How to Improve Vocabulary | Reporting and Attribution — Use Say, State, Claim, Argue, Suggest, Imply, Concede and According To Precisely

This specialist guide continues the How to Improve Vocabulary series by treating reporting and attribution as a precision problem rather than a synonym-list problem. The learner’s task is to identify the relation, choose the form whose grammar and meaning fit, test a near alternative, and explain what the substitution changes.

Reporting verbs do not merely replace ‘said’. They describe what kind of communicative act a source performed, and a careless verb can distort the source’s certainty, reasoning or stance. The complete vocabulary system owns the broader learning cycle. This article owns the narrower decisions named in its title and keeps neighbouring meanings behind explicit boundaries so that one useful page does not cannibalise another.

The surrounding strand already distinguishes scope and quantifiers, contrast and concession, conditions and exceptions, addition and emphasis, frequency and habit and approximation and precision. The same rule applies here: choose by meaning first, grammar second, register third, decoration last.

A learner has not mastered a form merely by recognising it. Control appears when the learner can reject a tempting alternative, repair an error, move the distinction into a new subject and predict what a careful reader will infer. The long workbook structure exists to create those repeated decisions.

Quick Read: Attribute the Source Without Rewriting Its Position

Say is relatively neutral and flexible for reporting words or propositions.

State often suggests an explicit, relatively formal assertion; it should not be used simply to make the source sound authoritative.

Claim reports an assertion and can carry sceptical colouring in some contexts, but it does not automatically mean the assertion is false.

Argue presents a position as supported by reasons or reasoning.

Suggest can report a proposal or a tentative indication; the intended sense must be clear.

Imply means the source communicates something indirectly; infer is what a reader or listener does from evidence.

Concede reports an acknowledgement that often runs against or limits the speaker’s main position.

According to attributes information or a position to a source without itself guaranteeing endorsement or truth.

Scope Boundary

This owner covers reporting verbs and source attribution. It does not decide whether a source is true, trustworthy or persuasive; those are evidence-evaluation questions. The task here is faithful representation.

When the problem belongs elsewhere, use the specialist owner rather than stretching this page. The Word-Choice Decision System is the general decision framework, while the linked precision articles handle their own logical relations.

1. Core Principles

1. Represent the source’s act

Choose a verb that matches what the source actually did.

2. Do not manufacture certainty

A tentative suggestion should not become a statement of fact in the reporting sentence.

3. Claim is not a verdict

Writers should not use claim as a covert accusation unless the context justifies that stance.

4. Argue needs reasoning

If the source merely mentions a view, argue may overstate the structure of the source’s case.

5. Suggest has multiple senses

It can propose an action or indicate a possibility.

6. Imply and infer differ

Sources imply; readers infer.

7. Concede is relational

It makes sense against a position, expectation or argument that the acknowledged point complicates.

8. According to is attribution

It identifies provenance rather than proving accuracy.

9. Quotation and paraphrase both need fidelity

Changing the reporting verb can change the source even when the paraphrased content is accurate.

10. Tense and grammar matter

Reporting structures constrain complements such as that-clauses, infinitives and noun phrases.

11. Citation does not remove judgement

Readers still need to know what is source material and what is the writer’s analysis.

12. Neutrality can be deliberate

Sometimes says or writes is more accurate than an interpretive reporting verb.

2. Core Forms

1. say

say has a defined job in this article. It is a relatively neutral reporting verb for words, propositions or communicated content. The nearest trap is tell, which normally takes a person or recipient in constructions where say does not. Ask: Am I reporting content without adding an interpretive stance?

Model: The report says that water use fell during the second quarter. Near miss: The report tells that water use fell. Repair: The report says that water use fell. Grammar as well as stance distinguishes reporting verbs.

2. state

state has a defined job in this article. It reports an explicit assertion, often in formal or official communication. The nearest trap is say, which is broader and often more neutral in register. Ask: Was the proposition explicitly presented rather than merely implied?

Model: The notice states that the building will close at 6 p.m. Near miss: The photograph states that the street was busy. Repair: The photograph suggests that the street was busy. Nonverbal evidence may suggest rather than state.

3. claim

claim has a defined job in this article. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. The nearest trap is state, which can sound more neutral about contestability. Ask: Am I accurately reporting an assertion without smuggling in a judgement that it is false?

Model: The manufacturer claims that the coating lasts for ten years. Near miss: The manufacturer lies that the coating lasts for ten years, without evidence of deception. Repair: The manufacturer claims that the coating lasts for ten years. Claim preserves room for verification without deciding the verdict.

4. argue

argue has a defined job in this article. It reports a position presented with reasons, evidence or an argumentative line. The nearest trap is say, which does not itself imply a reasoned case. Ask: Does the source actually support the position with reasoning?

Model: The author argues that the policy should be revised because its costs fall unevenly across households. Near miss: The author argues ‘hello’ at the beginning of the talk. Repair: The author says ‘hello’ at the beginning of the talk. Not every utterance is an argument.

5. suggest

suggest has a defined job in this article. It can report a proposal or indicate that evidence points tentatively toward an interpretation. The nearest trap is prove, which makes a much stronger evidential claim. Ask: Is the source proposing, or is the evidence indicating without establishing?

Model: The pattern suggests that demand increases during school holidays. Near miss: The pattern proves that holidays cause the increase. Repair: The pattern suggests an association between school holidays and higher demand. Keep evidential strength aligned with what the data support.

6. imply

imply has a defined job in this article. It reports meaning communicated indirectly rather than stated explicitly. The nearest trap is infer, which describes the reader’s reasoning from evidence. Ask: Who is doing the communicating and who is doing the interpretation?

Model: By calling the change ‘temporary’, the statement implies that the earlier arrangement may return. Near miss: The reader implies that the earlier arrangement may return. Repair: The reader infers that the earlier arrangement may return. Imply and infer have opposite participant roles.

7. concede

concede has a defined job in this article. It reports an acknowledgement that limits, complicates or runs against the source’s main position. The nearest trap is admit, which can carry different pragmatic colouring and is not always argumentative. Ask: What larger position is being qualified by this acknowledgement?

Model: The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them. Near miss: The author concedes that the proposal is excellent, with no opposing position or limitation. Repair: The author states that the proposal is excellent. Concession needs argumentative tension.

8. according to

according to has a defined job in this article. It attributes information, wording or a position to an identified source. The nearest trap is the writer’s own assertion, which should not be hidden behind vague attribution. Ask: Is the source clearly identified and is the attributed content faithful?

Model: According to the annual report, total enrolment increased by 6 percent. Near miss: According to experts, the method is best, when no experts are identified. Repair: According to the named review panel, the method produced the highest score in the reported trial. Attribution should be specific enough to inspect.

3. Pairwise Decisions

1. say versus state

say and state can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It reports an explicit assertion, often in formal or official communication. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

2. say versus claim

say and claim can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

3. say versus argue

say and argue can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It reports a position presented with reasons, evidence or an argumentative line. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

4. say versus suggest

say and suggest can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It can report a proposal or indicate that evidence points tentatively toward an interpretation. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

5. say versus imply

say and imply can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It reports meaning communicated indirectly rather than stated explicitly. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

6. say versus concede

say and concede can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

7. say versus according to

say and according to can appear near the same topic, but they should not be treated as prestige variants. It is a relatively neutral reporting verb for words, propositions or communicated content. It attributes information, wording or a position to an identified source. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

8. state versus claim

state and claim can appear near the same topic, but they should not be treated as prestige variants. It reports an explicit assertion, often in formal or official communication. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

9. state versus argue

state and argue can appear near the same topic, but they should not be treated as prestige variants. It reports an explicit assertion, often in formal or official communication. It reports a position presented with reasons, evidence or an argumentative line. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

10. state versus suggest

state and suggest can appear near the same topic, but they should not be treated as prestige variants. It reports an explicit assertion, often in formal or official communication. It can report a proposal or indicate that evidence points tentatively toward an interpretation. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

11. state versus imply

state and imply can appear near the same topic, but they should not be treated as prestige variants. It reports an explicit assertion, often in formal or official communication. It reports meaning communicated indirectly rather than stated explicitly. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

12. state versus concede

state and concede can appear near the same topic, but they should not be treated as prestige variants. It reports an explicit assertion, often in formal or official communication. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

13. state versus according to

state and according to can appear near the same topic, but they should not be treated as prestige variants. It reports an explicit assertion, often in formal or official communication. It attributes information, wording or a position to an identified source. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

14. claim versus argue

claim and argue can appear near the same topic, but they should not be treated as prestige variants. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. It reports a position presented with reasons, evidence or an argumentative line. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

15. claim versus suggest

claim and suggest can appear near the same topic, but they should not be treated as prestige variants. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. It can report a proposal or indicate that evidence points tentatively toward an interpretation. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

16. claim versus imply

claim and imply can appear near the same topic, but they should not be treated as prestige variants. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. It reports meaning communicated indirectly rather than stated explicitly. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

17. claim versus concede

claim and concede can appear near the same topic, but they should not be treated as prestige variants. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

18. claim versus according to

claim and according to can appear near the same topic, but they should not be treated as prestige variants. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. It attributes information, wording or a position to an identified source. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

19. argue versus suggest

argue and suggest can appear near the same topic, but they should not be treated as prestige variants. It reports a position presented with reasons, evidence or an argumentative line. It can report a proposal or indicate that evidence points tentatively toward an interpretation. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

20. argue versus imply

argue and imply can appear near the same topic, but they should not be treated as prestige variants. It reports a position presented with reasons, evidence or an argumentative line. It reports meaning communicated indirectly rather than stated explicitly. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

21. argue versus concede

argue and concede can appear near the same topic, but they should not be treated as prestige variants. It reports a position presented with reasons, evidence or an argumentative line. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

22. argue versus according to

argue and according to can appear near the same topic, but they should not be treated as prestige variants. It reports a position presented with reasons, evidence or an argumentative line. It attributes information, wording or a position to an identified source. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

23. suggest versus imply

suggest and imply can appear near the same topic, but they should not be treated as prestige variants. It can report a proposal or indicate that evidence points tentatively toward an interpretation. It reports meaning communicated indirectly rather than stated explicitly. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

24. suggest versus concede

suggest and concede can appear near the same topic, but they should not be treated as prestige variants. It can report a proposal or indicate that evidence points tentatively toward an interpretation. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Test the intended relation, the structure that follows, the strength of commitment and the likely reader inference. A substitution is useful only when the learner can say exactly what changed.

4. Precision Laboratory: Ninety Decisions

1. say: a data commentary

In a data commentary, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check logical direction. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

2. state: a comprehension answer

In a comprehension answer, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check attribution. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

3. claim: a formal email

In a formal email, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check route. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

4. argue: a research summary

In a research summary, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check claim strength. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

5. suggest: a literature response

In a literature response, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check sentence rhythm. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

6. imply: a comparison review

In a comparison review, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check obligation. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

7. concede: an argumentative paragraph

In an argumentative paragraph, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check register. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

8. according to: a reflective journal

In a reflective journal, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check ambiguity. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

9. say: a history essay

In a history essay, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check agency. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

10. state: a speech

In a speech, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check comparison set. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

11. claim: a public notice

In a public notice, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check reader expectation. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

12. argue: a parent-teacher note

In a parent-teacher note, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check cohesion. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

13. suggest: a problem-solution paragraph

In a problem-solution paragraph, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check instrument. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

14. imply: a school report

In a school report, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check punctuation. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

15. concede: a narrative scene

In a narrative scene, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check evidence. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

16. according to: a geography response

In a geography response, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check permission. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

17. say: a project evaluation

In a project evaluation, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check scope. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

18. state: a set of instructions

In a set of instructions, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check logical direction. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

19. claim: an oral answer

In an oral answer, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check attribution. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

20. argue: a science explanation

In a science explanation, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check route. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

21. suggest: a data commentary

In a data commentary, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check claim strength. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

22. imply: a comprehension answer

In a comprehension answer, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check sentence rhythm. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

23. concede: a formal email

In a formal email, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check obligation. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

24. according to: a research summary

In a research summary, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check register. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

25. say: a literature response

In a literature response, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check ambiguity. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

26. state: a comparison review

In a comparison review, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check agency. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

27. claim: an argumentative paragraph

In an argumentative paragraph, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check comparison set. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

28. argue: a reflective journal

In a reflective journal, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check reader expectation. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

29. suggest: a history essay

In a history essay, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check cohesion. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

30. imply: a speech

In a speech, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check instrument. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

31. concede: a public notice

In a public notice, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check punctuation. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

32. according to: a parent-teacher note

In a parent-teacher note, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check evidence. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

33. say: a problem-solution paragraph

In a problem-solution paragraph, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check permission. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

34. state: a school report

In a school report, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check scope. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

35. claim: a narrative scene

In a narrative scene, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check logical direction. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

36. argue: a geography response

In a geography response, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check attribution. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

37. suggest: a project evaluation

In a project evaluation, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check route. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

38. imply: a set of instructions

In a set of instructions, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check claim strength. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

39. concede: an oral answer

In an oral answer, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check sentence rhythm. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

40. according to: a science explanation

In a science explanation, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check obligation. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

41. say: a data commentary

In a data commentary, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check register. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

42. state: a comprehension answer

In a comprehension answer, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check ambiguity. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

43. claim: a formal email

In a formal email, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check agency. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

44. argue: a research summary

In a research summary, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check comparison set. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

45. suggest: a literature response

In a literature response, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check reader expectation. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

46. imply: a comparison review

In a comparison review, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check cohesion. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

47. concede: an argumentative paragraph

In an argumentative paragraph, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check instrument. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

48. according to: a reflective journal

In a reflective journal, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check punctuation. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

49. say: a history essay

In a history essay, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check evidence. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

50. state: a speech

In a speech, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check permission. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

51. claim: a public notice

In a public notice, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check scope. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

52. argue: a parent-teacher note

In a parent-teacher note, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check logical direction. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

53. suggest: a problem-solution paragraph

In a problem-solution paragraph, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check attribution. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

54. imply: a school report

In a school report, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check route. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

55. concede: a narrative scene

In a narrative scene, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check claim strength. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

56. according to: a geography response

In a geography response, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check sentence rhythm. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

57. say: a project evaluation

In a project evaluation, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check obligation. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

58. state: a set of instructions

In a set of instructions, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check register. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

59. claim: an oral answer

In an oral answer, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check ambiguity. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

60. argue: a science explanation

In a science explanation, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check agency. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

61. suggest: a data commentary

In a data commentary, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check comparison set. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

62. imply: a comprehension answer

In a comprehension answer, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check reader expectation. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

63. concede: a formal email

In a formal email, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check cohesion. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

64. according to: a research summary

In a research summary, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check instrument. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

65. say: a literature response

In a literature response, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check punctuation. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

66. state: a comparison review

In a comparison review, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check evidence. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

67. claim: an argumentative paragraph

In an argumentative paragraph, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check permission. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

68. argue: a reflective journal

In a reflective journal, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check scope. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

69. suggest: a history essay

In a history essay, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check logical direction. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

70. imply: a speech

In a speech, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check attribution. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

71. concede: a public notice

In a public notice, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check route. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

72. according to: a parent-teacher note

In a parent-teacher note, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check claim strength. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

73. say: a problem-solution paragraph

In a problem-solution paragraph, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check sentence rhythm. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

74. state: a school report

In a school report, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check obligation. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

75. claim: a narrative scene

In a narrative scene, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check register. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

76. argue: a geography response

In a geography response, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check ambiguity. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

77. suggest: a project evaluation

In a project evaluation, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check agency. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

78. imply: a set of instructions

In a set of instructions, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check comparison set. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

79. concede: an oral answer

In an oral answer, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check reader expectation. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

80. according to: a science explanation

In a science explanation, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check cohesion. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

81. say: a data commentary

In a data commentary, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check instrument. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

82. state: a comprehension answer

In a comprehension answer, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check punctuation. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

83. claim: a formal email

In a formal email, begin from meaning rather than surface form. Compare “The manufacturer claims that the coating lasts for ten years.” with “The manufacturer lies that the coating lasts for ten years, without evidence of deception.” and “The manufacturer claims that the coating lasts for ten years.”. Check evidence. Then write one new sentence on an unrelated subject and one version that avoids claim entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

84. argue: a research summary

In a research summary, begin from meaning rather than surface form. Compare “The author argues that the policy should be revised because its costs fall unevenly across households.” with “The author argues ‘hello’ at the beginning of the talk.” and “The author says ‘hello’ at the beginning of the talk.”. Check permission. Then write one new sentence on an unrelated subject and one version that avoids argue entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

85. suggest: a literature response

In a literature response, begin from meaning rather than surface form. Compare “The pattern suggests that demand increases during school holidays.” with “The pattern proves that holidays cause the increase.” and “The pattern suggests an association between school holidays and higher demand.”. Check scope. Then write one new sentence on an unrelated subject and one version that avoids suggest entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

86. imply: a comparison review

In a comparison review, begin from meaning rather than surface form. Compare “By calling the change ‘temporary’, the statement implies that the earlier arrangement may return.” with “The reader implies that the earlier arrangement may return.” and “The reader infers that the earlier arrangement may return.”. Check logical direction. Then write one new sentence on an unrelated subject and one version that avoids imply entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

87. concede: an argumentative paragraph

In an argumentative paragraph, begin from meaning rather than surface form. Compare “The author concedes that the proposal would increase short-term costs but maintains that the long-term savings justify them.” with “The author concedes that the proposal is excellent, with no opposing position or limitation.” and “The author states that the proposal is excellent.”. Check attribution. Then write one new sentence on an unrelated subject and one version that avoids concede entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

88. according to: a reflective journal

In a reflective journal, begin from meaning rather than surface form. Compare “According to the annual report, total enrolment increased by 6 percent.” with “According to experts, the method is best, when no experts are identified.” and “According to the named review panel, the method produced the highest score in the reported trial.”. Check route. Then write one new sentence on an unrelated subject and one version that avoids according to entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

89. say: a history essay

In a history essay, begin from meaning rather than surface form. Compare “The report says that water use fell during the second quarter.” with “The report tells that water use fell.” and “The report says that water use fell.”. Check claim strength. Then write one new sentence on an unrelated subject and one version that avoids say entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

90. state: a speech

In a speech, begin from meaning rather than surface form. Compare “The notice states that the building will close at 6 p.m.” with “The photograph states that the street was busy.” and “The photograph suggests that the street was busy.”. Check sentence rhythm. Then write one new sentence on an unrelated subject and one version that avoids state entirely. If the learner cannot explain why the alternatives differ, the decision is not yet stable.

5. Genre Transfer

1. academic essay

In a academic essay, say should be selected for its function. It is a relatively neutral reporting verb for words, propositions or communicated content. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

2. science report

In a science report, state should be selected for its function. It reports an explicit assertion, often in formal or official communication. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

3. history essay

In a history essay, claim should be selected for its function. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

4. geography response

In a geography response, argue should be selected for its function. It reports a position presented with reasons, evidence or an argumentative line. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

5. literature analysis

In a literature analysis, suggest should be selected for its function. It can report a proposal or indicate that evidence points tentatively toward an interpretation. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

6. comprehension answer

In a comprehension answer, imply should be selected for its function. It reports meaning communicated indirectly rather than stated explicitly. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

7. creative writing

In a creative writing, concede should be selected for its function. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

8. reflective writing

In a reflective writing, according to should be selected for its function. It attributes information, wording or a position to an identified source. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

9. formal email

In a formal email, say should be selected for its function. It is a relatively neutral reporting verb for words, propositions or communicated content. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

10. speech

In a speech, state should be selected for its function. It reports an explicit assertion, often in formal or official communication. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

11. data commentary

In a data commentary, claim should be selected for its function. It reports that a source asserts something, often where the assertion is open to evaluation or dispute. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

12. project evaluation

In a project evaluation, argue should be selected for its function. It reports a position presented with reasons, evidence or an argumentative line. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

13. research summary

In a research summary, suggest should be selected for its function. It can report a proposal or indicate that evidence points tentatively toward an interpretation. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

14. instructions

In a instructions, imply should be selected for its function. It reports meaning communicated indirectly rather than stated explicitly. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

15. public notice

In a public notice, concede should be selected for its function. It reports an acknowledgement that limits, complicates or runs against the source’s main position. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

16. oral presentation

In a oral presentation, according to should be selected for its function. It attributes information, wording or a position to an identified source. Write an accurate sentence, a deliberate near miss and a plain paraphrase. Then move the same distinction to a different topic. Transfer matters because vocabulary knowledge that works only in the original example remains fragile.

6. Error Clinic

1. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test register, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

2. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test punctuation, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

3. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test claim strength, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

4. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test reader expectation, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

5. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test logical direction, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

6. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test ambiguity, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

7. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test evidence, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

8. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test sentence rhythm, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

9. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test cohesion, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

10. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test attribution, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

11. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test agency, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

12. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test permission, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

13. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test obligation, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

14. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test instrument, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

15. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test route, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

16. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test comparison set, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

17. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test scope, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

18. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test register, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

19. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test punctuation, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

20. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test claim strength, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

21. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test reader expectation, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

22. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test logical direction, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

23. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test ambiguity, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

24. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test evidence, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

25. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test sentence rhythm, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

26. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test cohesion, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

27. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test attribution, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

28. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test agency, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

29. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test permission, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

30. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test obligation, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

31. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test instrument, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

32. Diagnose before replacing

When say appears doubtful, state the intended relation, identify the grammatical structure, test route, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

33. Diagnose before replacing

When claim appears doubtful, state the intended relation, identify the grammatical structure, test comparison set, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

34. Diagnose before replacing

When suggest appears doubtful, state the intended relation, identify the grammatical structure, test scope, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

35. Diagnose before replacing

When concede appears doubtful, state the intended relation, identify the grammatical structure, test register, compare the nearest alternative and then revise. A replacement that sounds formal may still be wrong. The repair succeeds only when meaning, grammar and the surrounding paragraph agree.

7. Retrieval Workbook

1. Build, replace, explain

Create a sentence in a speech that genuinely needs argue. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

2. Build, replace, explain

Create a sentence in a project evaluation that genuinely needs suggest. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

3. Build, replace, explain

Create a sentence in a research summary that genuinely needs imply. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

4. Build, replace, explain

Create a sentence in a public notice that genuinely needs concede. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

5. Build, replace, explain

Create a sentence in a set of instructions that genuinely needs according to. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

6. Build, replace, explain

Create a sentence in a literature response that genuinely needs say. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

7. Build, replace, explain

Create a sentence in a parent-teacher note that genuinely needs state. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

8. Build, replace, explain

Create a sentence in an oral answer that genuinely needs claim. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

9. Build, replace, explain

Create a sentence in a comparison review that genuinely needs argue. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

10. Build, replace, explain

Create a sentence in a problem-solution paragraph that genuinely needs suggest. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

11. Build, replace, explain

Create a sentence in a science explanation that genuinely needs imply. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

12. Build, replace, explain

Create a sentence in an argumentative paragraph that genuinely needs concede. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

13. Build, replace, explain

Create a sentence in a school report that genuinely needs according to. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

14. Build, replace, explain

Create a sentence in a data commentary that genuinely needs say. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

15. Build, replace, explain

Create a sentence in a reflective journal that genuinely needs state. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

16. Build, replace, explain

Create a sentence in a narrative scene that genuinely needs claim. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

17. Build, replace, explain

Create a sentence in a comprehension answer that genuinely needs argue. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

18. Build, replace, explain

Create a sentence in a history essay that genuinely needs suggest. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

19. Build, replace, explain

Create a sentence in a geography response that genuinely needs imply. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

20. Build, replace, explain

Create a sentence in a formal email that genuinely needs concede. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

21. Build, replace, explain

Create a sentence in a speech that genuinely needs according to. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

22. Build, replace, explain

Create a sentence in a project evaluation that genuinely needs say. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

23. Build, replace, explain

Create a sentence in a research summary that genuinely needs state. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

24. Build, replace, explain

Create a sentence in a public notice that genuinely needs claim. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

25. Build, replace, explain

Create a sentence in a set of instructions that genuinely needs argue. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

26. Build, replace, explain

Create a sentence in a literature response that genuinely needs suggest. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

27. Build, replace, explain

Create a sentence in a parent-teacher note that genuinely needs imply. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

28. Build, replace, explain

Create a sentence in an oral answer that genuinely needs concede. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

29. Build, replace, explain

Create a sentence in a comparison review that genuinely needs according to. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

30. Build, replace, explain

Create a sentence in a problem-solution paragraph that genuinely needs say. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

31. Build, replace, explain

Create a sentence in a science explanation that genuinely needs state. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

32. Build, replace, explain

Create a sentence in an argumentative paragraph that genuinely needs claim. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

33. Build, replace, explain

Create a sentence in a school report that genuinely needs argue. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

34. Build, replace, explain

Create a sentence in a data commentary that genuinely needs suggest. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

35. Build, replace, explain

Create a sentence in a reflective journal that genuinely needs imply. Replace it with the nearest alternative and label four things: intended relation, grammar, reader inference and register. Finish with a third example on a different subject. If the distinction disappears, strengthen the contrast rather than memorising another definition.

8. Teaching and Independent Practice

For independent practice, collect authentic sentences from reading and personal writing. Mark the decision point, predict the function, test a replacement and record one contrastive note. Revisit the note later by rebuilding the example from meaning. Retrieval should force a decision, not merely a recognition response.

For teaching, ask “What would change if we used the other form?” and “What is the reader licensed to infer?” Those questions expose partial understanding faster than asking for a dictionary definition alone.

9. Further Study

For general reference, use the British Council English grammar reference and Obligation, Permission and Advice when a source’s say/state/claim reports what people are required or allowed to do rather than what is merely possible.

10. Final Checklist

1. What relation am I expressing?

Final check: 1. What relation am I expressing? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

2. What structure follows?

Final check: 2. What structure follows? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

3. What does the nearest alternative change?

Final check: 3. What does the nearest alternative change? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

4. Is the register natural?

Final check: 4. Is the register natural? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

5. Does the form strengthen the claim?

Final check: 5. Does the form strengthen the claim? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

6. Could the reader infer more than I mean?

Final check: 6. Could the reader infer more than I mean? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

7. Is the scope clear?

Final check: 7. Is the scope clear? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

8. Can I paraphrase it plainly?

Final check: 8. Can I paraphrase it plainly? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

9. Can I transfer it to a new context?

Final check: 9. Can I transfer it to a new context? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

10. Is a simpler form already accurate?

Final check: 10. Is a simpler form already accurate? Identify the intended function, compare the nearest competing form, and give one concrete reason the chosen wording is better for this context. If the explanation is only stylistic, revise again.

Advanced Application: Transfer Problems

Application 1: say — a science explanation

In a science explanation, write one sentence that genuinely requires say. Replace it with the closest alternative and explain what changes in faithfulness to the source. Then transfer the same distinction to a comprehension answer. Finish by naming a context in which say would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 2: state — an oral response

Build a short example in an oral response around state. First state the intended function in plain English. Next write the sentence and test strength of attribution. Then alter one fact so that state is no longer the best choice, and repair the sentence. Rebuild the decision in a geography explanation to prove that the distinction is transferable.

Application 3: claim — a public notice

Imagine an editor challenges claim in a public notice. Defend the choice by pointing to assertion versus argument and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a classroom discussion. This trains the boundary of use, not just the successful case.

Application 4: argue — a reflective paragraph

Write two versions of the same idea in a reflective paragraph: one with argue, one with a simpler alternative. Compare the reader’s likely interpretation, especially proposal versus inference. Change the audience and rewrite for a report conclusion. Keep argue only if it communicates a distinction that the simpler wording would lose.

Application 5: suggest — a short speech

Create a deliberate near miss with suggest in a short speech. Diagnose the error using imply versus infer, repair it with the smallest possible change, and then create a fresh correct example in a history paragraph. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 6: imply — a data commentary

In a data commentary, write one sentence that genuinely requires imply. Replace it with the closest alternative and explain what changes in concession. Then transfer the same distinction to a project evaluation. Finish by naming a context in which imply would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 7: concede — a narrative scene

Build a short example in a narrative scene around concede. First state the intended function in plain English. Next write the sentence and test source specificity. Then alter one fact so that concede is no longer the best choice, and repair the sentence. Rebuild the decision in a research summary to prove that the distinction is transferable.

Application 8: according to — a literature response

Imagine an editor challenges according to in a literature response. Defend the choice by pointing to faithfulness to the source and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a technical note. This trains the boundary of use, not just the successful case.

Application 9: say — a comparison review

Write two versions of the same idea in a comparison review: one with say, one with a simpler alternative. Compare the reader’s likely interpretation, especially strength of attribution. Change the audience and rewrite for a problem-solution paragraph. Keep say only if it communicates a distinction that the simpler wording would lose.

Application 10: state — a revision note

Create a deliberate near miss with state in a revision note. Diagnose the error using assertion versus argument, repair it with the smallest possible change, and then create a fresh correct example in a policy note. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 11: claim — a formal email

In a formal email, write one sentence that genuinely requires claim. Replace it with the closest alternative and explain what changes in proposal versus inference. Then transfer the same distinction to a parent update. Finish by naming a context in which claim would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 12: argue — a comprehension answer

Build a short example in a comprehension answer around argue. First state the intended function in plain English. Next write the sentence and test imply versus infer. Then alter one fact so that argue is no longer the best choice, and repair the sentence. Rebuild the decision in a set of instructions to prove that the distinction is transferable.

Application 13: suggest — a geography explanation

Imagine an editor challenges suggest in a geography explanation. Defend the choice by pointing to concession and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in an exam response. This trains the boundary of use, not just the successful case.

Application 14: imply — a classroom discussion

Write two versions of the same idea in a classroom discussion: one with imply, one with a simpler alternative. Compare the reader’s likely interpretation, especially source specificity. Change the audience and rewrite for a science explanation. Keep imply only if it communicates a distinction that the simpler wording would lose.

Application 15: concede — a report conclusion

Create a deliberate near miss with concede in a report conclusion. Diagnose the error using faithfulness to the source, repair it with the smallest possible change, and then create a fresh correct example in an oral response. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 16: according to — a history paragraph

In a history paragraph, write one sentence that genuinely requires according to. Replace it with the closest alternative and explain what changes in strength of attribution. Then transfer the same distinction to a public notice. Finish by naming a context in which according to would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 17: say — a project evaluation

Build a short example in a project evaluation around say. First state the intended function in plain English. Next write the sentence and test assertion versus argument. Then alter one fact so that say is no longer the best choice, and repair the sentence. Rebuild the decision in a reflective paragraph to prove that the distinction is transferable.

Application 18: state — a research summary

Imagine an editor challenges state in a research summary. Defend the choice by pointing to proposal versus inference and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a short speech. This trains the boundary of use, not just the successful case.

Application 19: claim — a technical note

Write two versions of the same idea in a technical note: one with claim, one with a simpler alternative. Compare the reader’s likely interpretation, especially imply versus infer. Change the audience and rewrite for a data commentary. Keep claim only if it communicates a distinction that the simpler wording would lose.

Application 20: argue — a problem-solution paragraph

Create a deliberate near miss with argue in a problem-solution paragraph. Diagnose the error using concession, repair it with the smallest possible change, and then create a fresh correct example in a narrative scene. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 21: suggest — a policy note

In a policy note, write one sentence that genuinely requires suggest. Replace it with the closest alternative and explain what changes in source specificity. Then transfer the same distinction to a literature response. Finish by naming a context in which suggest would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 22: imply — a parent update

Build a short example in a parent update around imply. First state the intended function in plain English. Next write the sentence and test faithfulness to the source. Then alter one fact so that imply is no longer the best choice, and repair the sentence. Rebuild the decision in a comparison review to prove that the distinction is transferable.

Application 23: concede — a set of instructions

Imagine an editor challenges concede in a set of instructions. Defend the choice by pointing to strength of attribution and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a revision note. This trains the boundary of use, not just the successful case.

Application 24: according to — an exam response

Write two versions of the same idea in an exam response: one with according to, one with a simpler alternative. Compare the reader’s likely interpretation, especially assertion versus argument. Change the audience and rewrite for a formal email. Keep according to only if it communicates a distinction that the simpler wording would lose.

Application 25: say — a science explanation

Create a deliberate near miss with say in a science explanation. Diagnose the error using proposal versus inference, repair it with the smallest possible change, and then create a fresh correct example in a comprehension answer. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 26: state — an oral response

In an oral response, write one sentence that genuinely requires state. Replace it with the closest alternative and explain what changes in imply versus infer. Then transfer the same distinction to a geography explanation. Finish by naming a context in which state would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 27: claim — a public notice

Build a short example in a public notice around claim. First state the intended function in plain English. Next write the sentence and test concession. Then alter one fact so that claim is no longer the best choice, and repair the sentence. Rebuild the decision in a classroom discussion to prove that the distinction is transferable.

Application 28: argue — a reflective paragraph

Imagine an editor challenges argue in a reflective paragraph. Defend the choice by pointing to source specificity and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a report conclusion. This trains the boundary of use, not just the successful case.

Application 29: suggest — a short speech

Write two versions of the same idea in a short speech: one with suggest, one with a simpler alternative. Compare the reader’s likely interpretation, especially faithfulness to the source. Change the audience and rewrite for a history paragraph. Keep suggest only if it communicates a distinction that the simpler wording would lose.

Application 30: imply — a data commentary

Create a deliberate near miss with imply in a data commentary. Diagnose the error using strength of attribution, repair it with the smallest possible change, and then create a fresh correct example in a project evaluation. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 31: concede — a narrative scene

In a narrative scene, write one sentence that genuinely requires concede. Replace it with the closest alternative and explain what changes in assertion versus argument. Then transfer the same distinction to a research summary. Finish by naming a context in which concede would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 32: according to — a literature response

Build a short example in a literature response around according to. First state the intended function in plain English. Next write the sentence and test proposal versus inference. Then alter one fact so that according to is no longer the best choice, and repair the sentence. Rebuild the decision in a technical note to prove that the distinction is transferable.

Application 33: say — a comparison review

Imagine an editor challenges say in a comparison review. Defend the choice by pointing to imply versus infer and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a problem-solution paragraph. This trains the boundary of use, not just the successful case.

Application 34: state — a revision note

Write two versions of the same idea in a revision note: one with state, one with a simpler alternative. Compare the reader’s likely interpretation, especially concession. Change the audience and rewrite for a policy note. Keep state only if it communicates a distinction that the simpler wording would lose.

Application 35: claim — a formal email

Create a deliberate near miss with claim in a formal email. Diagnose the error using source specificity, repair it with the smallest possible change, and then create a fresh correct example in a parent update. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 36: argue — a comprehension answer

In a comprehension answer, write one sentence that genuinely requires argue. Replace it with the closest alternative and explain what changes in faithfulness to the source. Then transfer the same distinction to a set of instructions. Finish by naming a context in which argue would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 37: suggest — a geography explanation

Build a short example in a geography explanation around suggest. First state the intended function in plain English. Next write the sentence and test strength of attribution. Then alter one fact so that suggest is no longer the best choice, and repair the sentence. Rebuild the decision in an exam response to prove that the distinction is transferable.

Application 38: imply — a classroom discussion

Imagine an editor challenges imply in a classroom discussion. Defend the choice by pointing to assertion versus argument and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a science explanation. This trains the boundary of use, not just the successful case.

Application 39: concede — a report conclusion

Write two versions of the same idea in a report conclusion: one with concede, one with a simpler alternative. Compare the reader’s likely interpretation, especially proposal versus inference. Change the audience and rewrite for an oral response. Keep concede only if it communicates a distinction that the simpler wording would lose.

Application 40: according to — a history paragraph

Create a deliberate near miss with according to in a history paragraph. Diagnose the error using imply versus infer, repair it with the smallest possible change, and then create a fresh correct example in a public notice. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 41: say — a project evaluation

In a project evaluation, write one sentence that genuinely requires say. Replace it with the closest alternative and explain what changes in concession. Then transfer the same distinction to a reflective paragraph. Finish by naming a context in which say would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 42: state — a research summary

Build a short example in a research summary around state. First state the intended function in plain English. Next write the sentence and test source specificity. Then alter one fact so that state is no longer the best choice, and repair the sentence. Rebuild the decision in a short speech to prove that the distinction is transferable.

Application 43: claim — a technical note

Imagine an editor challenges claim in a technical note. Defend the choice by pointing to faithfulness to the source and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a data commentary. This trains the boundary of use, not just the successful case.

Application 44: argue — a problem-solution paragraph

Write two versions of the same idea in a problem-solution paragraph: one with argue, one with a simpler alternative. Compare the reader’s likely interpretation, especially strength of attribution. Change the audience and rewrite for a narrative scene. Keep argue only if it communicates a distinction that the simpler wording would lose.

Application 45: suggest — a policy note

Create a deliberate near miss with suggest in a policy note. Diagnose the error using assertion versus argument, repair it with the smallest possible change, and then create a fresh correct example in a literature response. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 46: imply — a parent update

In a parent update, write one sentence that genuinely requires imply. Replace it with the closest alternative and explain what changes in proposal versus inference. Then transfer the same distinction to a comparison review. Finish by naming a context in which imply would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 47: concede — a set of instructions

Build a short example in a set of instructions around concede. First state the intended function in plain English. Next write the sentence and test imply versus infer. Then alter one fact so that concede is no longer the best choice, and repair the sentence. Rebuild the decision in a revision note to prove that the distinction is transferable.

Application 48: according to — an exam response

Imagine an editor challenges according to in an exam response. Defend the choice by pointing to concession and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a formal email. This trains the boundary of use, not just the successful case.

Application 49: say — a science explanation

Write two versions of the same idea in a science explanation: one with say, one with a simpler alternative. Compare the reader’s likely interpretation, especially source specificity. Change the audience and rewrite for a comprehension answer. Keep say only if it communicates a distinction that the simpler wording would lose.

Application 50: state — an oral response

Create a deliberate near miss with state in an oral response. Diagnose the error using faithfulness to the source, repair it with the smallest possible change, and then create a fresh correct example in a geography explanation. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 51: claim — a public notice

In a public notice, write one sentence that genuinely requires claim. Replace it with the closest alternative and explain what changes in strength of attribution. Then transfer the same distinction to a classroom discussion. Finish by naming a context in which claim would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 52: argue — a reflective paragraph

Build a short example in a reflective paragraph around argue. First state the intended function in plain English. Next write the sentence and test assertion versus argument. Then alter one fact so that argue is no longer the best choice, and repair the sentence. Rebuild the decision in a report conclusion to prove that the distinction is transferable.

Application 53: suggest — a short speech

Imagine an editor challenges suggest in a short speech. Defend the choice by pointing to proposal versus inference and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a history paragraph. This trains the boundary of use, not just the successful case.

Application 54: imply — a data commentary

Write two versions of the same idea in a data commentary: one with imply, one with a simpler alternative. Compare the reader’s likely interpretation, especially imply versus infer. Change the audience and rewrite for a project evaluation. Keep imply only if it communicates a distinction that the simpler wording would lose.

Application 55: concede — a narrative scene

Create a deliberate near miss with concede in a narrative scene. Diagnose the error using concession, repair it with the smallest possible change, and then create a fresh correct example in a research summary. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 56: according to — a literature response

In a literature response, write one sentence that genuinely requires according to. Replace it with the closest alternative and explain what changes in source specificity. Then transfer the same distinction to a technical note. Finish by naming a context in which according to would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 57: say — a comparison review

Build a short example in a comparison review around say. First state the intended function in plain English. Next write the sentence and test faithfulness to the source. Then alter one fact so that say is no longer the best choice, and repair the sentence. Rebuild the decision in a problem-solution paragraph to prove that the distinction is transferable.

Application 58: state — a revision note

Imagine an editor challenges state in a revision note. Defend the choice by pointing to strength of attribution and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a policy note. This trains the boundary of use, not just the successful case.

Application 59: claim — a formal email

Write two versions of the same idea in a formal email: one with claim, one with a simpler alternative. Compare the reader’s likely interpretation, especially assertion versus argument. Change the audience and rewrite for a parent update. Keep claim only if it communicates a distinction that the simpler wording would lose.

Application 60: argue — a comprehension answer

Create a deliberate near miss with argue in a comprehension answer. Diagnose the error using proposal versus inference, repair it with the smallest possible change, and then create a fresh correct example in a set of instructions. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 61: suggest — a geography explanation

In a geography explanation, write one sentence that genuinely requires suggest. Replace it with the closest alternative and explain what changes in imply versus infer. Then transfer the same distinction to an exam response. Finish by naming a context in which suggest would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 62: imply — a classroom discussion

Build a short example in a classroom discussion around imply. First state the intended function in plain English. Next write the sentence and test concession. Then alter one fact so that imply is no longer the best choice, and repair the sentence. Rebuild the decision in a science explanation to prove that the distinction is transferable.

Application 63: concede — a report conclusion

Imagine an editor challenges concede in a report conclusion. Defend the choice by pointing to source specificity and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in an oral response. This trains the boundary of use, not just the successful case.

Application 64: according to — a history paragraph

Write two versions of the same idea in a history paragraph: one with according to, one with a simpler alternative. Compare the reader’s likely interpretation, especially faithfulness to the source. Change the audience and rewrite for a public notice. Keep according to only if it communicates a distinction that the simpler wording would lose.

Application 65: say — a project evaluation

Create a deliberate near miss with say in a project evaluation. Diagnose the error using strength of attribution, repair it with the smallest possible change, and then create a fresh correct example in a reflective paragraph. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 66: state — a research summary

In a research summary, write one sentence that genuinely requires state. Replace it with the closest alternative and explain what changes in assertion versus argument. Then transfer the same distinction to a short speech. Finish by naming a context in which state would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 67: claim — a technical note

Build a short example in a technical note around claim. First state the intended function in plain English. Next write the sentence and test proposal versus inference. Then alter one fact so that claim is no longer the best choice, and repair the sentence. Rebuild the decision in a data commentary to prove that the distinction is transferable.

Application 68: argue — a problem-solution paragraph

Imagine an editor challenges argue in a problem-solution paragraph. Defend the choice by pointing to imply versus infer and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a narrative scene. This trains the boundary of use, not just the successful case.

Application 69: suggest — a policy note

Write two versions of the same idea in a policy note: one with suggest, one with a simpler alternative. Compare the reader’s likely interpretation, especially concession. Change the audience and rewrite for a literature response. Keep suggest only if it communicates a distinction that the simpler wording would lose.

Application 70: imply — a parent update

Create a deliberate near miss with imply in a parent update. Diagnose the error using source specificity, repair it with the smallest possible change, and then create a fresh correct example in a comparison review. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 71: concede — a set of instructions

In a set of instructions, write one sentence that genuinely requires concede. Replace it with the closest alternative and explain what changes in faithfulness to the source. Then transfer the same distinction to a revision note. Finish by naming a context in which concede would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 72: according to — an exam response

Build a short example in an exam response around according to. First state the intended function in plain English. Next write the sentence and test strength of attribution. Then alter one fact so that according to is no longer the best choice, and repair the sentence. Rebuild the decision in a formal email to prove that the distinction is transferable.

Application 73: say — a science explanation

Imagine an editor challenges say in a science explanation. Defend the choice by pointing to assertion versus argument and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a comprehension answer. This trains the boundary of use, not just the successful case.

Application 74: state — an oral response

Write two versions of the same idea in an oral response: one with state, one with a simpler alternative. Compare the reader’s likely interpretation, especially proposal versus inference. Change the audience and rewrite for a geography explanation. Keep state only if it communicates a distinction that the simpler wording would lose.

Application 75: claim — a public notice

Create a deliberate near miss with claim in a public notice. Diagnose the error using imply versus infer, repair it with the smallest possible change, and then create a fresh correct example in a classroom discussion. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Application 76: argue — a reflective paragraph

In a reflective paragraph, write one sentence that genuinely requires argue. Replace it with the closest alternative and explain what changes in concession. Then transfer the same distinction to a report conclusion. Finish by naming a context in which argue would mislead a careful reader. The explanation should identify meaning and structure, not merely say that one version sounds better.

Application 77: suggest — a short speech

Build a short example in a short speech around suggest. First state the intended function in plain English. Next write the sentence and test source specificity. Then alter one fact so that suggest is no longer the best choice, and repair the sentence. Rebuild the decision in a history paragraph to prove that the distinction is transferable.

Application 78: imply — a data commentary

Imagine an editor challenges imply in a data commentary. Defend the choice by pointing to faithfulness to the source and to the grammar of the surrounding sentence. Then write the editor’s strongest counterexample and revise accordingly. Repeat the exercise in a project evaluation. This trains the boundary of use, not just the successful case.

Application 79: concede — a narrative scene

Write two versions of the same idea in a narrative scene: one with concede, one with a simpler alternative. Compare the reader’s likely interpretation, especially strength of attribution. Change the audience and rewrite for a research summary. Keep concede only if it communicates a distinction that the simpler wording would lose.

Application 80: according to — a literature response

Create a deliberate near miss with according to in a literature response. Diagnose the error using assertion versus argument, repair it with the smallest possible change, and then create a fresh correct example in a technical note. A strong repair explains why the original failed and why the new form fits the intended reporting and attribution relation.

Closing Principle

Reporting precision protects both writer and source. Choose the verb that matches what the source actually did, preserve its level of certainty, and keep attribution separate from your own evaluation.