Gettier, E. (1963). "Is Justified True Belief Knowledge?" Analysis, 23 (6), 121-123.
Gettier challenges the traditional-analysis-of-knowledge (as well as Ayer-version-of-the-analysis-of-knowledge) and is an influential paper, leading to philosophers undertaking the task of re-analyzing the notion of propositional knowledge.
“I shall argue that [the traditional analysis] is false in that the conditions states therein do not constitute a sufficient condition for the truth of the proposition that S knows that P.”
Gettier is specifically challenging the notion that the three conditions of the analysis jointly constitute as a sufficient condition for knowledge. One way of doing this is by suggesting that they are cases where all three conditions are fulfilled, but they still don’t count as knowledge.
Gettier Cases / Gettier Counter-Examples
Simplified version of a Gettier case
Suppose that Smith has strong evidence for the proposition: (A) Jones owns a Ford.
His evidence is that Smith’s friend, Brown, who is generally reliable and honest, told Smit that Jones has always owned a Ford. So, Smith then infers that “Jones owns a Ford” on the strength of Brown’s testimony.
On this particular occasion, Brown has mixed Jones up with someone else. And in fact, Jones has, in the past, owned a Ford. However, by coincidence, Jones bought his first Ford just yesterday. Therefore, despite Brown’s inaccurate testimony, it is now true that Jones owns a Ford.
This example satisfies all three conditions of the traditional analysis.
- condition 1 = Smith believes that Jones owns a Ford
- condition 2 = it is true that Jones owns a Ford
- condition 3 = Smith’s belief is justified as it is based on strong evidence (honest and reliable testimony) The problem is that it does not seem right to say that Smith knows that Jones owns a Ford. Smith’s belief that Jones owns a Ford is based on a false belief deriving from Brown’s mistaken testimony, which happens to be true by coincidence. This seems like a classic example of “right answer, wrong equation.
The evidence brought Smith to the right belief, but it brought him to that belief in the wrong way because the evidence was obviously mistaken, and even false, since it wasn’t true that Jones always had own a Ford. So despite Smith’s belief being justified due to the reliable testimony, Smith still doesn’t have knowledge because the belief was accidently true.
simplified version of Gettier’s case II
Suppose that Smith has strong evidence, and is justified in believing, the proposition that (A) Jones owns a Ford.
In this case, the evidence is that in Smith’s own memory, Jones has always owned a Ford, and Jones just offered him a ride while driving a Ford. Smith has another friend, Brown, whose whereabouts “he is totally ignorant” From the justification that Jones owns a Ford, Smith makes the deductive inference that: (B) either Jones owns a Ford, or Brown is in Barcelona.
It turns out the Jones does not own a Ford. The Ford he was driving when he offered Smith a lift was a rental. But, it just so happens that Brown is, in fact, in Barcelona.
First point of order is to explain how Smith makes his deductive inference. Propositional logic states that the truth of a disjunctive claim (an “or” proposition) follows from the truth of either of the disjuncts. i.e. f you know , you can validly infer OR , for any .
From proposition (A), Smith can logically infer (B). If (A) is true, then (B) must also be true, since the truth of a disjunction requires at least one of its components be true.
Because Smith’s original belief (A) was justified, and the inference from (A) to (B) is logically valid, Smith is also justified in believing (B). The problem arises because (A) turns out to be false while (B) turns out to be true, not because of the first disjunct but because of the second that Smith has no knowledge of.
This example satisfies all three conditions:
- condition 1: Smith believes in proposition (B)
- condition 2: proposition (B) is true, because of the second disjunct = Brown is in Barcelona
- condition 3: Smith’s belief in proposition (B) is justified because he has strong evidence for the first disjunct (A) from which he makes a justified inference by deductive logic.
However, Smith’s evidence for (A) was misleading. His belief in (B) rests on a false premise, and its truth is merely accidental. Therefore, Smith cannot be said to know that “either Jones owns a Ford or Brown is in Barcelona”.
Analyzing the Gettier cases
In both examples, the three conditions of the traditional analysis of knowledge are fulfilled: S believes that P, P is true, and according to Gettier, S believes the P is justified. The Gettier cases posits that even with all the conditions fulfilled, there can still be cases where it is not knowledge.
In each of the cases, Smith had evidence for his belief, but the evidence was false, and yet, his belief happens to be true despite the false evidence. It could be argued that evidence, even strong evidence, doesn’t count as genuine justification unless the evidence itself is true. If this is the case, then Gettier would be wrong to suggest that the counter-examples fulfil the conditions of the traditional analysis, since the justification condition is no longer satisfied.
So, traditional-analysis-of-knowledge can be salvaged by requiring that the justification for our beliefs must involve only true claims. There is a problem that arises with this kind of thinking.
Note that Smith has good reasons to trust the evidence in both cases. In the first example, he is trusting a generally reliable source, and in the second example, he is trusting his own memory, perceptions, and deductive inference. In both examples, he is not in a position to assess the truth of the evidence, for he has good reason to trust the evidence. This shows that to demand as a condition for justification that the evidence be true would seem too demanding as a conception of justification; it would make justification into something we would rarely be able to access.