Explain a Gettier-style counterexample to the justified true belief analysis of knowledge, and explain and assess one response to Gettier's counterexamples.
Submitted for PH144 (Mind and Reality) Take Home Exam
According to the ‘justified true belief’ (JTB) analysis of knowledge, S knows p iff the following hold: S believes p; p is true; and S is justified in believing p. This is a definition for knowledge inasmuch as each of the criteria are individually necessary (S does not know p if any are not met), and jointly sufficient (S necessarily knows p if all are met).
However, Gettier (1963:121) shows that these criteria aren’t sufficient, given two ostensibly reasonable assumptions: justification is consistent with falsehood (it is possible to have a ‘justified false belief’); and justified belief is closed under entailment – “if S is justified in believing that p, and p entails r, and S deduces r from p and accepts r as a result of this deduction, then S is justified in believing r”.
Hence, we have a rubric for producing “Gettier-style counterexamples” to the JTB analysis: S holds a justified false belief, and deduces a justified true belief from it. The resultant belief is true only by luck, which is typically perceived as being inconsistent with genuine knowledge, undermining the sufficiency of the JTB criteria. For example, suppose that one’s Philosophy exam consists of two essays; one shows the first to the lecturer, who clandestinely assures one that it will get a First. One thus forms the justified belief that one of the essays will get a First. But, the grader gives this essay a lower grade, and unexpectedly gives the other essay a First; accordingly, one’s justified belief that one of the essays will get a First turns out to be true; but one did not know this – its truth was coincidental.
A possible response to Gettier is to add a condition to the JTB analysis: in particular, S knows p iff S has a justified true belief p, and S’ justification for believing p rests on no false lemmas. On this account, the reason one doesn’t have genuine knowledge in this example is that the justification for one’s true belief depended on a false lemma (that the essay showed to the lecturer would get a First). This analysis of knowledge evades all Gettier-style counterexamples, since by their nature, these generate a justified true belief inferred from a falsehood.
But this definition is vulnerable to counterexamples of a different sort. Suppose, for instance, that unbeknownst to me, my watch is broken, stuck showing the time as 9 o’clock. The next morning, I am accurately informed that it is 9am; I look at my watch, which seems to corroborate this, and hence form the justified true belief that it is 9am. Per the ‘no false lemmas’ definition, I do not know that it is 9am, since part of the justification for my belief – that my watch accurately shows the time – is false.
However, if we accept this, many beliefs which we hold based on a variety of sources may be invalidated if one is unreliable. Consider a detective, who on the basis of testimony from a dozen witnesses, forms the true belief that a crime took place; if eleven were truly there but one was lying, it seems counterintuitive to maintain that the detective does not know that the crime happened (Nagel 2014:49). Hence, the ‘no false lemmas’ response is unsuccessful, as it precludes everyday knowledge whenever it is incidentally supported by a false belief.
Result
Mark: 78% (Lower Mid First)
Feedback:
You argue that adding the no false lemmas condition to the JTB analysis doesn't provide an adequate analysis of knowledge.
This is an excellent answer. Your explanation of the JTB analysis and of your own Gettier-style counterexample is exemplary. You write beautifully. On the basis of reading this answer, I can see that you have outstanding potential as a philosophy student.
Despite these manifest strengths, there is a way in which this answer could have been better. It actually isn't clear that, in the stopped clock case, your justified true belief rests on a false lemma. It isn't clear because it isn't clear that your belief that it is 9 o'clock is inferred from the premise that the clock is working. (Normally, when one forms beliefs about the time on the basis of clocks, one doesn't consciously go through a process of reasoning like "The clock says it's 9 o'clock; the clock is working; therefore, it's 9 o'clock.)
This may seem like nitpicking but there are actually some interesting issues here about the formulation of the no false lemmas condition, and how we are to understand when a belief is arrived at by "inference" from a false premise.