Baltimore, 1927–2021
Edmund Gettier
In 1963 a little-known lecturer published three pages asking whether knowledge is justified true belief. The answer was no, and analytic philosophers have been trying to repair the definition ever since.
The view
Since Plato, the usual analysis of knowledge had three parts. You know that something is so when it is true, you believe it, and you are justified in believing it. Truth alone is not enough, since you might never have heard of it. True belief is not enough, since a lucky guess is not knowledge. Justification was supposed to be what rules out luck.
Edmund Gettier, then teaching at Wayne State University in Detroit, showed that it does not. His paper "Is Justified True Belief Knowledge?" describes two cases in which a person has all three and plainly does not know. He published almost nothing else in a long career, most of it at the University of Massachusetts Amherst.
The argument
Both cases have the same structure. Someone has good evidence for a false belief, and deduces from it something that happens to be true for an unrelated reason.
Smith has strong evidence that Jones will get the job and that Jones has ten coins in his pocket. He infers that the man who will get the job has ten coins in his pocket. In fact Smith gets the job, and Smith, though he does not know it, has ten coins in his own pocket. After Gettier (1963), case I
Smith's belief is true. He believes it. He reasoned validly from good evidence, so he is justified. Yet almost everyone judges that he does not know it. He is right by accident.
A simpler case of the same kind comes from Bertrand Russell, years before Gettier. You glance at a clock that has always been reliable. It says ten to three, and it is ten to three. But the clock stopped twelve hours ago.
The paper set off an industry. Perhaps knowledge needs a fourth condition: that the belief not rest on any false step, or that it be caused by the fact it is about, or that it come from a reliable process, or that no further truth would undermine the justification. Each proposal met a new counterexample. In 1994 Linda Zagzebski argued that this is no accident. As long as justification falls short of guaranteeing truth, a case can always be built in which bad luck cuts the belief off from the truth and good luck restores it.
Where he meets Deutsch
Gettier found the weak point in the traditional account, and it is the one Popper and Deutsch attack: justification. His cases show that being justified and being right can come apart completely, with nothing in the believer's position to tell them which is happening. Zagzebski's argument sharpens the lesson. Justification that does not guarantee truth cannot do the job, and justification that does guarantee truth does not exist.
Where they part
The literature after Gettier tried to mend the definition. Deutsch, following Popper, thinks all three clauses are mistaken, and for reasons that have nothing to do with unlucky clocks.
Knowledge is not justified, because nothing is. It is not a kind of belief, because it does not need a believer: the knowledge in a textbook or a genome is there whether or not anyone holds it in mind. And it need not be true. Newton's theory of gravity is false, strictly speaking, and it is among the greatest pieces of knowledge ever created. It solves problems, it is hard to vary, and its successor explains why it works as well as it does.
Popper also distrusted the whole approach. Asking "what is knowledge?" and testing proposed definitions against intuitions about cases treats a word as if it had an essence to be captured. Deutsch asks different questions: how knowledge is created, how it improves, and what it does in the world. On those, the cases of Smith and the clock are silent.
| Gettier | Deutsch | |
|---|---|---|
| Where does knowledge come from? | Not addressed. His question is what knowledge is, not where it comes from. | Conjecture: guesses, improved by criticism and experiment. |
| What does justification do? | It is necessary for knowledge, and his cases show it is not sufficient. | Nothing. No theory is ever justified, and none needs to be. |
| Can knowledge grow without limit? | Not addressed. | Yes. Problems are inevitable, and problems are soluble. |
| What is knowledge, physically? | A state of a believer: a belief that is true and suitably justified. | Information that, once embodied, tends to keep itself in existence. |
Further reading
- Edmund L. Gettier, "Is Justified True Belief Knowledge?", Analysis 23 (1963), pp. 121–123.
- Linda Zagzebski, "The Inescapability of Gettier Problems", Philosophical Quarterly 44 (1994).
- Karl Popper, "Epistemology Without a Knowing Subject", chapter 3 of Objective Knowledge (1972).
- The Analysis of Knowledge, Stanford Encyclopedia of Philosophy.