Why the IB Philosophy Paper 2 unseen stimulus decides the 7 before

IB Philosophy is one of the few subjects in the IB Diploma where the syllabus list alone will not lift a candidate from a 5 to a 7. The marks sit on the ability to handle a stimulus the candidate has never seen before, to identify the philosophical move it is making, and to argue against a strong objection rather than a straw one. This article focuses on that pressure point, with most of the discussion anchored to IB Philosophy Paper 2, where the unseen stimulus is the entire exam, and shorter detours into Paper 1, the HL Paper 3, and the internal assessment. By the end, the reader should be able to recognise each IB Philosophy command term on contact, sketch a counter-example for any central claim, and plan a Paper 2 response that survives a 90-minute desk.

What IB Philosophy Paper 2 actually tests beyond the syllabus list

IB Philosophy Paper 2 is the only externally assessed component in the IB Diploma that asks candidates to respond to material they have never studied in class. The paper is built around an unseen stimulus — usually a short extract from a philosophical text, a thought experiment, or a case study — and a menu of structured questions. Candidates choose two of those questions and write a response to each. There is no choice of topic in the traditional sense; the stimulus is fixed, and the questions are designed to probe a small number of philosophical skills rather than a body of content.

For most IB Philosophy candidates, the instinct is to read the stimulus once, recognise a word from the syllabus, and start writing. That instinct is the most common reason a candidate scores in the middle band. The paper does not reward recognition; it rewards identification of the underlying philosophical move. A stimulus that mentions "the veil of ignorance" is not a test of whether the candidate remembers Rawls. It is a test of whether the candidate can separate the contractarian move from the egalitarian conclusion, and then construct an objection that does not depend on attacking Rawls by name.

Three skills carry almost all of the available marks. The first is conceptual analysis: the ability to define a key term in the stimulus precisely, to distinguish it from neighbouring terms, and to show that the argument depends on that distinction. The second is argument construction: the ability to state a thesis, supply a reason, and then test the reason against a counter-example. The third is counter-example handling: the ability to propose a case that breaks the stimulus argument, and to explain why the case breaks it, in two or three sentences rather than a paragraph of throat-clearing.

A useful diagnostic: a candidate who reads a Paper 2 stimulus and immediately thinks of a philosopher is already on the wrong track. A candidate who reads the same stimulus and immediately asks "what is the central claim, what does that claim depend on, and what would have to be true for the claim to fail" is on the right one. The IB markscheme rewards the second move because it is the only one that scales across stimuli. A candidate who has memorised the syllabus cannot transfer that memorisation to a new text; a candidate who has practised the three skills above can.

Two further points are worth noting before moving to the command terms. First, IB Philosophy does not reward hedging. A response that says "some philosophers might argue X, while others might argue Y" without committing to either is a band 3 response by default, because it shows no philosophical judgement. Second, the rubric is explicit that the highest band requires the candidate to do something the stimulus does not. Restating the stimulus, however accurately, caps a response in the middle band. The candidate must push the stimulus somewhere the writer of the stimulus did not intend to go.

The four command-term families that reshape every Paper 2 response

IB Philosophy uses the IB Diploma's shared command-term vocabulary, but in this subject the command terms do more work than they do in, say, ESS or Psychology. The same stimulus can produce a band 2, a band 5, or a band 7 response depending on which command term the candidate treats as dominant. The four families that matter most on Paper 2 are examine, discuss, evaluate, and to what extent. Each one restructures the response in a way that candidates consistently underestimate.

The "examine" family (including "examine the claim", "examine the argument", "examine the distinction") asks the candidate to lay out a piece of reasoning and identify its load-bearing structure. A response to an examine question is judged primarily on the accuracy of the reconstruction. Marks are lost when the candidate summarises the stimulus in their own words without showing the logical shape of what is being claimed. A clean way to handle "examine" is to write the central claim in one sentence, identify the two or three premises it depends on, and then check each premise against the text. If a premise is missing from the text, the candidate should say so; that observation is itself a mark.

The "discuss" family is the most common command term on Paper 2 and the one that causes the most confusion. "Discuss" does not mean "give both sides". It means "engage with a tension inside the claim". A response to "discuss the claim that X" should identify what makes X attractive, identify what makes X fragile, and then take a position on whether the attraction outweighs the fragility. Candidates who treat "discuss" as a synonym for "present two views" produce a response that the IB markscheme caps in the middle band, because the response never arrives at a judgement. In my experience, the most reliable structure for a discuss response is: claim, strongest reason to accept, strongest reason to reject, judgement, response to a likely objection to the judgement.

The "evaluate" family asks the candidate to weigh a claim against a counter-example or an alternative position. "Evaluate the claim that X" requires a counter-example; without one, the response is descriptive rather than evaluative, and the top band is out of reach. A practical move: before writing, draft a one-sentence case that the stimulus author would not want to be true, and then check that the case actually breaks the claim. If the candidate cannot construct such a case, the stimulus has not been understood at the level the question is asking.

The "to what extent" family is the highest-leverage command term on Paper 2 and the one most candidates underuse. A "to what extent" question is not asking for a yes or no; it is asking the candidate to specify the conditions under which the stimulus claim holds and the conditions under which it fails. A strong response will say something like "the claim holds in cases where the agent can be reasonably expected to foresee the consequence, and fails in cases where structural factors make the consequence invisible to the agent". That kind of conditional answer is what separates a band 5 from a band 7.

Across all four families, the practical rule is the same: the command term tells the candidate what the conclusion of the response is supposed to look like. A response to "examine" should end with a labelled diagram of the argument. A response to "discuss" should end with a judgement. A response to "evaluate" should end with a counter-example. A response to "to what extent" should end with a conditional. Candidates who write the same kind of paragraph to every command term — claim, evidence, conclusion — are giving up two marks per question by default.

Quick-reference: command-term → response shape

• Examine → label the argument's structure; flag any premise the stimulus assumes without supplying.
• Discuss → state the tension, take a side, defend the side against a counter-objection.
• Evaluate → construct a counter-example; explain why it breaks the claim; concede the limit of the counter-example.
• To what extent → specify the conditions of success and the conditions of failure; refuse a blanket answer.

How the unseen stimulus is supposed to be used — and how most candidates waste it

The unseen stimulus on IB Philosophy Paper 2 is roughly 500 to 700 words of dense philosophical prose, often with a thought experiment embedded in the middle. The IB guidance to candidates is explicit on what the stimulus is for, but the guidance is widely misread. The stimulus is not a passage to be agreed or disagreed with in the abstract; it is a piece of reasoning that contains at least one move the candidate is supposed to identify, interrogate, and either defend or undermine.

The most common waste of the stimulus is to summarise it in the first paragraph and then write a response that could have been written without ever reading it. The IB markscheme reads for stimulus engagement as a discrete criterion. A response that ignores the stimulus, or treats it as decoration, loses marks on that criterion even if the philosophical content is strong. Conversely, a response that engages with the stimulus in every paragraph — that quotes a specific phrase, that names the move the phrase is making, that tests the move against an alternative — picks up marks on engagement even when the philosophical content is rougher.

A practical method: the first read of the stimulus should take about 4 minutes. During that read, the candidate should underline every term that is doing argumentative work, write a one-word label next to each, and then in the margin summarise the central claim in a single sentence. The second read, taking 2 to 3 minutes, should be for the specific questions in the menu. The candidate should check which question targets which underlined term. A candidate who has done the first read properly can usually answer two of the four menu questions without re-reading the stimulus a third time.

Engagement has three measurable forms. The first is lexical: the response uses a key term from the stimulus as a term of art, not as a synonym for something more familiar. The second is structural: the response identifies a move in the stimulus (a reductio, a distinction, an analogy) by name. The third is counterfactual: the response proposes a change to the stimulus — a modified case, a stronger premise, a different agent — and shows what changes in the argument as a result. A response that hits all three forms is a band 7 response; a response that hits one is a band 5; a response that hits none is a band 3.

A worked Paper 2 response on a stimulus from political philosophy

To make the above concrete, it helps to walk through a worked example. Consider a stimulus of the following shape (this is illustrative, not a past paper): a writer argues that civil disobedience is justifiable only when the law being broken is itself unjust, the disobedience is public, and the actor accepts the legal consequences. The candidate chooses the question "To what extent is the third condition necessary?".

The first move is to restate the third condition precisely. The condition is not "the actor regrets the act"; it is "the actor willingly accepts the punishment prescribed by the law being broken". That distinction matters, because a candidate who collapses the two will write a response about sincerity rather than about necessity. The second move is to identify what the third condition is doing in the argument. It is doing two things: it is filtering out bad-faith disobedience, and it is preserving the actor's standing as a member of the legal community. A response that engages with the stimulus must engage with both of these, not just the first.

The third move is the counter-example. A candidate might propose: a whistleblower who reveals a state secret and then flees the jurisdiction. The whistleblower has broken an unjust law, has done so publicly, but does not accept the legal consequences. Is the civil disobedience justifiable? The strongest answer is that it depends on whether the value of the disclosure outweighs the value of the actor's continuing participation in the legal community. That conditional answer is exactly the shape a "to what extent" response is supposed to take.

The fourth move is the response to a likely objection. The objection is that the whistleblower case shows only that the third condition is sometimes overridable, not that it is unnecessary. The candidate should concede the point, then reframe: the third condition is necessary in the default case, and overridable in cases where the legal community has itself forfeited its claim on the actor. That reframing is what pushes the response into the top band, because it shows the candidate has thought about the condition's scope, not just its content.

Note what is missing from the worked response: there is no appeal to a famous philosopher by name, no restatement of a textbook definition, and no generic claim about "philosophy shows that". The response engages with the stimulus, identifies a structural move, constructs a counter-example, and arrives at a conditional judgement. That is the architecture of a band 7 Paper 2 response, and it can be rehearsed in advance on any stimulus the candidate has not seen before.

HL Paper 3 and the conceptual analysis question: where the 7 separates from the 5

IB Philosophy HL candidates sit a third paper, often called Paper 3, that tests a deeper level of philosophical skill. The paper is built around a longer unseen stimulus and a smaller menu of questions, of which one is almost always a conceptual analysis question. A conceptual analysis question asks the candidate to take a term from the stimulus, distinguish it from two or three neighbouring terms, and show that the distinction is doing work in the argument. The question is the single most reliable discriminator between a band 5 and a band 7 on the HL paper.

The reason conceptual analysis is so demanding is that it cannot be faked by definition. A candidate who responds to "distinguish X from Y" by writing "X means A, Y means B" has not done the work; the markscheme wants the candidate to explain why a reader who conflates X and Y will misread the argument. A useful test: if the candidate can write a sentence of the form "if X and Y were the same, then the argument would conclude Z, but the argument is supposed to conclude W", the candidate is on the right track. If the candidate cannot, the distinction has not been made to do any work.

Three conceptual-analysis traps are worth flagging. The first is the etymology trap: candidates spend a paragraph on where a term comes from in Greek or Latin. The IB markscheme does not award marks for etymology. The second is the example trap: candidates give three examples of the term in action and stop. Examples are useful only if the candidate can say what the examples share and how that shared feature differs from the neighbouring term. The third is the stipulation trap: candidates announce that they will use the term in a particular way and then proceed as if the stipulation were a definition. The markscheme reads for engagement with how the stimulus is using the term, not for the candidate's preferred usage.

For candidates aiming at the 7 boundary on HL Paper 3, the practical move is to read at least one full Plato and one full Descartes extract in the months before the exam, not to study them as historical artefacts but to practise the move of pulling a single term out and tracing how the term shifts meaning across a short argument. That habit transfers directly to the unseen stimulus. A candidate who has practised the move on Plato will recognise the same move in an unfamiliar contemporary text.

The internal assessment in IB Philosophy: a textual analysis that lives or dies on three sentences

The IB Philosophy internal assessment is a textual analysis of a non-philosophical text — an extract from a political speech, a legal judgement, a corporate press release, a religious sermon, or similar. The IA is judged on the candidate's ability to identify the philosophical assumptions embedded in the text and to evaluate them. The IA is unusual among IB subjects in that the rubric rewards three specific sentences above almost everything else.

The first sentence is the identification sentence, in which the candidate states, in plain language, the philosophical assumption that the text is making. A common failure: the candidate writes a sentence that is too broad ("the text assumes that humans are rational") or too narrow ("the text assumes that the speaker's audience will agree with the third clause"). The IB markscheme looks for a sentence that is specific enough to be tested and general enough to be philosophical. A useful diagnostic: if the candidate can write a counter-example that contradicts the identification sentence, the sentence is at the right level of specificity.

The second sentence is the explanation sentence, in which the candidate shows that the text depends on the assumption, by tracing a passage of the text to the assumption. This is the sentence that most candidates skip, on the assumption that the identification is self-evidently present in the text. The markscheme disagrees; it requires the candidate to demonstrate the dependence, not assert it.

The third sentence is the evaluation sentence, in which the candidate proposes a counter-argument to the assumption, draws on a philosophical position the candidate has studied, and then limits the force of the counter-argument by stating a condition under which the original assumption would be defensible. The evaluation sentence is where the IA rubric is won or lost; candidates who stop at the counter-argument cap their response in the middle band, because they have not shown the philosophical judgement the rubric is looking for.

A practical IA planning move: before writing, the candidate should draft those three sentences in order, on a single page, and revise them until each one survives a friend who has not read the text. If the sentences survive, the rest of the IA is mostly structural work; if they do not, the rest of the IA will not save them.

Common pitfalls and how to avoid them across Paper 1, Paper 2 and the IA

IB Philosophy produces a small number of recurring failure modes across all three components. A candidate who knows the failure modes in advance can correct them in real time; a candidate who meets them for the first time on exam day usually cannot. The list below pairs each pitfall with a concrete correction.

1. The summary trap. The candidate spends the first half of the response restating the stimulus, the text, or the question, and the second half on a generic argument. Correction: cap the summary at two sentences, and require every subsequent paragraph to do something the summary did not do.

2. The authority trap. The candidate appeals to a named philosopher as if the appeal were itself an argument. Correction: replace every appeal of the form "Plato says X" with a sentence of the form "If we accept X, then Y, but Z shows that Y is false". The philosopher is irrelevant; the argument is what matters.

3. The hedging trap. The candidate uses "some would say, others would say" without taking a position. Correction: every response must end with a sentence the candidate is willing to defend in a follow-up question. If the candidate cannot defend it, the response is not finished.

4. The example-as-proof trap. The candidate offers a single example as if the example settled the question. Correction: every example must be followed by a sentence that explains what the example shows and what it does not show. A response that confuses the two is a band 4 response.

5. The command-term blur. The candidate writes the same kind of paragraph in response to "examine", "discuss", "evaluate", and "to what extent". Correction: see the response-shape table in section 2. The shape of the conclusion is determined by the command term, not by the candidate's preference.

6. The IA identification trap. The candidate writes an identification sentence that is either too broad to test or too narrow to be philosophical. Correction: the candidate should be able to construct a counter-example that contradicts the identification. If they cannot, the identification is too vague.

None of these corrections requires new content. They require the candidate to spend the same amount of thinking time on the response shape as on the content. In my experience this is the single largest gap between a 5 and a 7 in IB Philosophy: the 7 candidate writes roughly the same amount of philosophy as the 5 candidate, but in a shape that the rubric can reward.

A Paper 2 preparation plan that fits inside a realistic study window

IB Philosophy preparation is unusual in that the syllabus list is small, but the skill set is large. A candidate who tries to cover more content will not improve; a candidate who practises the same five moves on a wider range of stimuli will. The plan below assumes a 12-week window before Paper 2, with roughly 4 hours of focused work per week, and is designed to be repeatable.

Weeks 1 to 3 are for command-term fluency. The candidate should take four unseen stimuli — one from ethics, one from political philosophy, one from epistemology, one from philosophy of religion — and write a 200-word response to each, switching command terms. By the end of week 3, the candidate should be able to look at a command term and produce, in writing, the shape of the response it requires, without re-reading the question.

Weeks 4 to 7 are for counter-example construction. The candidate should take a single philosophical claim (for example, "lying is always impermissible") and produce, in writing, three distinct counter-examples, each one followed by a two-sentence explanation of why it breaks the claim. The exercise is harder than it sounds; the same counter-example rephrased three times is a single counter-example, not three.

Weeks 8 to 10 are for full Paper 2 simulations. The candidate should sit one full Paper 2 under timed conditions every week, marking strictly against the IB markscheme. The first simulation will probably score in the band 4 range; the third should be in the band 6 range if the previous seven weeks have been done honestly.

Weeks 11 to 12 are for shape review. The candidate re-reads the responses from weeks 8 to 10 and labels, in the margin, which paragraphs hit the three forms of engagement (lexical, structural, counterfactual) and which do not. The candidate then rewrites the weakest two paragraphs of each response, holding the content constant and changing only the shape.

Two practical notes close the plan. First, the plan does not include any new content reading after week 7; everything from week 8 onwards is on unseen stimuli. Second, the plan assumes the candidate is doing one IA alongside the Paper 2 preparation. The IA work should be slotted into weeks 4 to 7, when the candidate is in the counter-example construction phase, because the IA's evaluation sentence is structurally identical to a Paper 2 counter-example.

The single most useful diagnostic at the end of the 12 weeks is this: the candidate should be able to read a 600-word stimulus they have never seen before and, in under 5 minutes, write down the central claim, the two premises it depends on, and one counter-example that would break it. A candidate who can do that is in the band 6 to 7 range on Paper 2, regardless of how much of the syllabus they can recite.

IB Philosophy rewards a specific kind of reading and a specific kind of writing. The reading is slow, focused on the argumentative structure, and resistant to the temptation to recognise rather than to analyse. The writing is shaped by the command term, anchored to the stimulus, and willing to take a position it can defend. A candidate who internalises those two habits before the exam will outperform a candidate who has memorised twice as much content. For candidates building a study plan around IB Philosophy, the most efficient use of the final weeks is to drill the command-term response shapes against a stream of unseen stimuli, and to treat the IA's three load-bearing sentences as a parallel exercise in the same skill set.

IB Courses' one-to-one IB Philosophy HL programme drills the four Paper 2 command-term response shapes against a stream of unseen stimuli and rebuilds the IA around its three load-bearing sentences, so a 7 target becomes a concrete preparation plan rather than a wish.

Frequently asked questions