H1 General Paper Paper 2 Summary is a test of controlled reduction. The candidate must remove material without removing meaning, combine overlap without flattening distinctions, paraphrase without changing claim strength and write concisely without turning the response into notes. A weak summary is often not a language problem first. It is a selection problem.
This page is the SETC canonical owner for the Paper 2 Summary examination-performance layer. It does not replace the broader SETC owner for concise writing or general paraphrase and comprehension. Those pages own universal compression and sentence-control mechanics. This page owns the summary-specific sequence inside H1 GP Paper 2: identify the target, select candidate points, classify information level, cluster overlap, compress conceptually, verify semantics and finish within the paper’s time budget.
Under the current 2027 syllabus 8881, the Paper 2 summary is based on Passage 2 and carries 8 marks. Future candidates should verify the syllabus for their own examination year with SEAB before relying on exact mark or format details.
For deeper generic compression mechanics, use Concise Writing Under Exam Pressure. For sentence clarity, use How Sentence Control Works. For the whole Paper 2 system, return to H1 GP Paper 2 Examination Performance.
Summary is target control before word control
The first question is not “How do I shorten this sentence?” It is “Does this idea belong?” A beautifully paraphrased irrelevant point earns nothing. A clumsy but correctly selected point at least begins in the right place. Selection therefore precedes compression.
Students who begin summarising from the first sentence of the passage often produce orderly but unfocused responses. The task should be read as a filter. Only ideas that answer the specified target are candidates for inclusion.
The Summary operating model
- Identify the target. Rewrite it in a short phrase.
- Scan for candidate points. Collect only ideas directly within scope.
- Classify information level. Core point, explanation, example, evidence or consequence.
- Separate distinct from overlapping ideas. Do not double-count paraphrased versions of the same point.
- Cluster where safe. Merge ideas only when meaning remains intact.
- Paraphrase conceptually. Restate the idea, not every individual word.
- Compress sentence structure. Remove repetition and unnecessary framing while keeping logic.
- Verify semantics. Preserve quantity, certainty, causality, comparison and qualification.
- Check completeness and grammar. Concision should not become fragments.
- Close and move. Protect the rest of Paper 2.
Maren and Iona share the Summary job
Iona owns scope and meaning. She must decide whether a point answers the target, distinguish a reason from an example and preserve the writer’s actual claim strength.
Maren owns compression architecture. She must cluster overlapping material, build concise grammatical sentences and organise the selected points without creating a new argument that did not exist in the passage.
Leonie protects time. Her Summary risk is perfectionism: rewriting one sentence repeatedly while the Connections and Application tasks wait.
Write the target box first
Reduce the summary task to a five- to eight-word target box. Examples: reasons trust declines; problems caused by urban congestion; benefits of public green spaces; why misinformation persists. Keep the box visible during selection.
Every candidate point must fit the box. If the point describes a consequence while the target asks for causes, it does not belong unless the passage explicitly presents that consequence as part of the causal explanation.
Target words create boundaries
Reasons, effects, benefits, problems, ways, barriers and concerns point to different information classes. The noun in the target matters as much as the topic. Students who summarise the theme instead of the required class usually collect too much.
Core point, explanation and example are not the same
A passage may state: “People distrust automated decisions because they cannot see how the system reached its conclusion. For example, applicants rejected by an algorithm may receive no explanation.” The core point is lack of transparency. The explanation is inability to see the reasoning process. The example is rejected applicants. Depending on the target, one concise formulation may capture the point without preserving the example.
Examples usually support; they do not automatically become separate points
Students often count each example as an extra idea. Ask whether removing the example would leave the underlying point intact. If yes, the example may not deserve separate summary space. If the example introduces a distinct cause or consequence, it may contain a new point. Function decides.
Explanations may or may not be separate points
Sometimes a supporting explanation merely develops the same idea. Sometimes it adds a distinct mechanism that the summary target rewards separately. Train the learner to ask whether the explanation changes what is being claimed or simply clarifies it.
Collect candidate points before writing sentences
Do not paraphrase immediately. Mark or note candidate points in compressed conceptual form. This separates selection from expression. A student who writes polished sentences too early becomes reluctant to remove them even when later analysis shows they overlap or fall outside scope.
Distinguish overlap from adjacency
Two ideas can appear next to each other without being the same. Conversely, the same idea can reappear in different paragraphs. Compare underlying propositions, not location or wording. The question is: if I include both, do I gain a new meaning?
Safe clustering
Cluster two points when they express one broader mechanism and can be combined without losing a meaningful distinction. For example, “limited public transport” and “few safe cycling routes” may both belong under inadequate alternatives to driving if the target is reasons people continue using cars. But if the passage treats affordability and safety as separate barriers with different consequences, merging them into “transport problems” may be too vague.
Unsafe clustering
Do not merge ideas simply because they share vocabulary. “People lack information” and “people distrust information” are different. One concerns access, the other credibility. “Costs are high” and “benefits are unequal” are also different. Compression that erases the distinction loses content.
Paraphrase the concept, not each word
Word-by-word synonym substitution creates awkwardness and semantic drift. First understand the proposition. Then put the source aside and say the idea naturally. Finally compare the new sentence with the original for quantity, certainty, causality and scope.
Use grammatical transformation carefully
Changing noun to verb or clause to phrase can compress efficiently. “The inability of regulators to enforce standards” can become “regulators cannot enforce standards”. But transformation must preserve who acts, what is affected and the strength of the relationship.
Delete framing before meaning
Writers often use framing such as “It is important to note that”, “One reason for this is that”, or “This can be seen in the fact that”. Summary can often remove the frame and keep the proposition. Compression should first target verbal packaging, not substantive content.
Reduce repeated nouns and clauses
If several points share the same subject, sentence architecture can reduce repetition. Instead of repeating “governments” in three separate sentences, a coordinated structure may preserve clarity more efficiently. But never compress so tightly that relationships become ambiguous.
Concision is not telegraphese
Fragments such as “lack trust; high costs; poor access” may be efficient notes but weak final prose. The summary response still needs grammatical, readable English. Current Paper 2 language marks include the Summary and Application, so controlled expression matters.
Preserve quantity
Do not turn some communities into communities, or many users into all users. Quantifiers carry meaning even when they take little space.
Preserve certainty
A writer who says something may happen is not saying it will. A writer who says an outcome is likely is not merely saying it is possible. Modal strength must survive compression.
Preserve causality
Do not turn “contributes to” into “causes”. Do not turn “is associated with” into “results from”. Logical relationships are content.
Preserve conditions
If a benefit occurs only when a system is well regulated, the condition matters. Removing it may turn a limited claim into a universal one. Conditions are often precisely what make a writer’s argument nuanced.
Preserve comparison
“Cheaper than”, “more accessible than” and “less effective than” depend on relational meaning. A compressed sentence must retain the comparison if it is part of the point.
Avoid accidental evaluation
A summary should represent the passage, not add the candidate’s judgement. Words such as “unfairly”, “wisely” or “obviously” may introduce evaluation the source did not make. Keep the writer’s stance, not the candidate’s opinion.
Avoid causal repair that the writer never supplied
Students sometimes make a summary sentence more logical by inserting a reason that feels obvious. That is not compression; it is addition. Summary should preserve, not improve, the original argument.
Word economy through structure
Coordination, subordination and participial structures can compress repeated material, but clarity comes first. One precise twenty-word sentence is better than a twelve-word sentence whose pronoun reference or causal relationship is unclear.
Build compression ladders
Take a 60-word source explanation and reduce it to 40, then 25, then 15 words. At each stage ask which deletion changes meaning. The goal is to discover the boundary between efficient compression and semantic loss.
Build merge-or-separate drills
Give pairs of candidate points and ask whether they should be merged. Require justification. This trains one of the most important Summary judgements: when two expressions are variants of the same point and when they are genuinely distinct.
Build target/no-target drills
Present ten passage ideas and one summary target. The learner marks include, exclude or uncertain. Then discuss borderline cases. This isolates scope control from language production.
Build semantic-verification drills
Show an original point and three paraphrases: one accurate, one overstrong and one too vague. Ask which meaning feature changed. Repetition builds sensitivity to small semantic errors.
Time: selection and expression are separate bottlenecks
A student can be slow because they cannot identify points, because they overthink clustering, or because they struggle to write concise grammatical sentences. Time each stage separately. The repair depends on the bottleneck.
Do not over-polish one summary sentence
Leonie should use a sufficiency rule. If the sentence preserves the point and is grammatically clear, move. Perfectionism can steal time from later high-value tasks.
Summary Diagnostic Casebook | Sixty Failure Patterns and Repairs
Case 1: Student starts at first sentence
Repair identifies target before reading for points.
Case 2: Student summarises whole theme
Repair writes a target box and tests every point against it.
Case 3: Causes mixed with consequences
Repair labels information class before selection.
Case 4: Benefits mixed with examples
Repair distinguishes core point from illustration.
Case 5: Example counted as separate point
Repair asks whether removing the example leaves the point intact.
Case 6: Explanation counted twice
Repair compares underlying proposition rather than sentence count.
Case 7: Two distinct causes merged
Repair preserves different mechanisms.
Case 8: Same cause counted twice
Repair clusters paraphrased repetition.
Case 9: Student writes polished irrelevant point
Repair separates selection from expression.
Case 10: Student removes qualifier to save words
Repair preserves claim boundary before compression.
Case 11: Some becomes all
Repair checks quantifier.
Case 12: May becomes will
Repair checks modality.
Case 13: Contributes becomes causes
Repair checks causal relation.
Case 14: More effective becomes effective
Repair preserves comparison.
Case 15: Condition removed
Repair treats “only when”, “unless” and similar structures as content.
Case 16: Student inserts own evaluation
Repair removes words not present in writer stance.
Case 17: Student improves the writer’s logic
Repair returns to faithful representation.
Case 18: Word-by-word synonyms create awkward prose
Repair paraphrases conceptually.
Case 19: Paraphrase retains original syntax too closely
Repair reconstructs the idea in a different grammatical shape.
Case 20: Paraphrase changes subject
Repair checks who or what performs the action.
Case 21: Pronoun becomes ambiguous
Repair names the noun where necessary.
Case 22: Compression creates fragment
Repair restores a complete grammatical sentence.
Case 23: Compression creates run-on sentence
Repair separates relationships more clearly.
Case 24: Sentence is concise but too vague
Repair restores the distinct concept that was compressed away.
Case 25: Student writes notes instead of prose
Repair converts keywords into concise complete sentences.
Case 26: Student overuses semicolons to save words
Repair uses punctuation only when grammatical relationships are secure.
Case 27: Student repeats subject in every point
Repair combines structures where clarity remains high.
Case 28: Student combines too aggressively
Repair separates points when causal or comparative distinctions matter.
Case 29: Student leaves redundant framing
Repair removes “one reason is that” and similar packaging where unnecessary.
Case 30: Student deletes relationship word
Repair preserves because, although, therefore or equivalent logic.
Case 31: Student selects only explicit topic sentences
Repair searches for implied or developed points across paragraphs.
Case 32: Student ignores point hidden in example
Repair abstracts the underlying idea when the example carries distinct content.
Case 33: Student selects same idea from two paragraphs
Repair compares propositions globally before drafting.
Case 34: Student misses one point because wording is unfamiliar
Repair identifies function and relation, not familiar vocabulary.
Case 35: Student includes counterexample outside target
Repair asks whether it answers the task or merely enriches the passage.
Case 36: Student includes conclusion when target asks causes
Repair keeps information class aligned.
Case 37: Student omits causal explanation that is itself distinct
Repair distinguishes mere elaboration from a separate mechanism.
Case 38: Student cannot decide whether to merge
Repair uses the test: does the second idea add a new mechanism, condition or outcome?
Case 39: Student selects correctly but writes too slowly
Repair isolates compression and sentence production drills.
Case 40: Student writes quickly but selects poorly
Repair isolates target/no-target selection drills.
Case 41: Student rewrites one sentence repeatedly
Leonie uses sufficiency rule and moves on.
Case 42: Student never verifies semantics
Repair adds quick quantity-certainty-causality check.
Case 43: Student checks semantics but runs out of time
Repair verifies only high-risk transformations rather than every word.
Case 44: Student counts words obsessively
Repair focuses on concise architecture and uses checkpoints rather than constant recounting.
Case 45: Student forgets summary target midway
Repair keeps target box visible.
Case 46: Student follows passage order rigidly
Repair allows logical clustering when it preserves meaning.
Case 47: Student rearranges points and changes cause-effect order
Repair preserves logical dependencies even when reorganising.
Case 48: Student uses advanced vocabulary that distorts meaning
Repair chooses precise ordinary words.
Case 49: Student uses vague general nouns
Repair restores specific concepts such as access, trust, cost or regulation.
Case 50: Student repeats passage wording too closely
Repair reconstructs the idea after briefly looking away from source.
Case 51: Student paraphrases accurately but changes tone
Repair preserves evaluative force where it is part of content.
Case 52: Student omits a condition to reduce length
Repair treats conditions as non-negotiable when they bound the claim.
Case 53: Student includes two versions of one consequence
Repair clusters semantic overlap.
Case 54: Student merges direct and indirect consequences
Repair preserves causal distinction if target rewards both.
Case 55: Student runs out of summary time
Segment diagnosis identifies selection, clustering or writing bottleneck.
Case 56: Student spends too little time and misses points
Repair sets a minimum selection check before writing.
Case 57: Summary strong in familiar themes only
Repair uses unfamiliar-topic transfer; the target system should not depend on background knowledge.
Case 58: Summary content strong, language weak
Repair uses concise grammatical sentence drills.
Case 59: Summary language strong, content weak
Repair prioritises target selection and point hierarchy before style.
Case 60: Candidate panics because source seems dense
Repair returns to target, information class, candidate points and semantic verification. Density does not change the operating sequence.
Six Summary training cycles
Cycle 1: Target control. Train include/exclude decisions before writing.
Cycle 2: Information level. Distinguish point, explanation, example and consequence.
Cycle 3: Clustering. Practise merge-or-separate judgements across varied passages.
Cycle 4: Conceptual paraphrase. Restate ideas while preserving semantic properties.
Cycle 5: Compression. Build concise grammatical structures without flattening meaning.
Cycle 6: Timed integration. Complete the summary under a tested section budget, classify losses and repair the dominant stage.
Summary tutor audit
- Does the learner identify the summary target before selecting?
- Can core points be separated from explanation and example?
- Does the learner distinguish overlap from genuinely distinct ideas?
- Are clusters semantically safe?
- Can the learner paraphrase conceptually rather than synonym-by-synonym?
- Are quantity, certainty, cause, condition and comparison preserved?
- Does compression remove framing before substantive meaning?
- Are sentences concise but grammatical?
- Is the candidate adding personal evaluation?
- Can the learner identify whether selection or expression is the time bottleneck?
- Does the summary close in time to protect later Paper 2 tasks?
Summary examination-day card
- Write the target box.
- Collect only candidate points inside scope.
- Label core point, explanation and example.
- Remove duplicates.
- Merge only safe overlap.
- Paraphrase the concept, not each word.
- Compress while preserving logical relationships.
- Verify quantity, certainty, causality, conditions and comparison.
- Check grammar and completeness.
- Stop when the job is done and move to the next Paper 2 task.
The Summary standard
World-class Summary performance is not merely writing fewer words. It is deciding what the task actually needs, preserving the writer’s meaning, combining only what can safely be combined and expressing the result in concise grammatical English. The candidate should be able to cut heavily without changing quantity, cause, certainty, condition or relationship.
Properly taught kids shine a bright light into the future. In Summary, that light is disciplined reduction: know what matters, remove what does not, and leave the meaning intact.
Where to go next
- Return to H1 GP Paper 2 Examination Performance.
- Use the Passage 1, Connections and Application child guides under the same owner for their distinct systems.
- Use Concise Writing Under Exam Pressure for deeper universal compression mechanics.