Russell, Descartes and the Sceptic.

As Russell asks: “Is there any knowledge in the world which is so certain that no rational man can doubt it?”

A definition of knowledge, first given by Plato in his ‘Meno’, states that knowledge has three criteria. He claims that one only knows something if one believes in it, if it is true and if the belief that it is true is justified. All three criteria are necessary for knowledge. This definition has been challenged and altered over the years but the essence remains the same and so is sufficient for this essay. This definition seems sensible enough but the question of ‘what is justification?’ arises. The sceptic claims that anything which cannot be fully justified, i.e. can be doubted, cannot pass as knowledge because there is a chance of error.

The sceptical epistemological argument claims that all our sources of knowledge can be doubted and because, therefore, we can never be justified in our belief, we can never have knowledge, certain or otherwise. To support this argument, it is important to understand what the main sources of knowledge are. The majority of our supposed knowledge comes from our senses, our faculty of reason and our memory. The sceptic argues that all these sources are unreliable and therefore not sufficient justifiers for knowledge. Our senses are easily fooled by illusions; a parched wanderer in an arid desert may think he sees an oasis up ahead but when he gets there he finds he has been deceived by his senses and that there is no oasis at all. Our memory can also fail us when we think we have remembered the year of the Battle of Waterloo as 1816 when in fact we are just wrong. The argument for doubting rational thought is perhaps not quite as simple. Rene Descartes gives a support for the argument in his ‘First Meditation’.

Descartes, speaking through the voice of a first-person meditator, uses the example of mathematics as knowledge gained from the faculty of reason; “For whether I am awake or sleeping, two and three added together always makes five, and a square never has more than four sides.” This seems indubitable but Descartes then goes on to create a situation in which it is possible for the consistent set of rational ‘truths’ we hold, to be doubted. Once he has created some doubt, however small, says the sceptic, we can longer use it as knowledge.

Descartes says “it is possible that God has wished that I should be deceived every time I add two and three or count the sides of a square.” He then goes on to say that it would be against God’s benevolent nature to deceive the meditator all the time and so he postulates the existence of a malicious demon that has the same abilities to deceive as God but is, of course, malicious. It must be noted that Descartes is not attempting to (and does not need to) prove the existence of God or the malicious demon. He merely has to show that it is conceivable that the meditator’s mind is being controlled by a malicious demon who is able to make him believe 2+3=5 is true every time he thinks it, but in fact be wrong. If it is conceivable, the meditator can no longer be certain his faculty of reason is indubitable and if it can be doubted, we cannot use it for knowledge.

So far the sceptic philosopher has argued that it is not possible to have knowledge of anything, let alone certain knowledge, on the basis that if we can doubt our sources of knowledge and therefore make errors in the construction of our beliefs, we can never be justified in our beliefs. Both Descartes and Bertrand Russell give counter-examples to the sceptical argument. Descartes attempts to refute the claim that we can never have knowledge by finding a belief that can never be doubted. If it can’t be doubted, then the sceptic must agree that it is certain knowledge. Referring to the original question, Russell attempts to show by method of induction that we can have certain knowledge if we appeal to the rational man as opposed to the strong sceptic.

In the ‘First Meditation’ Descartes travels the sceptics path and doubts everything that can be doubted; “I will suppose that the heavens, the air, the earth, colours, shapes, sounds and all external things that we see, are only illusions and deceptions…I will consider myself as having no hands, eyes, flesh, blood or senses, but as believing wrongly that I have all these things.” The meditator also doubts his rational thought using the malicious demon hypothesis. Now in the ‘Second Meditation’, Descartes looks for some knowledge that cannot be doubted. Having doubted that the external world and other minds exist, the meditator attempts to doubt his own mind’s existence. He says, “Was I not, therefore, also persuaded that I did not exist? No indeed; I existed without doubt, by the fact that I was persuaded, or indeed by the mere fact that I thought at all.”

This seems to be the first glimpse of certain knowledge that can now be analysed. Descartes is arguing that the thought, ‘I don’t exist’ is self-refuting. In order to convince himself he doesn’t exist, the meditator must exist to do so. This is clearly a pragmatic paradox and the only conclusion is that if the meditator is thinking, he exists, or as Descartes put it; ‘Cogito ergo sum - I think, therefore I am’. It has not been necessary to rely on the sense, reason or memory to arrive at the statement ‘I am, I exist’ and so is not prone to attack by the sceptic. Even if there is no external world and there does exist a malicious demon that is deceiving the meditator in his faculty of reason, the meditator can still know he exists because in order to be deceived, he must necessarily exist. Descartes seems to have hit upon some certain knowledge but what exactly is the nature of this knowledge?

What is the nature of the ‘I’ in the statement ‘I am, I exist’? It refers specifically to whoever or whatever is having the thought at the time it is thought. It is not necessarily the man Rene Descartes or any embodied human being at all. The doubt of the external world still holds and so it is important not to assume that the statement ‘I think, therefore I am’ establishes anything more than a thing that thinks at the time it is thinking. However, even though the ‘Cogito’ does not afford us useful knowledge perhaps, it does offer us certain knowledge that no rational man can doubt, because to doubt it, is to exist. Bertrand Russell in his ‘Problems of Philosophy’ offers a perhaps more useful method of discerning knowledge.

Russell, as has been said earlier, offers a different approach to gaining knowledge. He accepts Descartes ‘Cogito’ argument as knowledge and also adds to it by saying the we have knowledge of our ‘sense-data’. ‘Sense-data’ is the word that Russell uses to describe the experiences we have while we are sensing things. Russell argues that if we experience something from our ‘sense-data’, even if the data is erroneous, we cannot deny that we have had the experience. For example, if one experiences the smell of a fragrant perfume, even if the perfume doesn’t exist and the person is not in fact smelling anything but is being programmed to do so, the experience is still valid and it is possible to say one has knowledge of the experience or ‘sense-data’. Using Plato’s definition, the person having the sense experience, believes he is having the experience, it is true he is having the experience and he is justified in believing that he is having the experience because it is not possible to deny an experience.

If we are to proceed in our quest for knowledge, says Russell, we must find a way of applying our established knowledge in order to draw inferences from them. He says, “It must be known to us that the existence of some one sort of thing, A, is a sign of the existence of some other sort of thing, B…as for example, thunder is a sign of the earlier existence of lightning. If this were not known to us, we could never extend our knowledge beyond the sphere of our [limited] private experience.” Note that Russell is merely saying what he thinks ought to be the case, not necessarily what is the case. He has stated his aim and now he must support it.

Russell says that if some thing A is often associated or followed by another thing B, it is reasonable to suggest that the probability of B following A increases the more actual occurrences there are of B following A. Before he can be criticised for not offering a systematic proof to his knowledge, Russell says, “It is to be observed that all such expectations are only probable; thus we have not to seek for a proof that they must be fulfilled, but only for some reason in favour of the view that they are likely to be fulfilled”. This is an important point when distinguishing between certain knowledge that the sceptic doubts and certain knowledge that the rational man doubts. The sceptic, who claims that if there is any element of doubt in a proposition, it cannot be viewed as knowledge, will argue that Russell’s method of induction, i.e. the more something happens the more likely it is always to happen, is open to an enormous amount of doubt. If an occurrence of something is only probable, there is no certainty that it will happen again in the same way. If there is no absolute certainty, says the sceptic, it cannot be knowledge.
However, the question asks if there is any certain knowledge that no rational man could doubt and Russell believes knowledge gained via the principle of induction fulfils these criteria.

The general principle of induction states that:

a) The greater the number of cases in which a thing of the sort A has been found associated with a thing of the sort B, the more probable it is (if no cases of failure of association are known) that A is always associated with B;
b) b) Under the same circumstances, a sufficient number of cases of the association of A with B will make it nearly certain that A is always associate with B, and will make this general law approach certainty without limit.

Russell offers a few caveats when applying the principle of induction. He says “we have therefore to distinguish the fact that past uniformities cause expectations as to the future, from the question whether there is any reasonable ground for giving weight to such expectations after the question of their validity has been raised.” By this he means that just because we expect something to occur because it has happened many times in the past this does not give the necessary evidence to support the claim that it will happen again. We must have a reason to believe it will occur again not just an expectance. He gives the example of a chicken which, having been fed by its master every day of its life, gets a nasty shock when its slaughtered for its meat. In a rather tongue-in-cheek fashion Russell notes that “more refined views as to the uniformity of nature would have been useful to the chicken”.

So expecting that B will follow A is not enough, we must have a good enough reason. Russell says we must use the uniformities of nature, such as laws of motion and gravity to give us good reason. He also acknowledges that the original question can be applied to the uniformities of nature, namely; have we any reason, assuming that they have always held in the past, to suppose that they will hold in the future? He answers that the probability of such uniformities increases the more times that they occur without contradiction. It is these uniformities of nature, which science has played a great role in producing, which enable us to have knowledge about the world. We can have knowledge that the sun will rise tomorrow, not only because it has done so in the past for as long as can be remembered, but because the laws by which it rises have also stayed uniform. He also says that if two events are found together often enough, the probability will amount almost to certainty.

In answer to his own question, Russell provides a way for the rational man to have certain knowledge about sense experiences and the external world. For the more sceptical rationalist, Descartes opens the doors to some certain knowledge, even if it isn’t very useful.


  1. Descartes, René. Discourse on Method and The Meditations. London; Penguin, 1968
  2. Russell, Bertrand. The Problems of Philosophy. Oxford; Oxford University Press, 2001.

So according to your interpretation of Russel’s sense data formulation (talking about the perfume), if we were to see what we believed to be an extra terrestrial being under the influence of LSD…it is valid …please explain

Yes that is what I’m saying. Note the difference between “knowledge of the existence of an extra terrestrial being” and “knowledge of an experience of an extra terrestrial being.” You would be quite right in saying, and Russell agrees with you, that our senses can be deceived (in your case with LSD), however, we cannot deny the actual experience. The knowledge we have, is that of the experience, as opposed to knowledge of the actual thing that’s happening.

  • ben

interesting because i am now just writing a paper on a similar topic but taking a different approach. the sceptics (i assume you were referring to the ancient school in post-aristotliean times) caused the death of certainty because they clearly pointed to the discrepincey between what is in the self and what is in the world. aristotle really motivated this change in distinguishing between actuality and potientiality —> which completely refuted the presocratics and plato. still, even russel’s ‘solution’ gravitates back to the ‘priori-universals’ claim that the socratic school held. overall, knowledge will always remain dependent on what constitutes the self and the world. descartes tried to divorce that, and could only do it to a certain extent. would have liked to see you take this paper farther, the topic deserves it, really.