top of page
  • Writer's pictureBurak Yetistiren

Does Nozick’s tracking theory represent an adequate response to the Gettier problem?

Updated: Mar 2


Gettier introduces a problem which is begotten by the traditional conditions for knowledge. He discovers the luck factor, which could be introduced in examples of knowledge acquisition (Gettier, p. 193). He gives two examples of such a case, and argues, that the knowledge cannot be based on the said factors of traditional knowledge. Nevertheless, settling for his view on knowledge, there has to be a proposition of knowledge, in which the proposed conditions for knowledge will eliminate the explained problem, discovered by Gettier. Subsequently, we are introduced with a modified version of the traditional conditions for knowledge introduced by Nozick. Nozick’s view on the conditions for knowledge intents to eliminate the luck factor embraced in the traditional conditions for knowledge, which result in erroneous instances of knowledge (one could state erroneous knowledge is not knowledge), which were exemplified in the previously explained study of Gettier. For our question, we want to see if indeed, the new conditions of knowledge offered by Nozick, solves the problem of erroneous instances of knowledge introduced by some luck factor.


Responses of the Philosophers

In the following, we will elaborate the responses of the philosophers about the problem of knowledge acquisition. The traditional conditions for knowledge is as follows (Gettier, p. 192):

(i). P is true. (ii). S believes that P. (iii). S is justified in believing that P.

Gettier argues that these conditions fail to suffice for knowledge (Gettier, p. 192). He argues that there can be some examples, that will embrace the three conditions stated, but still fail to be the bases for knowledge.

For example, he illustrates his argumentation with the Smith and the coins example. In this example Smith and Jones apply for the same job, and Smith has strong evidence that his friend Jones will get the job, moreover he also has strong evidence that Jones has ten coins in his pocket. By the strong evidence he has the previous proposition entails a new proposition, which offers that the man with ten coins in his pocket will get the job (Gettier, p. 192).

According to Gettier, this is where the problem starts. He introduces the luck factor here, he gives a possibility where unknown to Smith he himself will get the job, also again unknown to Smith, he himself has ten coins in his pocket (Gettier, p. 193). This is where Gettier argues that the traditional conditions of knowledge fail to suffice for reliable conditions for knowledge (Gettier, p. 192). In other words, Gettier argues that, even, the given three conditions are satisfied, an example could be given, in which the subject proceeds to end up with a wrong interpretation, in which her belief would be wrong, hence her belief would not constitute knowledge (Gettier, p. 193).

There are some responses to Gettier, evaluating the problem he introduced, and try to come up with possible solutions. Gilbert Harman is one of the respondents suggesting that Reasoning that consists primarily of incorrect conclusions, whether intermediate or final, cannot provide knowledge (Harman, p. 195). To solve this dilemma we observe the solution Nozick provides:

Nozick’s Truth Tracking theory reforms the original conditions for knowledge, intending to solve the previous problem introduced by Gettier. Nozick, wants to connect mere true belief to knowledge, by introducing a reliable connection between and the truth, by reforming the last condition for knowledge by introducing two new conditions (Nozick, s. 257):

(i). P is true. (ii). S believes that P. (iii). If P was wrong, S would not believe that P. (Sensitive requirement)

(iv). If P is true, S believes that P. (Adherence requirement)

These conditions try to work as eliminators of luck factors that we saw in the examples introduced by Gettier. To illustrate the condition (iii) offered by Nozick, let us consider the example of fake barns. For this example, we consider a field, which is full of barn facades (i.e., fake barns). The facades are indistinguishable from the normal barns, hence if the subject would take a look at the facade and try to identify what it is, the knowledge acquisition procedure would be compliant to the traditional conditions of knowledge, that the belief of the subject would be true and justified. In the proposition of Nozick, for this problem we see that if we consider the third condition, we will accept the following: If “that” was not a barn, the subject would not believe that it was a barn (Nozick, s. 256).



Having explained the underlying problem, we shall answer the question, if indeed Nozick’s response is sufficient in “solving” the problem Gettier introduced. I shall say that I am persuaded by the arguments stated by both philosophers. I agree with the argumentation, followingly the examples Gettier introduces in his study. The examples he gives indeed introduce some faults for a reliable instance of knowledge. At this point, I have also the view of the need for a singleton knowledge instance, in which with the given condition, one cannot end up with different conclusions, by some factors like luck. Hence, I am in the decision, that a redefinition of the traditional view of knowledge is required. I believe that Nozick was also in a similar conclusion, therefore he introduced his study, in which he gives a reform for the traditional theory on knowledge. There indeed, has to be some breakpoints, which act like a checkpoint for the disclosure of some aspects like, the luck factor. Hence, I am in favor to the (iii), and (iv) conditions introduced in the proposal for conditions for knowledge of Nozick.

New conditions Nozick offered, act in a way that any “open doors were closed”, meaning that any ambiguities left by the previous conditions, were eliminated by the new two conditions. The luck factor introduced by the ambiguities found out by Gettier were addressed with the updated version of the conditions of knowledge by Nozick. This situation confirms my view of choosing the approaches of both philosophers about the question. I agree with Gettier, in terms of discovering the existent problem with the traditional conditions of knowledge, and since we cannot reject knowledge, that knowledge exist, and humans seemingly construct their daily activities using knowledge, there should be any possible modification, not necessarily Nozick’s, in order to solve the existing problem.

On the other hand, I agree with Nozick, considering his theory’s power to solve some of the problems that became problematic after the points made by Gettier. In order to better assess the success or failure of the new proposition of Nozick, I believe that it is valuable to consider some of the examples that were left with ambiguity without the updated version of the conditions for knowledge.

The first instance of a Gettier problem I first want to introduce is the one I previously explained, which explained that Smith inferred an incorrect conclusion, because of his strong beliefs, which turned out to be wrong. Smith had strong evidence that Jones would get the job, and Jones had ten coins in his pocket. From this point it entailed that the person with ten coins would get the job. Accordingly, we considered a possible world, where Smith instead of Jones got the job, and Smith instead of Jones had ten coins in his pocket. If we apply the new proposition of Nozick to this problem, we would have, if Jones did not have ten coins in his pocket, Smith would not believe that Jones would get the job. This is the path we choose for the formation of the response of Nozick to the Gettier problems.

Let us consider another example offered by Gettier. In this one, Smith has now strong evidence for the proposition of Jones owning a Ford car. Also, Smith has another friend called Brown. Smith has no ideas where Brown is. Smith forms three proposition considering some possible situations about Jones and Brown:

  • Either Jones owns a Ford, or Brown is in Boston.

  • Either Jones owns a Ford, or Brown is in Barcelona.

  • Either Jones owns a Ford, or Brown is in Brest-Litovsk.

Gettier argues that (Gettier, p. 193) these propositions are entailed by the proposition, for which Smith had strong evidence, which is, Jones owning a Ford car. Followingly, Smith is justified in believing each of these new propositions. As a follow-up Gettier states that Jones might be driving a rental car, and again employs the luck factor, stating that by coincidence Brown can be present at any of these stated places. Now, the propositions that are entailed by the first proposition, that Smith had strong belief to, but by the luck factor.

Nozick’s response can be used again in terms of stating that, if Smith knew that Jones did not own a Ford car, he would not believe that he did. This time the answer is seemingly more trivial to the previous example as it is straightforward, but we cannot deny that the Nozick’s condition fail to solve this particular Gettier problem.


Possible Counter Argument

The trivial characteristic of Nozick’s response in effort to solve the Gettier’s questions might be unfulfilling for some audience, that one might think that the answers are pretty trivial, why one would some subject would consider the incorrectness of a proposition that they believe in the first place. In other words, why would a subject stop having her belief if it had been false. I agree, this is a fair point to consider, when we are considering the scope of the subject, but I believe that the problem Nozick is trying to solve is the structure, or the formation of the traditional knowledge acquisition, that in a way it causes some ambiguities to have a singleton way of knowledge definition for a given proposition. Hence, at this point what Nozick is trying to do is propose some additional constraints on justified true belief, to be able to defeat possible ambiguities like we had in the Gettier problems. The concerns that could be had for the belief of the subject appears to be out of the scope for our argument here.



Gettier, E. L. (2008). Is Justified True Belief Knowledge? In E. Sosa, J. Kim, J. Fantl, & M. McGrath, Epistemology an Anthology (pp. 192-193). Malden: Blackwell Publishing.

Harman, G. (2008). Thought, Selection. In E. Sosa, J. Kim, J. Fantl, & M. McGrath, Epistemology an Anthology (pp. 194-206). Malden: Blackwell Publishing.

Nozick, R. (2008). Knowledge and Skepticism. In E. Sosa, J. Kim, J. Fantl, & M. McGrath, Epistemology an Anthology (pp. 255-279). Malden: Blackwell Publishing.

6 views0 comments


Post: Blog2_Post
bottom of page