What Is The Gettier Problem?

The Tripartite Theory of Knowledge, a theory which goes back as far as Plato, states that in order for one to possess knowledge, one must have justified true belief. In other words, if you carry a belief, that belief must be true and you must be able to justify it. The Tripartite Theory exists to provide a response to the question of what exactly knowledge is. Whilst the theory might have been widely accepted in the past, it was in 1963 that an epistemological debate on the topic ensued. Edmund Gettier’s three-page paper “Is Justified True Belief Knowledge?” shows two conditions where the criteria for justified true belief appear to be met but mostly due to luck rather than knowledge. Whilst these counter-examples have led to a number of attempted solutions, my opinion is that the Gettier problem cannot be resolved. Alvin Goldman attempted to solve the Gettier problem with a theory of causal connection which states that justified true belief must be achieved through a causal connection to the truth. In this essay, I will highlight the ways in which Goldman’s proposed solution falls short and so fails to solve the Gettier problem and thus demonstrate one way in which the Gettier problem cannot be resolved.

The Gettier problem fundamentally demonstrates the problem of luck in the Tripartite Theory and thus shows how justified true belief doesn’t necessarily mean knowledge. I will present this through the following Gettier-style example: Susie walks past the same clock everyday on her way to class at 2pm. Today, as she walks past, the clock indeed says 2pm and she has no reason to believe otherwise since the clock is well maintained and has always been right in the past. Susie has justified true belief. However, unbeknownst to Susie, the clock stopped 12 hours earlier at 2am. So whilst Susie believes the clock reads 2pm, it actually reads 2am. Thus, the fact that the clock stopped at the time it did is purely down to luck. So Susie’s belief fails to be knowledge. If knowledge were to be gained via a causal connection, the problem of luck would essentially not exist as the agent would have acquired knowledge through a reliable, cognitive process. The purpose of the causal theory is to reinforce the Tripartite Theory but with the addition of a clause that states that a causal chain between beliefs and truths is required for knowledge. Goldman argues that perception is one such mechanism that can allow a causal connection and thus justify one’s true beliefs. He states in his 1967 paper “A Causal Theory of Knowing”; “Perhaps the simplest case of a causal chain connecting some fact p with someone’s belief of p is that of perception” (Goldman 358). Goldman uses the example of the vase to reinforce his argument about perception which is put forth as follows: S sees a vase but that vase is actually being hidden by a laser photograph of a vase. Because of the photograph, S believes that there is a vase in front of him. Goldman argues that in actual fact, S does not really see the vase because of the lack of causal connection that exists between S’s belief and the real vase. So ultimately, what Goldman is saying is that knowledge acquired by sight requires some sort of causal connection for it to truly be knowledge. And so the Gettier problem is essentially resolved, according to Goldman, with the addition of the causal connection clause.

Goldman continues his paper by discussing knowledge based on memory. Memory can be considered a causal process because a current belief could be caused and therefore traced back to an earlier cause. Knowing a fact at one time and then knowing it again at another is not necessarily knowledge based on memory as a causal connection would have to be included as an essential component in memory. Like perception, the causal connection of memory is used by Goldman to solve the Gettier problem through the addition of an extra element in the Tripartite Theory. Goldman goes on further to discuss knowledge based on inference through the lava case. S finds lava on the countryside and infers that there was an eruption. But in actual fact, someone has placed lava there in order to make it look like there had been a real eruption, unaware of the actual volcano. Goldman states that S doesn’t really know that there was an eruption because it is not causally connected to any beliefs about there being an eruption. So, inferential knowledge requires a connection through some sort of causal chain or as Goldman puts it: “when someone bases his belief of one proposition on his belief of a set of other propositions, then his belief of the latter propositions can be considered a cause of his belief of the former proposition.” (Goldman 362).

Whilst Goldman’s “A Causal Theory of Knowing” is a stimulating essay on the subject of knowledge, his argument or rather so-called solution to the Gettier problem in my opinion is ultimately flawed. One such criticism lies in the argument that our senses could so easily deceive us and therefore can a causal connection based on perception really be used to prove the acquirement of knowledge? The answer is surely no and this can be seen in the barn facade example, in which an agent drives through a town filled with fake barns. Standing in front of one of these fake barns, the agent can say that he has justified true belief through perception. Even though there is a causal connection in this example, the agent does not really have knowledge. So false conclusions can be drawn from a process which, according to Goldman would ordinarily produce true beliefs highlighting the fact that Goldman’s theory is flawed. So this does not provide a solution to the Gettier problem.

Another problem with Goldman’s theory lies in the fact that not everything in the world can have a causal connection. In particular, knowledge that is a priori such as 2 + 2 = 4 seems to be a fact about numbers but it doesn’t have a causal effect. It is a truth that you know based on no causal connection just like many series of numbers are. So it is possible to say that the problem with Goldman’s theory here is that it focuses on a posteriori knowledge but seems to ignore a priori knowledge and thus is too restrictive. It can therefore not be a solution to the Gettier problem because it doesn’t take all types of knowledge into account, which is vital in epistemology. So again, the Gettier problem remains unresolved.

The problem of incorrect reconstruction is one that further weakens Goldman’s theory. There are times when one may have a true belief but the causal chain which allowed them to have that knowledge is incorrect. For example, an agent discovers his best friend’s body and believes that he is dead due to the fact that he is missing body parts. The agent assumes that the friend died by decapitation but actually, he was poisoned and decapitated following his death. This doesn’t count as knowledge due to the false nature of the causal chain. Kenneth Collier in his 1973 essay “Contra the Causal Theory of Knowing” finds a counter-example to Goldman’s theory which indicates no causal chain. Collier uses the example of an agent on a hallucinogenic drug which highlights the fact that any beliefs the subject has which turn out to be true, cannot be explained causally due to the hallucinogenic state that they are in. In the paper, Collier has given Smith the hallucinogenic drug and he hallucinates that his wife is having an affair. In actual fact, his wife is having an affair but can this count as knowledge? Surely this is just an example of incorrect reconstruction due to the hallucinations or an example of no causal chain at all. Collier argues that “the difficulty here seems to be that it is possible for Smith’s belief that p to be caused by the fact that p in ways that are irrelevant to his knowledge. But this means that the causal connection cannot be a sufficient condition for knowledge.” (Collier 351)

Ultimately, Goldman’s causal theory attempts to resolve the Gettier problem by adding an extra clause which states that true belief has to be a result of a causal chain. He uses examples of perception, memory and inference to highlight the ways in which causality can solve the Gettier problem. But his theory is flawed overall due to the unreliability of the senses, a lack of a focus on a priori knowledge and the problem of incorrect reconstruction. I chose to focus on Goldman’s theory but it is just one example out of a large number of attempts to resolve the Gettier problem. Other proposed solutions include eliminating conditions such as luck, defeat and false evidence as put forth by philosophers like Unger, Lehrer and Lycan. These proposed solutions, like the causal connection theory are all flawed and thus do not provide an adequate response to the Gettier problem. Therefore, I do not believe that the Gettier problem can be resolved due to my agreement that justified true belief does not necessarily imply knowledge.