As I understand it a Gettier case happens when there is a true belief that is justified but only by luck.

Common scenarios include

  • looking at a broken watch and it just happens to be the time on that watch. You believe it's that time, it's true that it is that time and you're justified in that belief because the watch says so but it's not real knowledge because the clock is broken.
  • A farmer sees a large fluffy dog in the distance thinking it's a sheep and based on his observation he believes there is a sheep in the field. Because he thinks it's a sheep he's justified in his belief. Out of the farmers view there happens to be a sheep in the field making his belief true.

In those two cases and the few other cases I've heard it seems to me that the justification part of the equation isn't good enough for it to be considered real justification. What is the criteria for justification?

3 Answers 3


(As a preliminary, it's not "luck" that the belief is justified. The perceptual evidence is the justification in both cases - it's just luck that the belief described is in fact true.)

I'd say responding to the cases by looking at what we mean by justification is a pretty standard response to the Gettier cases, and projects like Goldman's Reliabilism or Dretske's Relevant Alternatives restriction can be read as being more strict about what it might take to count as sufficient justification for knowledge claims. But this makes justification the central puzzling question; is justification in a knowledge attribution strictly stronger than justification as "good reason to believe", or stronger still than "reason to hold with a certain credence value"? Do we have a standard notion of justification simpliciter to appeal to?

Gettier's cases make a strong argument to the effect that the "True Justified Belief" explanation of knowledge is at best a partial explanation in the absence of a more fine-grained account of (1) what justification is, and (2) the particular role of justification in knowledge.

  • What defines Justification? Can prejudice be considered justification? Feb 25, 2014 at 16:19
  • 1
    Well that's the question! Post-Gettier, giving a particular account of justification would seem important to amending the tripartite theory of knowledge. On a Reliabilist account, some prejudgements might well prove reliable, even if they do tend to fail occasionally or have no obvious causal connection to the truth of the beliefs formed. And the Relevant Alternatives account would seem to rule out many prejudices on the grounds that the possibility that the prejudice fails is salient more often than not.
    – Paul Ross
    Feb 25, 2014 at 19:24
  • I suppose I should clarify - the epistemology literature tends to read Gettier's challenges as objections to the tripartite theory, suggesting that knowledge might be reformulated into "JTB+C". Reading the extra conditions as specific constraints on justification can be a useful way to go, though it opens up a problem of whether there is something particular about the kind of justification involved in knowledge over justification in other contexts.
    – Paul Ross
    Feb 25, 2014 at 20:00

No, the idea of gettier cases is that knowledge can't be justified true belief, because you can have a justified true belief but lack knowledge.

The reason that the beliefs really are justified, is that the procedure the knower uses would generally count as providing a good justification in ordinary circumstances. Usually the fact that your watch says 9am is good justification for the belief that it is 9 am, etc.

The problem is that justification does not guarantee truth. I can have a justified belief that simply fails to be true by bad luck. But since justification is a different concept than truth, I can have a true belief that is justified, although the fact that my belief is true is not grounded in the actual justification I have for that belief. That's Gettier's insight.

  • In college my philosophy prof said you can only know if your assertion is correct after the fact. This is where I stopped taking philosophy classes. It is no more useful than ordinary reasoning.
    – Scott Rowe
    Nov 3, 2022 at 21:53

I agree with you: the Gettier problems fail to justify the claim that knowledge is not justified true belief because none of the examples are examples of justified beliefs. I think that justification for knowledge consists of logic and evidence - logic being the means of correct/valid reasoning, and evidence being facts that justify a statement.
Now most beliefs are not knowledge despite some being held with certainty, and people hold such beliefs anyway because it's useful to operate on assumptions, but we should recognize that they are in fact assumptions/faith. In that sense one is "justified" to hold such working beliefs/assumptions, but that is not knowledge justification. So I do not think you can have a justified true belief that is not knowledge, and if justification does not guarantee truth then it is not really knowledge justification.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .