amazon.co.ukAQA 7172 Paper 1 Section A: the tripartite definition of knowledge, Gettier counter-examples, direct realism, indirect realism, idealism and knowledge from reason — with 25-mark essay technique and full mark-scheme answers.
Three types of knowledge. The spec opens with the distinction between propositional knowledge ('knowing that' — Paris is the capital of France), ability knowledge ('knowing how' — knowing how to ride a bike) and acquaintance knowledge ('knowing of' — knowing Oxford). The tripartite analysis targets propositional knowledge only. AQA examiners regularly report candidates losing marks by illustrating 'knowledge' with skills — use propositions.
The tripartite view. S knows that p iff: (1) p is true — you cannot know something false; (2) S believes that p — knowledge requires belief, not mere assent; (3) S is justified in believing that p — the belief must be backed by good grounds (perception, memory, testimony, reasoning), excluding coin-flip guesses that happen to be right. Conditions (1) and (3) must be independent: if justification consists in 'it seems true', the definition collapses.
Gettier's challenge (1963). Gettier produced cases where all three conditions hold yet intuition denies knowledge. Standard case: Smith has strong evidence that Jones will get the job and that Jones has ten coins in his pocket, so Smith infers and believes 'the man who will get the job has ten coins in his pocket'. Unknown to Smith, he gets the job and also has ten coins. The belief is true, justified — and pure epistemic luck. Zagzebski's recipe shows the problem is general: engineer any justified false belief (a false lemma), then bolt the truth on by luck; the tripartite conditions will be met.
Responses. (1) No false lemmas: knowledge requires that the justification not rest on any false proposition. Smith's rests on the false 'Jones will get the job'. Critique: Gettier-style cases with no false lemmas (fake-barn variants where every belief in the chain is true) still generate luck, so the fix is incomplete. (2) Reliabilism (Goldman; Nozick's truth-tracking): knowledge is true belief produced by a reliable process. In Gettier cases the belief-forming process (inference from a false lemma) is unreliable. Critique: in a field of papier-mâché barns, the one genuine barn seen in normal conditions yields a reliably-formed true belief that most still judge not knowledge — reliabilism either still admits luck or rigs the notion of 'reliable process'. (3) Virtue epistemology (Zagzebski, Sosa): knowledge is true belief because of intellectual virtue (careful perception, good reasoning); Gettier luck means the truth is not creditable to the agent's virtue. Critique: beliefs from a virtuous process in an unvirtuous environment (the barn field) still seem lucky, so 'because of' must do heavy lifting. (4) Infallibilism: strengthen the justification condition so that only beliefs beyond any possibility of error count as knowledge (Descartes' certainty standard). Critique: the bar is too high — it yields virtually no perceptual knowledge and cannot explain fallible everyday knowledge.
Direct realism holds that in perception we are in immediate, direct contact with mind-independent physical objects themselves: when I look at the table, I perceive the table, and (in good conditions) as it really is. Attractions: it is the common-sense view, it invokes no hidden intermediaries, and it explains how perception straightforwardly yields knowledge of the world.
Argument from perceptual variation. Russell's table: from one angle it looks distinctly brown and rectangular; from another, paler and diamond-shaped; Locke's straight stick looks bent in water. The same single object appears F to one observer and G to another; if we directly perceived the object as it is, we would contradict ourselves ('the table is brown and not brown'). Conclusion: what we directly perceive is not the object itself but an appearance (sense-data), so direct realism is false. Direct realist response: the argument shows only that objects look different from different points of view — 'looks' reports how the object appears from a perspective, not a second object perceived; there is no contradiction in one object appearing different ways from different angles.
Argument from illusion. In illusion something with property F appears to have property G (the straight stick appears bent). What we are directly aware of in illusion — the bent appearance — is not the physical object as it is. But illusion is subjectively indistinguishable from veridical perception; so in veridical perception too we are directly aware only of appearances. Conclusion: we never directly perceive physical objects. Response: the inference overreaches — from 'in illusion we do not perceive a bent object' it does not follow that we never perceive objects directly; in illusion we may still perceive the very stick, with the experience merely misdescribing it (we perceive that it appears bent, not that it is bent).
Argument from hallucination. A subject hallucinating a pink rat has an experience subjectively indistinguishable from really seeing one. Since in hallucination there is no pink rat being perceived, what the subject is directly aware of is a mere appearance; by indistinguishability, veridical perception is the same kind of awareness — so perception is never direct. Response: hallucination is a degenerate, abnormal case; we cannot generalise from it about what normal perception is like. (This reply is weaker than for illusion: the indistinguishability premise is harder to shift here.)
Time-lag. Light and sound take time: we see the star as it was, not as it is now; so we never perceive objects as they presently are. Response: 'as it is now' sets the bar too high — seeing the star is directly seeing a (past-stage) physical object. Examiner tip: in 12-markers on this section, present the argument's premises and conclusion cleanly, then the specified response — no evaluation (AO1 only).
Indirect realism (Locke). We perceive mind-independent physical objects indirectly: the direct objects of perception are sense-data caused by, and in some way representing, the objects. Locke's primary/secondary quality distinction does the representational work: primary qualities (shape, size, motion, number, solidity) are inseparable from the object and our ideas of them resemble them; secondary qualities (colours, tastes, sounds) are merely powers objects have to produce sensations in us — the redness is in the experience, not in the tomato. This underwrites a realist world of minute particles and explains illusion and perceptual variation: the senses reliably tell us 'when something has a property, it seems to have it' (Locke), even when appearance misleads.
Issue: the veil of perception. If we only ever perceive sense-data, we can never 'step outside' our experiences to compare them with the objects they allegedly represent. So the claims that sense-data are caused by and resemble mind-independent objects are unverifiable — we are sealed behind a veil. Berkeley presses further: an idea can only be like another idea (the likeness principle), so mind-dependent sense-data cannot resemble mind-independent objects at all — the resemblance claim is incoherent, not merely unverifiable. Related sceptical issue: the very existence of the external world no longer follows from perception: all my experiences would be exactly as they are if caused by something else (a dream, a demon, nothing at all). Responses: (1) Inference to the best explanation (Russell): the hypothesis that mind-independent objects cause our sense-data is the best explanation of their involuntary character, their coherent cross-sensory structure (what I see, I can touch), and their patterned, partly unpredictable regularities; a dream or demon explains the pattern less well. (2) 'Resemblance' is needed only for primary qualities, where our ideas are structural (shapes, sizes) and comparison is less mysterious.
Idealism (Berkeley). Esse est percipi — to be is to be perceived. Only minds and their ideas exist; there is no mind-independent matter. Berkeley first attacks the primary/secondary quality distinction: secondary qualities (colour, taste) are plainly mind-dependent — but primary qualities are no better off, since size, shape and motion are all relative to the perceiver (what looks small to me is large to a mite); so if secondary qualities exist only in the mind, primary qualities do too — both are ideas. What we call 'physical objects' are stable, lawful collections of ideas; the immediate objects of perception are exactly these ideas, so idealism takes the indirect realist's sense-data and stops inferring anything behind them. Berkeley's master argument: try to conceive a tree existing unperceived — in the very act of conceiving it, you are perceiving it; so an unperceived object is inconceivable, so mind-independent matter is impossible. Critique: this conflates conceiving a tree with conceiving a tree-unperceived: the content of the second thought is different, and it is the second that is coherent.
God's role. Objects persist when no human perceives them because God perceives everything always; the regularity of our ideas reflects God's uniform will, which also explains why the world is orderly and why perception yields knowledge. Critiques of idealism: (1) it makes the existence of tables and mountains depend on minds — implausible for a world that apparently existed before any minds; God rescues continuity only by loading theology into metaphysics; (2) idealism appears to conflate the act of perceiving with the object perceived; (3) problems with God's role: Berkeley holds that God cannot feel pain or have sensations — so how can our painful, colour-like ideas exist in God's mind? God's perception explains the existence of objects but not, on Berkeley's own view of God, what our ideas are like; (4) idealism is said to lead to solipsism: if only minds and ideas exist, my evidence for other minds weakens; the veil-of-perception worry dissolves — but at the price that what exists is not the world common sense believes in.
Rationalism vs empiricism. A priori knowledge is had independently of experience; a posteriori knowledge depends on experience. Rationalists (Plato, Descartes) hold that some substantive knowledge of the world is a priori; empiricists (Locke, Hume) deny this: all substantive knowledge of the world is a posteriori, with reason merely operating on material supplied by the senses.
Innate knowledge (Plato). In the Meno, Socrates questions an untutored slave boy and gets him to 'discover' a geometrical truth (doubling the square). Since the boy was never taught it, his knowledge must be innate — recollected (anamnesis) from a prior existence; the argument from recollection generalises: learning is remembering. Locke's critique: (1) argument from ignorance: children and 'idiots' plainly do not know geometry or the supposed universal maxims — a truth universally present in the mind cannot be unknown to the person whose mind it is; (2) there is no universal consent on the speculative candidates (moral and religious principles vary across cultures), and where agreement exists (logic, mathematics) it can be explained by shared cognitive equipment plus instruction; (3) the mind at birth is white paper (tabula rasa): what feels 'innate' is merely what experience delivers very early and effortlessly. Platonist reply: innate knowledge may be dispositional — a capacity triggered rather than consciously possessed from infancy. Leibniz's argument from necessary truths (New Essays): experience only ever supplies instances — however many white swans we see, never the necessity of 'all swans are white' — yet we grasp necessary truths (mathematics, logic) whose denial is inconceivable; so necessary truths cannot come from experience and must be innate, as dispositions or 'inclinations' (Leibniz's veined marble: the veins are in the block from the start, though work is needed to reveal them).
Intuition and deduction thesis (Descartes). Some truths are grasped by intellectual intuition as self-evident; from these, deduction yields further substantive knowledge. In the Meditations: the cogito — 'I think, I am' — is indubitable, since the attempt to doubt it presupposes a thinking 'I'; this establishes that I am a thinking thing (res cogitans). The trademark argument: the idea of an infinite, perfect being could not have originated in a finite, imperfect mind; by the causal principle (a cause must contain at least as much reality as its effect), only God could cause such an idea; so God exists — and the idea is God's 'trademark'. Meditation 5 adds the a priori proof from the essence of a triangle: existence belongs to God's essence as necessarily as three angles to two right angles belong to a triangle.
Hume's fork (empiricist critique). Every proposition is either a relation of ideas (a priori, analytic, necessary — to deny it is a contradiction: '2+2=5') or a matter of fact (a posteriori, contingent — its denial is conceivable: 'the sun will not rise'). No proposition can be both substantive about the world and knowable a priori. Applied to Descartes: (1) the causal principle powering the trademark argument is not a relation of ideas — 'every event must have a cause' is, for Hume, a matter of fact known only from experience; (2) the cogito, if true, is singular and fleeting — it establishes a thinking self only 'while I think'; whether the self persists needs a further, a posteriori argument. Rationalist reply: intuition and deduction demonstrably yield necessity that experience never can (mathematics); Hume's fork is itself a philosophical claim not certified by his own dichotomy.
Paper format. Paper 1 (3 hours, 100 marks) has Section A: Epistemology and Section B: Moral philosophy — five questions per section worth 3, 5, 5, 12 and 25 marks. Every question except the 25-marker is pure AO1; the 25-marker is 5 AO1 + 20 AO2. Overall the qualification weights AO1 60% / AO2 40%.
Command words. 'Explain…' questions (3/5/12 marks) demand explanation only: define the key terms, set out the claim, argument or distinction, illustrate where asked. Evaluation in short questions is redundant and loses marks — examiners report penalising 'it could be objected…' material on 12-markers. The 25-marker presents a quoted claim followed by 'Evaluate this claim': here you must argue with intent — commit to a thesis in the introduction, defend it through objections, and finish with a consistent conclusion.
25-mark essay structure (top band 21–25). (1) Introduction: define key terms, state your thesis. (2) Argument for your view, with premises and conclusion; consider the best objection; respond. (3) Argument against your view; give a weak response, then the strong response showing why the weak one fails. (4) Conclusion: restate the thesis as supported by the argument — no new ideas. The mark scheme rewards 'crucial arguments clearly identified against less crucial ones': develop one or two debates in depth rather than listing everything.
Common pitfalls and examiner tips (click to expand)