Category: epistemology
Type: thought experiment
Associated thinker(s): Laurence BonJour
Original publication or approximate period: Laurence BonJour, “Externalist Theories of Empirical Knowledge,” Midwest Studies in Philosophy 5 (1980), 53–74
Difficulty: Intermediate
Themes: reliabilism, justification, internalism and externalism
Related experiments: Gettier Cases; Fake-Barn Country; Brain in a Vat
Norman, under conditions that usually obtain, has a completely reliable clairvoyant faculty for certain kinds of subject matter. He has no evidence or reasons for or against the general possibility of such a power, or for or against his possessing it. One day he believes that the President is in New York City. He has no evidence for or against that particular belief; it is true and is produced by his clairvoyance in circumstances in which the faculty is completely reliable.
Does Norman know that the President is in New York? Is his belief justified?
Can a belief be justified—or amount to knowledge—solely because it comes from a reliable process, when the believer has no reason or evidence for trusting that process?
The case is commonly used to challenge simple process reliabilism, the view that a belief is justified when it is produced by a sufficiently reliable process. BonJour argues that, from Norman’s own perspective, the belief seems like an unsupported hunch: Norman has no reason to regard clairvoyance as a way of finding out where the President is. BonJour concludes that actual reliability alone does not make Norman’s belief epistemically justified. The case puts pressure on the sufficiency of reliability for justification; by itself, it does not settle whether every theory that gives reliability a role must accept the same result.
The debate is often framed in terms of internalism and externalism, but those labels cover different views. Accessibilist internalists require relevant reasons or evidence to be accessible to the believer. Some externalist theories deny that justification depends only on such accessible factors, while other theories combine external reliability with additional conditions. Nor do all philosophers agree that knowledge always requires justification or rational belief.
BonJour’s objection: Norman’s belief is epistemically irrational or irresponsible because he has no grounds, from his own point of view, for treating the belief as a reliable way of learning the President’s location. BonJour argues that this undermines its justification, even though the belief is in fact true and reliably produced.
Simple process reliabilism: On the simple account targeted by the case, the stipulated reliability appears sufficient for justification, contrary to BonJour’s judgment. This is why the case is a counterexample candidate, not a demonstration that reliability is irrelevant.
Replies and revisions: Reliabilists have offered different responses. Some defend the externalist result; others deny that their view implies that Norman’s belief is justified, or add conditions alongside reliability. Proposals discussed in the literature include approved-list approaches to justification attributions, requirements on the kind of cognitive system producing a belief, and views combining evidence with reliability. These strategies are not one uniform reply, and the case does not automatically refute every reliabilist or externalist theory.
The scenario stipulates a reliable clairvoyant faculty; it does not claim that clairvoyance exists. BonJour’s original discussion places Norman within a broader challenge to externalist accounts of justification and knowledge, while later reference treatments present the case as a problem for process reliabilism. The force of the example depends on what a theory says about justification, knowledge, and the conditions required in addition to reliability.
Norman is also not simply a Gettier case. Standard Gettier cases concern true, justified beliefs that still appear not to be knowledge because of epistemic luck; Norman’s case asks whether a reliably produced belief with no accessible reason for trusting its source is justified in the first place. This is a teaching contrast, not a claim that the cases raise no related questions about knowledge.
Keith Lehrer’s Mr. Truetemp is a related but distinct case: a person forms accurate temperature beliefs through an implanted thermometer-like device, without knowing about the device. It raises a similar question about reliable belief formation without the believer’s awareness of its source; the details differ from Norman’s clairvoyance case.
The scenario and BonJour’s objection are based on Case IV of “Externalist Theories of Empirical Knowledge” (1980), especially pp. 62–67. The primary text argues that Norman’s belief is unjustified from Norman’s own perspective, then considers possible externalist responses; its conclusion is BonJour’s argument, not a settled result. The Stanford Encyclopedia of Philosophy entries on reliabilism (§2.1 and §4.2), internalism/externalism (§§1 and 2.1), and the analysis of knowledge (§3), and the Internet Encyclopedia of Philosophy entry on reliabilism (§3(a)), were consulted for later framing, replies, and the comparison with Gettier cases. These sources discuss more than one response and do not establish that every reliabilist theory accepts Norman as justified or that the case is decisive against all externalism.
The Mr. Truetemp comparison is presented as a related case, not as an identical variation. The thermometer question in §6 is a teaching analogy. Research check: 2026-11-01. Relevant passages were consulted rather than every cited work being read in full; this educational overview is not a substitute for the primary text or a claim of expert review.