to barbarian invasion, Augustine objected to skepticism on the ground that some truths are certainly known to us, and that this includes some truths about the world.

I still know something about physics. For I am certain that (1) there is either one world or not. And (2) if there is not just one, the number of them is either finite or infinite… In the same way, I know that (3) our world is disposed as it is either by the nature of bodies or by some plan. And I know that (4) (a) either it always did exist and always will, or (b) it started to exist and will never stop, or (c) it did not start in time but will have an end, or (d) it started and will not last forever…These truths are disjunctions, and no one can confuse a likeness of something false with them.[2]

And what about mathematical or logical truths? Is it possible to doubt that 2 + 2 = 4? Or that if Socrates is a man and all men are mortal, then Socrates is mortal? Aren’t these completely indubitable? Why would we wish to suspend judgement upon such things? Can any rational argument (appealing to logic) be given to show that appealing to logic is unjustifiable? The very idea is absurd, he thinks.

And more than just such contentless truths (contentless because they would be true whether there was anything in the world or not,) Augustine believes that we can certainly know that there are some things actually in the world.

…who would doubt that he lives, remembers, understands, wills, thinks, knows, and judges? For even if he doubts, he lives; if he doubts, he remembers why he doubts; if he doubts, he understands that he doubts; if he doubts, he wishes to be certain; if he doubts, he thinks; if he doubts, he knows that he does not know; if he doubts, he judges that he ought not to consent rashly. Whoever then doubts about anything else ought never to doubt about all of these…[3]

For such reasons, and because of the triumph of Christianity in the intellectual world, skepticism vanished as a philosophical position of much interest. In the world that followed, Aristotle became the accepted authority on just about everything. Not an unquestioned authority, by any means, but the authority that set the terms of the philosophical debate for over a thousand years.

Descartes’s  Project

Skepticism of any sort didn’t really become important again until Descartes, disillusioned with the stagnant philosophy that had barely moved on from the time of Augustine, decided upon a root and branch rejection of that tradition. He would begin by taking a radical and, as he thought, a purely original approach to finding out what he really knew (p. 116.)

For several years now, I have been aware that I accepted many falsehoods as true in my youth, that what I had built on the foundations of those falsehoods was dubious, and therefore that, once in my life, I would need to tear down everything and begin anew from the foundations if I wanted to establish any firm and lasting knowledge.

Descartes had decided that he was not at all certain of all of his beliefs and wanted to remove everything that could be in error in order to leave himself with a remainder of beliefs that really were  true. Then, once this work of destruction was completed, if he was careful to add to his beliefs only those things that had to be true given the things that he already believed, he would be able to rebuild a structure of beliefs that could not be other than true.

This initial project of destruction, he decided would best be aimed at the most fundamental beliefs that he held. He saw that, if one viewed the collection of beliefs that one had as a structure in which some beliefs were held only on the grounds of other beliefs – and tis was a reasonable way of looking at things – then there was no need to investigate the certainty of each belief separately. It would be sufficient to investigate the beliefs that underlay all the other beliefs, and if they were found wanting and required to be discarded then all those beliefs that were supported by them would have to be discarded too.

The Method of Doubt

We can see from the attitude that Descartes took to his present beliefs that his view of knowledge was one that required absolute certainty in a justification before the belief that it justified could be accepted as knowledge. This means that for Descartes, he can know something only if it is not possible for that to be false.

A view such as this means that if there is some way, however absurd it might seem, in which things might be otherwise than he believed them to be then his belief could not count as knowledge. It is this that provided Descartes with his celebrated Method of Doubt. He tested each of the most fundamental beliefs (as he identified them) in his system of beliefs against the very possibility of their being incorrect, and if that possibility was found to exist he discarded those beliefs.

Sense Data

It’s widely held that sense data are at the root of many of our beliefs, which is quite reasonable: we believe that the sun is risen because we can see the sun. We believe that the door is open because we can see the door and it is open. The birds are singing because we can hear them. And so on. But Descartes points out what was well known to everyone, that the senses are not to be trusted completely (p. 116.)

... I have occasionally caught the senses deceiving me, and it would be prudent for me never completely to trust those who have cheated me even once.

Nevertheless, there seem to be some things at least that the senses do not mislead one about. Notably, information about oneself and one’s own situation.

But, while my senses may deceive me about what is small or far away, there may still be other things taken in by the senses which I cannot possibly doubt — such as that I am here, sitting before the fire, wearing a dressing gown, touching this paper.

If that is the case then here is a certain basis of knowledge that cannot be doubted. But is it the case?

Dreams

Descartes thinks not, because there’s always the possibility that he is dreaming (117).

This [certainty of knowledge about oneself] would be perfectly obvious — if I weren't a man accustomed to sleeping at night whose experiences while asleep are at least as far-fetched as those that madmen have while awake. How often a dream has convinced me that I was here, sitting before the fire, wearing my dressing gown, when, in fact, I was undressed and between the covers of my bed! ... When I think very carefully about this I see so plainly that there are no reliable signs by which I can distinguish sleeping from waking that I am stupefied ...

It’s pretty clear what’s going on here. Descartes sees that at the time of occurrence a dream is quite as convincingly real as the perceptions one has when one isn’t dreaming and there is no way to distinguish the two experiences. This means that whatever sensations he is having that he might think are giving him certain knowledge of the world could just as easily be merely dreamt sensations, and the knowledge he has from them would be merely beliefs about things that might be otherwise. Because certainty is thus denied to him, sensations cannot be the source of knowledge in that direct way.

On the other hand, Descartes does think that if the danger of dreaming is the only obstacle to knowledge from the senses, then there may still be indirect knowledge of the world that is indubitable, as well as certain other types of knowledge that are not affected by the truth or falsity of beliefs about the external world (p. 117.)

… [W]hile things like eyes, heads, and hands may be imaginary, it must be granted that some simpler and more universal things are real – the ‘real colours’ from which the true and the false images in our thoughts are formed. Among things of this sort seem to be general bodily nature and its extension, the shape of extended things, their quantity (that is, their magnitude and number), the place in which they exist, and the time through which they endure.

That is to say, we may imagine strange and false combinations of these things, but we can’t really doubt that that there are such things as extension or colour or shape, and these are certain facts about the world. He continues:

Perhaps we can correctly infer that, while physics, astronomy, medicine and other disciplines requiring the study of composites [i.e. material objects] are dubious, disciplines like arithmetic and geometry ... are somehow certain and indubitable. Whether we are awake or asleep, two plus three is always five, and the square never has more than four sides. It seems impossible even to suspect such obvious truths of falsity.

Demons

You’ll recall that this was one of the points that Augustine identified as being beyond the bounds of doubt: but now Descartes brings out his big guns in the form of a possible ground of doubt that is so fundamental that even the certain knowledge that was left to us after admitting the possibility of dream deceptions is no longer acceptable. He asks us to consider the situation that might arise if God used his omnipotence to deceive us about the things that we think we are certain of. In that case the mere fact that we thought that we had this certain knowledge could be no guarantee that we did actually have that knowledge. Even if one is certain that 2 + 2 = 4, this feeling of certainty is no guarantee that 2 + 2 is not actually a zebra. We are simply not justified in taking our intuitions of certainty as being

Some might claim that God’s goodness is such that he would never permit such a deception; but God evidently finds it acceptable to deceive from time to time, so there could be no difficulty in doing it more often. Or perhaps we could claim that God lacks omnipotence and is therefore incapable of such a thorough deception; but this (apart from being blasphemous) means that we are the product of an inferior being and so one can expect our truth-determining capabilities to be correspondingly reduced, and error to be even more common, and knowledge to be even less possible.

To these arguments, I have no reply; I am forced to admit that nothing that I used to believe is beyond legitimate doubt — not because I have been careless or playful, but because I have valid and well-considered grounds for doubt. (p. 117.)

If anything like this is the case, then even fewer of Descartes’s fundamental beliefs are going to qualify as knowledge. This means that when Descartes is trying to reconstruct his world of known things he will have to keep this possibility always in mind. Eventually, however, he decides not to slander God in this way, but instead to propose a perfectly equivalent possibility (p. 118)

I will suppose then, not that there is a supremely good God who is the source of all truth, but that there is an evil demon, supremely powerful and cunning, who works as hard as he can to deceive me. I will say that sky, air, earth, color, shape, sound, and other external things are just dreamed illusions which the demon uses to ensnare my judgement.

Cogito

At this point Descartes has apparently determined that, because there may be an evil demon he may be mistaken about every proposition. And, given his position that unless it is not possible to be wrong about something it is not correct to claim that you know that thing, this means that he does not know anything.

Descartes does not stop with this conclusion of course. He is not really a skeptic. He begins to question whether there really is nothing that he can be said to know and, once again following in the footsteps of Augustine (though he did not know it,) he comes up with the idea that he can at least be sure that he exists in order that he may doubt. This he expresses in the famous phrase cogito ergo sum – I think, therefore I am. Of course, he can’t initially express it like this, because to say ‘therefore’ is to appeal to rules of logic which have just previously been shown to be fallible (on the assumption that a demon is possible.) Instead, he has to say ‘I think; I am.’

It’s not until later in his work, and with a few rather slippery assumptions, that Descartes is able to reconstruct the bases of our knowledge sufficiently that we can appeal as we normally do to rules of reason. And from there he begins to rebuild the structure of his knowledge. It turns out that he thinks that we really do know a lot of things, but most philosophers think that the reconstructive part of the exercise is much less convincing, much less valuable, and much less interesting, than the destructive work. We won’t go into this project any further, but will now consider other problems with the view of knowledge that we derived from Aristotle.