Back to the on-screen lesson ·
explain why a lucky true belief can fall short of knowledge
Paper packet. Every task here also exists on screen, where it is checked automatically; answers written on paper are not assessed by Nydus. When you are back at a device, enter your answers there.
You will explain why a lucky true belief can fall short of knowledge, explaining the inference and its limits in a supplied case.
Belief, truth, and justification are distinct. A proposed analysis must say whether its conditions are necessary, sufficient, or both.
| Term | What it means |
|---|---|
| JTB | The proposed account of knowledge as justified true belief. |
| Sufficient condition | A condition whose satisfaction guarantees the target property. |
| Epistemic luck | Accident affecting the connection between cognitive success and the believer's grounds. |
| Counterexample | A case meeting a proposed condition while lacking what it supposedly guarantees. |
An analysis of knowledge tries to explain what makes a belief count as knowledge. A familiar proposal says that a subject knows p exactly when p is true, the subject believes p, and the subject is justified in believing p. These three conditions are often abbreviated JTB: justified true belief. The proposal is stronger than saying that the conditions are often present when people know. It says that meeting them is sufficient, as well as necessary, for knowledge.
To challenge that sufficiency claim, construct a case in which all three conditions appear to hold but knowledge remains absent. A case lacking truth would not do this job, because the proposal already excludes false belief. A case lacking belief would likewise miss the target. Nor would a bare lucky guess, with no reason at all, show that justified true belief is insufficient. The point is to grant enough support to make the justification condition plausible and then identify a different defect.
Gettier's challenge concerns the relationship between the subject's route to belief and what makes the belief true. A reason can lead someone toward error, while an independent coincidence happens to make the final conclusion true. The resulting belief may be supported and true yet seem too lucky to count as knowledge. This is epistemic luck: luck affecting whether the cognitive success is connected to the believer's grounds in the right way. It is different from the harmless luck of happening to meet a knowledgeable teacher.
The examples in this lesson are original teaching cases, not quotations from Gettier's paper. They isolate the structure of the challenge using explicitly stated facts. Philosophers disagree about which account best explains the defect, and some dispute particular intuitions about cases. You are asked to reconstruct the challenge and its target, not to treat a disputed philosophical analysis as a measurement result. The assessment can be precise even when the philosophical debate remains open.
Another way: A true conclusion for the wrong reason
Consider a worker who sees a convincing demonstration that machine A is operating. In fact, the demonstration is a recording and machine A is off. The worker concludes, 'At least one machine in the workshop is operating.' Unknown to the worker, machine B is operating in a back room. The conclusion is true because of B, while the worker reached it through apparent evidence about A. The source of truth and the source of confidence come apart.
Reconstruct this carefully. The first belief is that A operates. The final belief is that at least one workshop machine operates. The inference from the first proposition to the second is valid: if A operates, then some machine operates. The problem is not an invalid existential inference. The problem is that the premise is false while an unrelated fact about B makes the conclusion true. If the appearance about A supplies reasonable support under the case's conditions, the final belief can appear justified, true, and believed without being knowledge.
The justification claim needs attention. If the worker already knows the display is a recording, then accepting it as a live demonstration would be poorly supported. That altered case no longer gives the same challenge. A well-constructed example therefore tells us that the display is ordinarily dependable and that the worker has no available warning. These details are not excuses added after the fact. They specify why the case is directed at sufficiency rather than merely illustrating careless reasoning.
Now remove B while leaving the worker's evidence unchanged. The final belief becomes false. The worker would make the same inference and experience the same confidence. This nearby variation makes the accidental connection vivid. However, do not confuse the variation with the original case: the original conclusion is true. A counterfactual check helps diagnose the original success; it does not rewrite the facts so that a true belief becomes a false one.
Another way: Stopped clocks and environmental luck
A stopped clock gives another accessible illustration. A traveler consults a clock that has always worked well and has no visible sign of failure. It stopped twelve hours earlier, but the traveler happens to look at precisely the time it displays. The resulting belief about the time is true. The normal history of the clock can make consulting it reasonable. Yet the matching time is accidental: the display is no longer responding to the passing hours.
This example draws attention to the method's current connection with the fact. The traveler is not simply lucky to have encountered a clock; everyday knowledge depends on many fortunate opportunities. The troubling luck is that an insensitive display happens to coincide with reality. Had the traveler looked a few minutes earlier or later, the same method would have produced a false answer. Again, the point is not that every possibility of error destroys knowledge. It is that this particular successful result comes from a currently broken route.
An environmental case is different. Imagine a district filled with convincing imitation weather stations, only one of which actually measures the weather. A visitor unknowingly consults the sole functioning station, sees a correct reading, and forms a true belief through its genuine measurement. The actual causal route is better than in the stopped-clock case. Nevertheless, some philosophers judge the surrounding risk of easily choosing a fake to threaten knowledge. Others analyze the case differently. This shows why one simple repair may not settle every form of epistemic luck.
Keep the two patterns distinct. In the stopped-clock case, the actual source is defective. In the environmental case, the actual source works but the surrounding alternatives are misleading. A proposed condition excluding false premises may address some cases of reasoning from error, yet it does not automatically settle an environmental case that uses a true observation. When comparing repairs, ask exactly which defect each condition excludes. The fact that two examples both involve luck does not mean their structures are identical.
Another way: Repairs and their limits
One repair proposes adding a no-false-premises condition. The machine case then fails because the belief about A is false. That is a clear response to that particular construction. But knowledge can sometimes survive incidental false beliefs that do no work in supporting a conclusion, and environmental luck may arise without a false supporting premise. A useful analysis must say which premises matter and why. Simply requiring a person to have no false beliefs anywhere would make knowledge depend on irrelevant mistakes.
Another approach emphasizes a reliable belief-forming process. An operating calibrated clock usually produces accurate time beliefs; a stopped clock does not. This directs attention away from a single successful output toward a method's tendency. It raises further questions about how to describe the relevant process and environment. 'Reading this functioning station' and 'choosing a station at random in a district of fakes' describe the same episode at different levels, with different reliability implications. The next reliability lesson examines that problem more closely.
Safety approaches ask, roughly, whether a belief formed in this way could easily have been false. Sensitivity approaches ask a different counterfactual question: if p were false, would the subject still believe p? These formulations need refinement in philosophical work, and they are not interchangeable definitions. Here they serve as examples of repair strategies that examine the relationship between a method, nearby alternatives, and truth. When reconstructing them, preserve the direction of the counterfactual rather than treating every conditional as the same test.
Do not infer from the failure of one analysis that knowledge does not exist. A failed definition shows that the proposed conditions need revision, not that the thing being analyzed is impossible. If a proposed definition of a triangle accidentally includes curved shapes, rejecting the definition does not eliminate triangles. Similarly, Gettier-style cases challenge a sufficiency claim about knowledge. They can motivate better accounts without requiring global scepticism or the dismissal of ordinary inquiry.
A good written analysis contains four parts: the proposition believed, the subject's grounds, the fact that makes the proposition true, and the coincidence separating those last two. Finish by naming the target: the claim that JTB is sufficient. This final sentence is essential because otherwise an interesting story may never become an argument. The philosophical work lies in showing how the story supplies a counterexample to a stated general account.
A librarian checks a shelf label that normally records the location of the only large-print atlas. The label points to cabinet A. Without opening it, she reports that the library has a large-print atlas available. Cabinet A actually contains an ordinary-print atlas because a delivery was shelved incorrectly. Independently, a donor has just left a large-print atlas in cabinet B, and the librarian does not know about that donation.
The proposition to analyze is existential: at least one large-print atlas is available in the library. It is true because the donated copy is in B. The librarian believes it on the basis of the label for A. Under the case's assumptions, the label normally gives reasonable support and there was no available warning of the shelving error. Nevertheless, the correctness of this particular report depends on a fact outside her reasoning. The error about A and the unnoticed arrival in B cancel at the level of the final yes-or-no inventory claim.
To see the luck, vary only the donation. If the donor had arrived tomorrow, the librarian would have made the same report from the same label, but the report would have been false. This does not prove that all catalogs are untrustworthy. It identifies why this specific true result does not demonstrate that the relevant item was actually located by the reporting process.
The operational response is to inspect the item and correct the record for its actual location. If the librarian subsequently opens B, verifies the large print, and updates the catalog, she acquires a different evidential route. A good audit records the original error even though the final availability statement happened to be true. Counting only correct final outputs would hide the faulty process, leaving future readers exposed to the same mistake without another donation to rescue the result.
A bare guess does not challenge JTB, since the justification condition is missing. A false belief does not challenge its sufficiency either, since truth is missing. To reconstruct a Gettier-style case, preserve the true final conclusion and the apparently adequate grounds, then explain the accidental connection. The result challenges an analysis; it does not by itself establish that no one ever knows anything.
Specify the believer's proposition.
p: it is eight o'clock
The traveler forms this time belief after consulting a familiar clock.
Record the supplied support.
The clock has an excellent past record and no visible warning
These details make the belief reasonably supported in the stipulated case.
State the actual time.
It is eight o'clock
The truth condition is satisfied despite the hidden fault.
Identify the hidden defect.
The clock stopped twelve hours earlier
Its present display no longer tracks the passing time.
Locate the epistemic luck.
The frozen display happens to match the actual time
The success is accidental relative to the current method.
Name the first belief.
Machine A operates
A deceptive live-looking display supplies the worker's apparent evidence.
Name the inferred proposition.
At least one machine operates
The existential inference would be valid if the first belief were true.
State the hidden facts.
A is off and unseen B is on
These facts separate the false supporting premise from the true conclusion.
Identify the truth-maker in the case.
B's operation makes the existential conclusion true
B was not part of the worker's grounds.
State the challenge to the analysis.
JTB appears insufficient
The conclusion is true and supported, yet its truth is rescued by coincidence.
State the proposed additional condition.
The belief uses no false supporting premise
This condition targets reasoning that succeeds despite an erroneous premise.
Apply it to the machine case.
The false premise about A violates the condition
The added condition excludes that particular example.
Introduce the environmental case.
The consulted station works among many indistinguishable fakes
The actual observation and measurement can both be accurate.
Check for a stated false supporting premise.
None is stipulated
Do not invent one merely to make the repair succeed.
Identify the remaining concern.
The visitor could easily have consulted a fake
This concern concerns the environment rather than an actual false premise.
Give the bounded assessment.
The repair addresses one pattern but leaves another disputed
Defeating one counterexample does not establish a complete analysis of knowledge.
A reliable-looking label says crate A contains a compass.
The agent concludes that some crate contains a compass
The conclusion existentially generalizes the apparent evidence about A.
A is empty, but unseen crate B contains a compass.
The conclusion is true
The object in B satisfies the existential claim.
Identify what rescues the conclusion.
An agent reasonably believes token A is silver from misleading evidence and infers some token is silver. A is copper; unseen B is silver. Answer Y or N in both fields. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Response symbols: Y = yes; N = no.
| Response | |
|---|---|
| Truth | |
| Luck |
A stopped clock is read at exactly its displayed time by someone with no warning of failure. Complete the analysis. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Truth symbols: T = true; F = false. Connection symbols: L = luck; G = guarantee.
Evaluate the resulting time belief.
The belief is truth
The actual time matches the displayed time in this stipulated case.
Identify the unusual connection.
The matching result depends on cause
The display does not currently track passing time.
Name the target of the challenge.
The sufficiency of JTB
The case preserves the proposed conditions while questioning knowledge.
The same kind of misleading evidence concerns A. The agent infers some token is silver, but every token is copper. Answer Y or N in both fields. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Response symbols: Y = yes; N = no.
Is the conclusion true? truth. Is its truth rescued by the stated independent coincidence? luck.
An agent directly verifies that A is silver and concludes some token is silver. A really is silver, and no misleading route or independent rescue is stipulated. Answer Y or N in both fields. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Response symbols: Y = yes; N = no.
Is the conclusion true? truth. Is its truth rescued by the stated independent coincidence? luck.
A misleading but ordinarily dependable display supports 'machine M operates'. M is stopped. Unknown N operates, making 'some machine operates' true. Name the machine supporting the belief and the machine making its conclusion true; then say Y/N whether removing N leaves the conclusion true when all other machines are stopped. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Response symbols: Y = yes; N = no.
| Your reconstruction | |
|---|---|
| Apparent support | |
| Truth maker | |
| True without N |
A normally dependable record incorrectly says atlas A is large-print. A librarian infers a large-print atlas is available. A is ordinary print, but an unknown donated atlas B is large-print and available. Answer Y or N in both fields. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Response symbols: Y = yes; N = no.
Is the conclusion true? truth. Is its truth rescued by the stated independent coincidence? luck.
Lesson test: one question per skill, one attempt each, no hints. Your answers are checked when you submit.
A researcher reasonably believes sample C contains quartz from a deceptive test and infers some sample contains quartz. C contains no quartz. Unknown sample D does contain quartz. Answer Y or N in both fields. Use the following symbols in your response fields; these codes replace word answers and remain the same in every language. Response symbols: Y = yes; N = no.
Is the conclusion true? truth. Is its truth rescued by the stated independent coincidence? luck.
Without rereading, explain how to explain why a lucky true belief can fall short of knowledge. Give a fresh case and identify what would change your analysis.
10. The misleading crate label, step 3
The unknown compass in B
The final truth comes from a fact absent from the agent's grounds.