If one subject, more than any other, has come to dominate philosophical inquiry it is epistemology.  Epistemology deals with questions concerning the nature of knowledge, what we can know, and how we come to know it.  Modern philosophy’s focus on epistemology might be considered the legacy of René Descartes; however, as we have already seen, epistemological concerns were being addressed as early as Plato.

As previously mentioned, one of the concerns of philosophy involves coming up with definitions.  These, more often than not, are provided in terms of necessary and sufficient conditions.  For example, a necessary condition for being a sister is that you are female.  However, this alone is not sufficient for sisterhood; it is also necessary that you be a sibling.  As such there are two necessary conditions for being a sister and these are jointly sufficient.  Therefore, we can define a sister as a female sibling.

There are generally thought to be three individually necessary and jointly sufficient conditions for knowledge. Firstly, in order to know something you must believe it (ignore the fact that sometimes people will say such things as ‘I know Boris Johnson is the Mayor of London but I don’t believe it!’); this is known as the belief condition.  Obviously, it’s not sufficient for having knowledge that one has a belief.  I, for example, may believe that as you read this you are slowly transforming into a giant beetle.  But, could I be said to know this?  If not, why not?

Of course, I don’t know that you currently undergoing metamorphosis.  Why not?  Well, simply because it’s not true (I hope).  The second condition, then, for a particular claim to be known is that it be true; the truth condition.  Having knowledge, then, involves having some sort of true belief, but is this enough?  Consider the following claim:  ‘I know that you are a Capricorn’.  Now, roughly 8.33% of you will be Capricorns, and let’s suppose that you’re one of them (even if you aren’t).  In that case I have a belief that you’re a Capricorn and my belief is true.  But do I know that you are a Capricorn?

Most people would, I believe, say not.  Why not?  Well, perhaps the reason that my true belief would not count as knowledge has something to do with the fact that this claim is only accidentally true, if it is true at all.  That is to say, if my belief is true, it is due merely to the fact that you happen to be one of the 8.33% of people who are Capricorn and not due to my having any particular reason or justification for holding this belief.  If, however, my true belief that you are a Capricorn was justified in some way (for example, by my knowing that you were born on December 25th) then most people, I think, would agree that I know you are a Capricorn.  This final necessary condition is known as the justification condition.

As such, we may define knowledge as justified true belief; these three conditions being individually necessary and jointly sufficient for knowledge.  That is to say, if you have knowledge you must have a justified true belief and if you have a justified true belief then you have knowledge.  Or not!

In 1963 Edmund Gettier published a very short paper entitled ‘Is Knowledge Justified True Belief?’ in which he presented counterexamples to the tripartite definition of knowledge. Gettier’s paper had a huge impact on philosophy and a number of Gettier-like counterexamples have been constructed to show that a justified true belief is not always sufficient for knowledge.

Consider the following:

Jennifer believes that Brad is having an affair with Angelina and, in fact, it’s true that Brad and Angelina are having an affair.  Jennifer’s suspicions were aroused by the fact that the paparazzi photographed the pair together in a restaurant on a night when Brad had said that he was busy filming.  As such, Jennifer’s belief is justified.  However, what Jennifer doesn’t know is that Angelina was actually out with a Brad impersonator that night (Brad was actually filming and Angelina was so besotted that she hired a look-alike).  So Jennifer has a justified true belief that Brad and Angelina are having an affair, but does she have knowledge?  Does Jennifer know about Brangelina?

Most people here are inclined to say ‘no’. Perhaps they would say, though, that this is not a counterexample to knowledge as justified true belief as Jennifer’s belief was not really justified in the right kind of way.  Either way, what these examples highlight is the major difficulty with justifying one’s beliefs; what, if anything, counts as an adequate justification?  When can we be said to know something – for certain?  These are exactly the type of questions which struck the aforementioned Descartes some 300 years earlier.

In his Meditations on First Philosophy (1641), René Descartes (1596-1650) set out to establish a firm foundation for the sciences.  In attempting to do so he succeeded, via two of the most famous sceptical arguments in philosophy, only in confirming how little we actually know for certain.

In the Medieval world [and indeed still to this day in some dark corners] the best way of acquiring knowledge was via Revelation or Authority.  But Descartes was writing at a time of scientific revolution when many doctrines which had hitherto been accepted as certain were being overturned and, as such, he was struck by the instability and unreliability of scientific ‘knowledge’.

In his First Meditation, Descartes aims to sweep away all of his previously held opinions and start afresh. His ‘method of doubt’ entails that if anything can be doubted, however slightly, then we are to treat it as if it is manifestly false and reject it outright.  It is not, however, necessary that we subject each and every one of our opinions to this hyperbolic (exaggerated) doubt, as this would be a Sisyphean task.  Rather, Descartes aims to test the ‘foundations’ of what we claim to know – ‘as the removal from below of the foundation necessarily involves the downfall of the whole edifice’.

Descartes claims that our knowledge comes either from the senses or from reason, and in order to subject these foundations to his method of doubt he unleashes two powerful arguments: the dream and demon hypotheses.  Firstly, let us consider the Dream Hypothesis.  What Descartes hopes to establish via this argument is that, given any particular experience, we can never know that that experience is not a dream.  As such, we have reason to doubt the senses and therefore should treat all beliefs that we receive from them as manifestly false.  [Note that all Descartes requires here is the universal possibility of illusion (that for any experience we have, that experience may be a dream) not the possibility of universal illusion (that we may be dreaming all the time).]

Descartes believes, however, that some beliefs are immune from the dream hypothesis – namely, those that we gain not from experience but from reason alone – ‘for whether I am awake or dreaming, it remains true that two and three make five, and that a square has but four sides’.  In order to subject this realm to the method of doubt Descartes conjures the second of his sceptical arguments: the demon hypothesis.

I will suppose … that some malignant demon, who is at once exceedingly potent and deceitful, has employed all his artifice to deceive me; I will suppose that the sky, the air, the earth, colors, figures, sounds, and all external things, are nothing better than the illusions of dreams.

In case you find the introduction of a malevolent genie too much to bear, Hilary Putnam (Reason, Truth, and History, 1981) updates the argument by asking you to consider the possibility you are a brain in a vat and that your experiences are just electronic signals being sent to your disembodied brain by an evil scientist.  [‘Utter nonsense!’ you exclaim, but that’s exactly what a brain in a vat would say.]  This thought experiment also forms the plot of the Hollywood blockbuster The Matrix.

By the end of the First Meditation Descartes has done a terrific demolition job.  By applying his method of exaggerated doubt to the foundations of knowledge the entire edifice has indeed crumbled.  At the outset of Meditation II, he states that:

The Meditation of yesterday has filled my mind with so many doubts, that it is no longer in my power to forget them… I suppose, accordingly, that all the things which I see are false (fictitious); I believe that none of those objects which my fallacious memory represents ever existed; I suppose that I possess no senses; I believe that body, figure, extension, motion, and place are merely fictions of my mind. What is there, then, that can be esteemed true? Perhaps this only, that there is absolutely nothing certain.

In order to progress, Descartes must establish one thing that is certain and indubitable.  This eventually leads him to the cogito: ‘this proposition I am, I exist, is necessarily true each time it is expressed by me, or conceived in my mind’.  This one truth, believes Descartes, is immune from all doubt for even a malignant demon cannot deceive me into thinking that I exist if I do not; if I am being deceived, then I exist.

It is worth questioning why Descartes employs both the Dream and Demon hypotheses when it seems clear that the latter covers everything included in the former; that is to say, the Demon argument attacks both the senses (experience) and reason, while the Dream argument is directed only at the unreliability of the senses.  The answer to this is, I think, that Descartes believes (wrongly) that he can defeat the Demon argument.  [Descartes’ rejection of the Demon argument is actually risible in light of his own demanding method.]  As such, he still requires the Dream argument to cast doubt on the reliability of the senses.

Descartes is a Rationalist; he believes that some knowledge can be had via reason alone, unaided by the senses.  This form of knowledge is known as a priori knowledge.  In opposition to this, Empiricists believe that knowledge must come via experience derived from the senses: a posteriori knowledge.  This division between Rationalism and Empiricism can be traced back as early as Plato and Aristotle.

One of the major benefits of a priori knowledge is that it delivers certainty; it is indubitable.  For example, the proposition that ‘2+2=4’ or that ‘either Jim is in Edinburgh or he’s not’ just have to be true (they are necessarily true).  Some a priori claims, such as the aforementioned ‘All sisters are female siblings’, are also called analytic.  This is due to the fact that they contain no more information than is already contained in the meaning of the terms (‘sister’ just means ‘female sibling’).  In establishing the truth of all of these propositions there is no need to appeal to the senses; the truth of these claims is not established by your counting your fingers (hopefully), going to check whether I’m here or not, or interrogating sisters.

The problem, an empiricist might claim, is that while such a priori/ analytic propositions may be necessarily true, they are actually rather trivial and worthless.  What we really want to know, they argue, is whether ‘Jim is in Edinburgh’ or if ‘Lucy is a sister’, and they only way we can establish the truth (or falsity) of these propositions is by looking for Jim or asking Lucy.  If such propositions are true, they are only contingently so; that is, they happen to be true but could have been false (Jim need not be in Edinburgh and Lucy could have been an only child).  Note also, that these propositions are not analytic but rather synthetic; they bring together different concepts and, as such, offer meaningful information.  The problem, of course, is that, as Descartes and Gettier discovered, they are far from certain.  For example, it may be that it was Jim’s identical twin brother you saw in Edinburgh while Jim is in fact sunning himself on a beach somewhere, and Lucy may be a compulsive liar.

So, there we have it. In simple terms, knowledge can be defined as justified true belief, however, it is difficult to establish what exactly counts as an adequate form of justification. Rationalists maintain that useful knowledge can be gained through the use of reason alone. Such a priori/ analytic claims are self justifying and, therefore, necessarily true. Empiricists, however, claim that the price of this certainty is triviality and that the only route to useful ‘knowledge’ about the world is experience.

If you enjoyed this post, you may also find my discussion of Abre los ojos interesting.

5 replies on “Epistemology”

The last para summarises the issues very well. However, as a newcomer to Philosophy I find it surprising that the source of “received wisdom” on the subject of “knowledge” are the great minds of 300 or more years ago! This is not to criticise their pioneering thoughts, more to ask whether modern Philosphers have anything to add!

There are two issues that I’d like to raise for discussion
1. Why are the “rationalist” and “empiricist” approaches presented as mutually exclusive alternatives? Surely we need both. There is ample evidence of this mutually supportive approach in the advances made in our scientific knowledge. Thus our ”empirical” observations of the world need to be “rationalised” in the form of a hypothesis which then leads to further observation and rationalisation.
2. Traditional philosophical thinking appears to be rooted in “absolute” ways of thinking- true or false. Clearly, there are areas where that is appropriate, as in logic and the rationalist way of thinking, because that is the way the premises are set-up. But in considering the wider subject of knowledge, is there not a place for a probabilistic way of thinking? If the recent history of human activity has taught us anything, surely it is that there is often more than one way of explaining what we see in our world, be it in science, politics and art. Plato’s cave illustrates this, but again it assumes that there is one “true” physical world. I don.t understand it but when it comes to more complex areas of “knowledge” such as origin of the Universe (Big bang v steady state) or particle physics, it seems we can only think of knowledge in terms of what best explains the observation, or in lay terms, what is most probably true.


Some really interesting points here. It’s correct to say that there has been a great deal of contemporary work done on the subject of epistemology that is not covered in this short introduction. It’s also fair to say, though, that these contemporary epistemologists are still, in a way, attempting to respond to the sceptical arguments put forward by Descartes.

I think you are also correct in pointing out that rationalism and empiricism (as ways of acquiring knowledge) are not mutually exclusive – the hypethetico-deductive model in science is evidence of this. However, particular philosophers do tend to belong to one school or another, dependant on the particular set of beliefs they hold concerning such things as the Mind, etc. We will be looking at some of these beliefs throughout the course.

Lastly, there is indeed a need for probabilistic thinking (actually this is the more generally adopted in terms of science and everyday experience). Inductive arguments (as opposed to the deductive ones we looked at) progress from particular experiences to general claims based on probability. The problem here (and there’s always a problem) is that we can never determine with certainty the truth of general claims (all swans are white) by recourse to a finite list of particular instances (swan 1 is white…swan 2,367,875 is white). Of course, we do this all the time – and most people are content with claims of verisimilitude – but we always have to be aware of black swans!


I don’t know if this is a forum for discussion, but…

In reponse to John I feel the need to assert that ‘good’ ideas will remain for centuries, millennia, which is why the names of philosophers will outlast the names of scientists. Good ideas are not pioneered. They exist independently of the ‘pioneer’. Scepticism will exist infinitely longer than the knowledge of string theories as, paradoxically, string theories will be disproven. No amount of evidence equates to knowledge as science itself evidences. Scepticism can respond to science, but science cannot respond to scepticism with anything other than empty refutation. Berkeley kicking a rock does not prove his existence. It merely calls into question the existence of the rock.

Scepticism cannot be disproven, yet the existence of such an idea does ‘prove’ that science is not a means of attaining knowledge. We cannot redefine knowledge to suit science. Knowledge is truth and science cannot provide us with knowledge as truth. Truth, as I mentioned in class, is ‘self’ evident. It requires no belief, nor justification. It is true in itself. It’s own truth IS it’s justification. It’s own truth negates belief. Redundancy merely serves to complicate the complicated.


You make a couple of good points here, but there must be some compromise between certainty and utility. Philosophers such as Popper would agree with your view that science cannot provide truth, however, the hypotheses they advance are incredibly useful in helping us to understand phenomena (Occam’s razor: that theory is best which is simplest and works).

I see your point on truth as being self-evident. Be sure to distinguish though between something’s being true (which perhaps entails no belief nor requires justification) and knowledge, which involves our (correctly) justified belief that something is true.


Compromise is difficult when dealing with ‘knowledge’. All principles of utility are fallacious and I cannot progress from an erroneous principle to a conclusion. As useful as it may be to others, I fail to see the use of utility when dealing with abstractions such as truth and knowledge.

I do not see any difference between truth and knowledge. Truth is knowledge. Knowledge is truth. Certainly from a sceptical viewpoint, and as scepticism is undoubtedly the theory that “is simplest and works” (in principle) I stand by it.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s