#139: Gettier Counter-Example Lessons [Warning: Boring Philosophy]

April 30, 2018

[These are reflections really only relevant to people of the Epistemology persuasion — my apologies to all others.]

I: Providing Context

From about 1900 on philosophical conceptual analysis increasingly took on a very specific form. Instead of leaving it to our intuitions to determine the “meaning” of a concept, philosophers left any reference to “meaning” behind and became interested instead in what had to be true in order for a particular concept to be properly applied.

A classic instance of such an analysis is that of the justified-true-belief account of the concept of knowledge.

According to the JTB:

S knows that p if and only if

p is true,

S believes that p, and

S is justified in believing that p.

The new method had the distinct advantage of avoiding all sorts of vexatious questions associated with talking about “meanings.” It also had the advantage of specifying exactly how a criticism of an account is to take place, viz. by way of a “counter-example.”

A counter-example, as the expression indicates, is a case, real or fictitious, in which an account fails either by being too demanding (it has too many conditions or too constraining conditions), on the one hand, or by being not demanding enough (requiring an additional condition(s)).

The JTB was considered a rock-solid analysis, a model of the method, until a very short paper was published by Edmund Gettier in 1963. Gettier argued that the JTB is deficient in that the three conditions while necessary were not sufficient. He argued this by providing a counter-example to the classic account, an example in which the three conditions were satisfied, but putatively S did not know.

His argument is instructive for our discussion of the method of philosophical analysis. He provides, in fact, two counter-examples, though I’ll only cite his “case 2” as both are based on the same strategy.

Here’s case 2:

“Let us suppose that Smith has strong evidence for the following proposition:

Jones owns a Ford.

Smith’s evidence might be that Jones has at all times in the past within Smith’s memory owned a car, and always a Ford, and that Jones has just offered Smith a ride while driving a Ford. Let us imagine, now, that Smith has another friend, Brown, of whose whereabouts he is totally ignorant. Smith selects three place names quite at random and constructs the following three propositions:

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.

Each of these propositions is entailed by (f). Imagine that Smith realizes the entailment of each of these propositions he has constructed by (f), and proceeds to accept (g), (h), and (i) on the basis of (f). Smith has correctly inferred (g), (h), and (i) from a proposition for which be has strong evidence. Smith is therefore completely justified in believing each of these three propositions, Smith, of course, has no idea where Brown is.

But imagine now that two further conditions hold. First Jones does not own a Ford, but is at present driving a rented car. And secondly, by the sheerest coincidence, and entirely unknown to Smith, the place mentioned in proposition (h) happens really to be the place where Brown is. If these two conditions hold, then Smith does not know that (h) is true, even though (i) (h) is true, (ii) Smith does believe that (h) is true, and (iii) Smith is justified in believing that (h) is true.”

The strategy is this:

It is to construct examples in which a proposition (h in the above, the one supposedly known) is validly derived from a false, though justified, proposition f, but because h can be true under more than one condition, it is in fact true, though accidentally so.

I chose Case 2 because the “known” proposition (h) is a disjunction (an “or” proposition) and it is obvious with disjunctions that they can be true under more than one circumstance.

The Gettier examples rely on the following:

That a proposition can be well justified even though false;

That justification is transferred through entailment.

That there are propositions which can be true under various different circumstances.

II: What Should We Conclude from these Cases?

Now, the two Gettier cases have been taken to mean that the JTB account of knowledge is defective because it fails to exclude the cases. I want to suggest that that is not the significance of the cases; rather, I think it points to a problem with this method of analysis.

In both of his cases, Gettier simply assumes that the conclusion must be that his protagonist does not know the proposition in question and that we will automatically agree with that. And it is critically important that we do agree, for without that, the counter-example fails. However, I don’t want to suggest that he does know the proposition; rather I want us to ask ourselves what our own intuitions are about the case he presents. Note that it is a highly artificial case based on a logical trick. Given that, my own intuition is that I want to respond that I have no idea what this example implies with respect to knowledge.

The clear uses of words allow for confident responses, but things get murky as we move away from the center. And this is not a case of a vagueness built into a word, like it is with the word “heap.” In “heap,” we have a vague band between “yes” and “no,” but “no” is ultimately reached. In the Gettier case of knowledge, we are not moving towards uncertainty by tiny increments, not knowing exactly where “knowledge” definitely ends; no, we are given a bizarre, contrived example for which we are not prepared by natural language or intuition, so we really cannot say “no.”

With the Gettier case, we like an ornithologist confronted by a winged, flying pig. Ok, we say to our ornithologist, is it a bird? He would, should, say I have no idea.

Does my belief that a square has four sides count as knowledge? Yes. Does my belief that the sun is 93 million miles from the earth? Yes. But does Smith’s belief that Jones owns a Ford or Brown is in Barcelona? I have no idea. The word “knowledge” as I learned to use it was never applied to such a case.

Thus, the only way in which the Gettier cases can be made to count as counter-examples to the JTB is if we make the assumption that analyses of the JTB kind must be proof against examples outside the core of normal natural language uses. But why would we make such an assumption?

I think it is because a number of very influential philosophers had a mostly unspoken model of language in which our ordinary usage was no more than a clue to an underlying, concealed perfect language which did indeed consist of universally applicable concepts. The ancient version of this fairy tale was Plato’s theory of Ideas, while the modern one is that of “logically perfect language.” According to this notion, a conceptual analysis must be proof against any counter-example, no matter how foreign to our natural language mastery. I suggest that this is a bar too high, set perhaps with one eye on the physical sciences where observed phenomena are used as signs of underlying natural laws.

A more plausible model, perhaps, would be that of the law, where the prosecution’s claim must be proven beyond a reasonable doubt. Not beyond any doubt, no matter how contrived. So, the prosecutors in the O.J. Simpson murder trial did not have to prove to the jury that aliens from another planet did not kill Nicole Brown and Ron Goldman in order to make their case; this was a doubt too far, they only had to dismiss reasonable doubts (i.e. doubts for which there was at least some evidence).

In a similar way, if conceptual analysis is to be continued, an analysis can fairly be only expected to be proof against any intuitively recognizable counter-example, but not more than that.