Epistemology — the philosophical study of knowledge — asks some of the most foundational questions in the discipline: What is knowledge? How is it different from mere belief? Can we ever be certain of anything? This guide walks through the classical theory of knowledge, the famous Gettier problem, the rationalism-empiricism debate, and skepticism, with the kind of detailed examples that make these abstract debates concrete enough to use in an assignment. These questions connect closely to philosophy of science and the scientific method, since both fields ask what justifies our beliefs about the world.
1. What Is Knowledge? The Classical (JTB) Analysis
Locke’s belief condition, discussed below, is also the starting point for his separate theory of what makes someone the same person over time, a puzzle developed fully in Personal Identity and Theories of the Self.
For most of the history of Western philosophy, “knowledge” was understood as justified true belief (JTB). On this analysis, a person S knows a proposition P if and only if:
- P is true (the truth condition)
- S believes P (the belief condition)
- S is justified in believing P (the justification condition)
Worked Example: Suppose Maria believes “it is raining outside.” For Maria to know it is raining, it must actually be raining (truth), she must genuinely hold that belief (belief), and she must have good reason to hold it — say, she can see rain through the window (justification). If it merely happens to be raining but Maria’s belief was based on a wild guess with no evidence, most epistemologists would say she has a true belief but not knowledge, because the justification condition fails.
1.1 Why Each Condition Matters
- Truth condition: You cannot “know” something false. You might sincerely and confidently believe the Earth is flat, but you cannot know it, because it is not true. Knowledge is factive.
- Belief condition: You cannot know something you do not believe. If you have overwhelming evidence for a claim but stubbornly refuse to believe it, you do not “know” it in the relevant sense — you merely have evidence available to you.
- Justification condition: This is what separates knowledge from lucky true belief. If you guess correctly on a coin flip, your true belief (“it landed heads”) isn’t knowledge because it wasn’t grounded in adequate evidence or reasoning.
2. The Gettier Problem
In 1963, Edmund Gettier published a short but devastating paper showing that the JTB analysis is not sufficient for knowledge — there are cases where someone has a justified true belief that we would intuitively deny counts as knowledge.
2.1 The Classic Gettier Case
Worked Example — “Stopped Clock”: Suppose you glance at a clock on the wall that reads 3:00 PM. The clock is, in fact, broken and has been stuck at 3:00 PM for hours — but by coincidence, it actually is 3:00 PM right now. You believe “it is 3:00 PM,” your belief is true, and you are justified (you have no reason to doubt a clock on the wall). Yet most people’s intuition is that you do not know it is 3:00 PM — you got the right answer by luck, not by a reliable process.
Worked Example — “Ford in the Barn”: Smith has strong evidence that his colleague Jones owns a Ford (Jones has driven one for years, shown him the title, etc.). From this, Smith validly infers “Jones owns a Ford, or Brown is in Barcelona” (even though Smith has no idea where Brown is). It turns out Jones sold his Ford yesterday and does not own one anymore — but by sheer coincidence, Brown really is in Barcelona. Smith’s belief is justified and true, but he does not know it, because the truth of his belief has nothing to do with the reasoning that justified it.
2.2 Responses to the Gettier Problem
Philosophers have proposed additional conditions to patch the JTB analysis:
- No false lemmas condition: knowledge requires that the justification not depend on any false intermediate belief (this rules out the “Ford in the Barn” case, since Smith’s justification relied on the false belief that Jones owns a Ford).
- Reliabilism: knowledge requires that the belief be produced by a reliable cognitive process (perception, memory, valid inference) — this rules out the “Stopped Clock” case, because reading a broken clock is not a reliable way to determine the time, even when it happens to be correct.
- Causal theory of knowledge: knowledge requires that the fact known actually cause (in an appropriate way) the belief that it is so.
None of these patches is universally accepted, and the search for a fully Gettier-proof analysis of knowledge remains an open problem in contemporary epistemology — a great example for assignments asking you to evaluate competing philosophical theories.
3. Rationalism vs. Empiricism
A central historical debate in epistemology concerns the source of knowledge.
3.1 Rationalism
Rationalists (e.g., Descartes, Spinoza, Leibniz) hold that at least some knowledge is gained through reason alone, independent of sensory experience (a priori knowledge).
Worked Example: Descartes’ cogito argument — “I think, therefore I am” (cogito ergo sum) — is presented as a truth known with certainty through pure reflection, not through sensory observation. Mathematical truths (e.g., “2 + 2 = 4,” or geometric theorems) are also classic rationalist examples: we do not need to go out and empirically survey triangles to know that their interior angles sum to 180 degrees in Euclidean geometry — we can prove it through reason.
3.2 Empiricism
Empiricists (e.g., Locke, Berkeley, Hume) hold that all (or nearly all) knowledge derives from sensory experience (a posteriori knowledge). Locke’s famous image is the mind as a tabula rasa (blank slate) at birth, filled in only through experience.
Worked Example: Hume’s empiricism led him to argue that even our belief in causation is not something we can rationally derive — we never directly observe “cause” itself, only one event constantly followed by another (e.g., the cue ball striking the eight ball, then the eight ball moving). Our belief that the first event causes the second is a habit of the mind formed by repeated observation, not a rational deduction — a radical empiricist claim with major implications, connecting directly to debates in philosophy of science about the justification of scientific laws.
3.3 Kant’s Synthesis
Kant famously attempted to reconcile the two positions, arguing that all knowledge begins with experience but is structured by innate categories of the mind (space, time, causality) that reason imposes on raw sensory data — famously summarized in his claim that “concepts without percepts are empty, percepts without concepts are blind.”
4. Types of Knowledge
Epistemologists commonly distinguish three kinds of knowledge, useful for framing assignment answers precisely:
- Propositional knowledge (“knowing that”): e.g., knowing that Paris is the capital of France.
- Procedural knowledge (“knowing how”): e.g., knowing how to ride a bicycle — this may not be fully expressible in propositions.
- Acquaintance knowledge (“knowing X”): e.g., knowing a person or a place through direct experience.
Most epistemological debates (including the JTB analysis and Gettier problem) concern propositional knowledge specifically.
5. Skepticism
Philosophical skepticism questions whether we can have knowledge (or certain kinds of knowledge) at all.
5.1 Cartesian Skepticism and the Evil Demon
Descartes used a method of radical doubt, imagining an “evil demon” of great power deceiving him about everything — even his sensory experiences and mathematical reasoning. He sought a foundation of belief immune to even this radical doubt, landing on the cogito (“I think, therefore I am”) as the one thing he could not doubt.
5.2 The Brain-in-a-Vat Thought Experiment
Worked Example: A modern version of Cartesian skepticism asks: how do you know you are not merely a brain in a vat, being fed simulated sensory experiences by a scientist (or, in contemporary versions, living in a simulated reality)? If your experiences would be indistinguishable either way, can you ever be certain your beliefs about an external world are true? This thought experiment (echoed in films like The Matrix) is frequently assigned precisely because it forces students to examine what, if anything, could count as an adequate response to radical doubt.
5.3 Responses to Skepticism
- Foundationalism: some beliefs are “basic” or self-justifying (e.g., immediate sensory experience, or Descartes’ cogito) and can serve as a non-circular foundation for all other justified beliefs.
- Coherentism: a belief is justified not by resting on a foundational belief, but by fitting coherently into a whole web of mutually supporting beliefs.
- Pragmatic responses: philosophers like G.E. Moore argued that our certainty in ordinary claims (“here is a hand”) is actually stronger than our certainty in the abstract skeptical premises used to doubt them, so skepticism should be rejected on that basis (the “Moorean shift”).
6. Sources and Limits of Justification
Beyond rationalism and empiricism as sources, epistemologists further examine the structure of justification:
- Foundationalism (as above): justification terminates in basic beliefs.
- Coherentism (as above): justification is holistic, not linear.
- Infinitism: justification consists of a potentially endless chain of reasons, with no ultimate foundation required.
Worked Example: If asked “why do you believe the sun will rise tomorrow?” — a foundationalist might appeal to a basic, direct perceptual/inductive starting point; a coherentist would point to how well this belief fits with the rest of your scientific worldview (physics, astronomy, past experience); an infinitist would argue there is always a further “why” question that could, in principle, be answered, without needing to terminate anywhere.
7. Common Assignment Pitfalls
- Treating the JTB analysis as if it were still universally accepted — always mention Gettier’s counterexamples when discussing JTB, since this is the expected critical move.
- Confusing skepticism (a philosophical position questioning the possibility of knowledge) with everyday “being skeptical” (mere doubt or caution).
- Conflating rationalism and empiricism as though they are mutually exclusive in every philosopher’s system — many philosophers (like Kant) synthesize elements of both.
- Failing to distinguish propositional knowledge from procedural or acquaintance knowledge when a question specifically asks about “knowing that” versus “knowing how.”
8. Applying These Concepts: A Sample Assignment Response
Question: “Does the Gettier problem show that knowledge cannot be defined as justified true belief? Explain with an example.”
Sample structured answer: Begin by stating the JTB analysis and its three conditions. Present a Gettier-style counterexample (e.g., the Stopped Clock case), showing that all three JTB conditions can be satisfied while intuition says the subject lacks knowledge. Conclude by evaluating at least one proposed solution (e.g., reliabilism), noting whether it successfully blocks the counterexample, and acknowledge that the debate remains unresolved in contemporary epistemology — this shows engagement with both the classic problem and the ongoing literature, which is exactly what graders look for for when preparing a strong philosophy assignment.
Epistemology forms the backbone of many other philosophical debates, including how we can be confident in ethical claims (see normative ethical theories), whether free choices can be known in advance (see free will and determinism), and how scientific theories are justified (see philosophy of science). Practice constructing and evaluating your own Gettier-style examples — this is the single most reliable way to demonstrate mastery of the topic in an exam or essay.
9. Contextualism and the Standards for Knowledge
A more recent development in epistemology, contextualism, argues that the standards required for a knowledge claim to count as true actually shift depending on conversational context.
Worked Example: In a casual conversation, saying “I know the bank is open on Saturday because I went last Saturday” seems perfectly acceptable. But if someone raises the possibility that the bank recently changed its hours and a large sum of money is at stake, the same evidence might no longer feel sufficient to justify the claim “I know it’s open” — even though nothing about your actual evidence has changed. Contextualists argue this shows that the word “know” tracks a variable epistemic standard that rises in high-stakes or skeptically-charged contexts, offering a distinctive diagnosis of why skeptical scenarios (Section 5) feel so gripping in philosophical discussion but seem irrelevant in ordinary life.
The question of whether we can be justified in our ethical beliefs is itself an epistemological one, a theme explored further in Normative Ethical Theories. The mind-body problem raises its own version of the justification question — how can we know another mind is conscious at all — which is examined in Philosophy of Mind and the Mind-Body Problem. The tools used above to evaluate the Gettier cases (validity, soundness, counterexample) are covered in more depth in Logic: Deductive and Inductive Reasoning. Locke’s empiricism also shaped his political theory, discussed in Social Contract Theory. Kierkegaard’s idea of subjective truth, touched on in Section 3, is explored at length in Existentialism. Finally, the question of whether aesthetic judgments like “this painting is beautiful” can ever be objectively justified is taken up in Aesthetics: Philosophy of Art and Beauty.
Frequently Asked Questions
What is the difference between knowledge and belief? A belief is simply something a person holds to be true, regardless of whether it actually is true or well-supported. Knowledge, on the traditional JTB analysis, additionally requires that the belief be true and justified. You can believe something false, but you cannot know something false.
What is the Gettier problem in simple terms? The Gettier problem shows that a person can have a justified true belief purely by luck, without that belief counting as genuine knowledge. It demonstrates that justified true belief is not sufficient for knowledge, even though most philosophers agree it is necessary.
Is rationalism or empiricism correct? Neither position is universally accepted as simply “correct” — most contemporary epistemologists hold a mixed view, often influenced by Kant, that some knowledge depends on structuring concepts supplied by reason while its content still derives from experience. Assignments typically ask you to evaluate the strongest arguments on each side rather than declare one the winner.
How do you respond to skepticism about the external world? Common responses include foundationalism (grounding belief in indubitable basic beliefs), coherentism (justification through a web of mutually supporting beliefs), and G.E. Moore’s pragmatic response that our certainty in ordinary claims outweighs our certainty in the skeptical premises used to doubt them.
Why does epistemology matter outside of philosophy class? Epistemology underlies how we evaluate evidence, expertise, and claims in everyday life, science, journalism, and law — any time you ask “how do we know that?” you are doing epistemology.
