I can say I know something, but how would I know I really knew it? A.J. Ayer attempted to answer this very question by identifying three necessary and sufficient conditions for someone (S) knowing a proposition (P)- firstly, that P is true; secondly, that S is sure that P is true; and thirdly, that S has the right to be sure that P is true. Simply put, this necessitates the justification of an objectively true, subjective belief for there to be knowledge- for something to be knowledge, it would have to be justified true belief.

The subjective component reinforces the objective component, and the objective component reinforces the subjective component. Ayer believed we could not have the one without the other. Plato, in his Theaetetus, makes a clear distinction between knowledge and unjustified belief. He gives the example of a jury who cannot convict a person they believe to be guilty as they have not enough evidence to convict them on. (He was also a rich aristocrat who loathed the democratic system of justice but that’s for somewhere else). Then we have ‘knowledge’ which implies correctness. We would never doubt knowledge, or claim a false proposition as knowledge. A belief is not always true and we would never say a conviction was knowledge. The third component, the justification, is what the rest of it depends upon. But ‘having a right to be sure’ is what many of the major arguments against Ayer’s definition, have been.

The major arguments against Ayer’s definition have been put forward by Edmund Gettier against the justification component of the definition but minor counterexamples have also been put forward to the second condition, the belief in what is true. In 1966 Colin Radford said it is actually possible to know without believing. The best way I have of understanding this is by using an example from Dostoyevsky’s beautiful novel Crime and Punishment. Raskolnikov is going to see Svridligaïlov- he takes a turning he would ordinarily have never thought of taking. Usually he would turn right to the direction of::: okay, i haven't read the book for a while. but today he turns left and, as he believes, by chance comes upon a dirty tavern where he sees Svridligaïlov in an upstairs window. Raskolnikov did not believe he knew to take this turning. But subconsciously he did believe. He did not remember Svridligaïlov telling him three times of the tavern where he would usually drink and pay a minor to sing for him, but the knowledge was engraved in his subconscious. So Raskolnikov knew, but he did not believe he knew. There may be better examples than this of knowing without believing, but it’s the one I find makes me understand it better. Luckily for Ayer, D. M. Armstrong put a counterexample against Radford’s by telling him he’s merely giving us something we already knew. A person can only know something if he on some level believes it. So what Radford is saying is no argument, it only reinforces Ayer’s definition. Even if someone does not believe a proposition is true consciously, he will have to believe it is true for him to know it, according to Ayer.

However, the main argument against Ayer is not whether conditions in the definition are necessary, but whether they are sufficient. Could it be that one fulfils all the conditions for knowledge, and yet still not possess knowledge? Edmund Gettier published an article in 1963 in which he argued that this is possible. His example involved two people, Smith and Jones, applying for the same job. The chairman of the company assures Smith that Jones will get the job (Jones is better qualified for the job, the chairman has known Jones since birth, &c.). Smith also knows a fact about Jones- that he has ten coins in his pocket. So from this Smith concludes that the man to get the job has ten coins in his pocket. It turns out that Smith gets the job. He checks his pocket and finds, much to his astonishment, that he has ten coins in his pocket. So did Smith know he would get the job? All three conditions for knowledge are met- the proposition that the man with ten coins in his pocket will get the job was certainly true. Smith did believe this proposition was true. Smith also had good reason (the right) to believe this was true. But nobody in their right minds, and truthfully, would say he knew this. It was merely a coincidence, that he had the justified true belief that the man who would get the job had ten coins in his pocket. The basis of his justification was in Jones’ pocket and not his own.

In response to Gettier's article, several measures have been taken to try and fix this definition. One approach has been to try to distinguish between reliability and unreliability in ways of collecting beliefs. It was the justification part of the definition that was the problem, the way people based their beliefs on other things and were justified in them. The solution to this problem has been given by Robert Nozick. He called it the ‘conditional theory of knowledge’ and with it he attempted to remove the problem of someone’s justification that is based on a reliable method acquiring a belief that is true, but where the method plays no part in its success. Basically it is a simple tweaking of the traditional analysis of knowledge. Nozick accepts the first two conditions i)P is true and ii)S is sure that P is true. To these he added two new conditions, in place of the ‘S has a right to be sure’ which he rejected, and attempted to expand. These two new conditions are iii)if P were not true then S would not believe P and iv)if, in changed circumstances, P were still true, S would still believe P.

I can thus conclude that more is needed in a definition of knowledge than justified true belief. This is my first argument for Ayer being a bit of a pretentious charlatan.