Opinion essay (300 words) at C1

At B2 you wrote a five-paragraph opinion essay of around 280 words. At C1 the form is the same on the outside, but everything inside is sharper. The thesis is more specific and more defensible. The body paragraphs no longer just give a reason — they make a claim, ground it in concrete evidence, and tie it explicitly back to the thesis. The counter-paragraph stops being a token some people might say and becomes a fair-minded representation of the strongest opposing view. The conclusion stops restating and starts broadening — pointing outward to the stakes.

A C1 reader (IELTS examiner, TOEFL rater, university tutor, op-ed editor) is looking for three things that distinguish an upper-band essay from a competent one: a thesis that takes a real risk, cohesion across paragraphs (not just within), and register control — confident but hedged, opinionated but not strident. This lesson walks through the C1 opinion essay end to end with a 320-word annotated model.

Structure — five paragraphs, sharper edges

1. Introduction (~50w) — hook, context, thesis. The hook does real work; no In this essay I will discuss openings.
2. Body 1 — strongest reason (~75w) — claim, evidence, analysis tied back to thesis.
3. Body 2 — second reason (~75w) — same shape, different angle. The two reasons should not overlap.
4. Body 3 — counter + refutation (~70w) — represent the strongest opposing view fairly, then refute or concede-and-limit.
5. Conclusion (~30w) — restate position in fresh language, point to broader stakes, no new arguments.

Word target: 300 minimum, 350 ceiling. Going under signals underdevelopment; going far over signals lack of discipline.

Step-by-step craft

1. Find a thesis that takes a real risk

A C1 thesis is specific, debatable, and single-sentence. The B2 thesis Universities should ban AI for writing gets sharpened to Universities should ban generative AI for first-year writing assignments because the cognitive work it bypasses is precisely the work the assignment exists to develop. The longer version names the audience (first-year), names the scope (writing assignments), and names the mechanism (cognitive work bypassed) — which gives the body paragraphs three places to land.

A test: can someone smart disagree with your thesis without sounding stupid? If not, the thesis is too obvious. Honesty matters is not a thesis; Honesty about layoffs costs companies in the short term but compounds trust over years is.

2. Hook that earns the next sentence

The hook is the first sentence and it has one job: make the reader want the second sentence. Three reliable hook types at C1:

• The surprising fact: More American teenagers now meet their close friends in voice chat than in person.
• The reframing: We argue about screen time as though screens were the only thing competing for a child’s attention.
• The pointed observation: Every company swears it values feedback, and every employee learns within a month not to give any.

Avoid the rhetorical question hook — overused, and at C1 it reads as B2.

3. Body paragraphs that argue, not just assert

Each body paragraph at C1 has four moves: claim, evidence, analysis, link. The B2 version often had three (claim, example, link). The new ingredient is analysis — explaining why the evidence supports the claim, not just citing it. A 2023 Stanford study found that students who drafted essays with AI assistance scored lower on subsequent unaided assessments is evidence; This suggests the offloading is not just convenient but actively interrupts skill formation is analysis.

4. Counter-argument done seriously

The C1 counter-paragraph is where most B2 essays still leak. The bad version: Some people might say AI is useful, but they are wrong. The C1 version names the strongest opposing argument, in its strongest form, then refutes it on its own terms or concedes part of it: Critics rightly point out that banning AI ignores its real value as a brainstorming partner. The objection has force; what it overlooks, however, is the difference between AI as a thinking aid for experienced writers and AI as a substitute for the cognitive work novices have not yet done.

5. Conclusion that broadens

A C1 conclusion does not repeat the thesis. It restates the position in fresh language and then points outward: to a related debate, a broader principle, a stake the reader hadn’t fully felt. The deeper question is what universities are for: credentialing or cognition. The answer should determine the policy.

6. Cohesion across paragraphs

Cohesion at C1 is built with explicit linking between paragraphs, not just within them. Open Body 2 with a reference back to Body 1: Beyond the cognitive cost, there is a second concern…. Open the counter-paragraph with explicit signaling: Critics of this position argue…. These bridges are what separate an essay from a list.

Full model essay — 320 words, annotated

Should universities ban generative AI for first-year writing?

Generative AI tools have arrived in higher education faster than the policies meant to govern them. Some universities have embraced them as teaching aids; others have banned them outright; most are improvising. The question is not whether AI is useful — it plainly is — but whether universities should permit it for the specific work where it does the most harm. I argue that universities should ban generative AI for first-year writing assignments, because the cognitive work it bypasses is precisely the work the assignment exists to develop.

The first reason is developmental. Writing is not output; it is the externalization of thought. When a student drafts a paragraph, they are not just producing prose but discovering what they think. A 2023 Stanford study found that students who drafted with AI assistance scored measurably lower on subsequent unaided assessments. This is not a story about cheating; it is a story about a skill that fails to form when the relevant cognitive struggle is offloaded.

A second reason is assessment integrity. A grade on a writing assignment is supposed to reflect a student’s competence; when AI produces the work, the grade reflects the tool. The whole feedback loop on which university teaching depends — diagnose, intervene, retest — breaks down silently. Faculty stop seeing the students they are supposed to teach.

Critics rightly note that banning AI ignores its real value as a brainstorming partner. The objection has force; what it overlooks is the difference between AI as a thinking aid for experienced writers and AI as a substitute for the cognitive work novices have not yet done. The remedy is not a blanket embrace but graduated access.

The deeper question is what universities are for. If they exist to credential, AI is irrelevant. If they exist to cultivate thinking, the policy that follows is not difficult to determine.

Annotations: thesis is the last sentence of the intro and names mechanism, scope, and audience. Body 1 cites evidence and provides analysis. Body 2 introduces a structurally different reason (developmental vs institutional). Counter concedes part of the opposing view, then limits it. Conclusion broadens to first principles.

Common pitfalls

• B1 thesis at C1 length: long word count, vague claim. AI is bad for students padded to 350 words is still B1.
• Restated conclusion: ending with the thesis in slightly different words. C1 conclusions broaden.
• Strawman counter: representing the opposing view as obviously wrong. Real C1 readers respect a serious counter.
• Stranded examples: dropping a fact and not explaining why it supports the claim. Always finish with analysis.
• Connector salad: Firstly, Moreover, Furthermore, In addition, Finally — every paragraph opens with a stock linker. Vary, or build cohesion through reference instead.

Connectors and phrases bank

Thesis framing: I argue that…, This essay contends that…, The position taken here is that…, On balance, the evidence supports…

Building the case: The first reason is…, A second consideration is…, Beyond the [previous angle], there is a deeper concern…, Equally significant is the fact that…

Evidence and analysis: A 2023 [study/report] found that…, Recent data suggests that…, This indicates that…, The implication is that…, What this means in practice is…

Counter-argument: Critics rightly point out that…, A common objection is that…, Proponents of the opposing view argue…, The objection has force; what it overlooks, however, is…

Conclusion that broadens: The deeper question is…, The underlying issue is one of…, What is ultimately at stake is…

Cohesion devices at the paragraph level

Cohesion at C1 operates at two scales: within paragraphs and across them.

Within a paragraph, cohesion comes from:

• Pronoun reference: The study found X. It suggests Y. The it anchors the second sentence to the first.
• Synonym chains: introduce a term, then refer to it with varied synonyms. AI use → the technology → the tool → algorithmic drafting.
• Old-information-first: each sentence opens with information the previous sentence established, then adds new information. The given-new pattern keeps the reader oriented.

Across paragraphs, cohesion comes from:

• Forward-pointing closings: end body 1 with a sentence that names what body 2 will address.
• Backward-pointing openings: open body 2 with explicit reference to body 1. Beyond the developmental cost, there is a second concern…
• Thematic threading: a metaphor or analogy introduced in the intro can return in the conclusion as a structural device.

The B2 essay relied on linkers (Firstly, Furthermore, In conclusion) for cross-paragraph cohesion. The C1 essay shifts much of that work into reference and threading, freeing up linker slots for variety.

Hedged confidence — the C1 register move

A common C1 register failure is over-hedging — softening every claim until the essay sounds tentative. The opposite failure is under-hedging — stating contested causal claims as fact.

The C1 balance:

• The thesis is unhedged. Universities should ban generative AI for first-year writing. Not Universities might possibly consider…
• Causal claims are hedged. AI use appears to be associated with measurable declines in performance. Not AI causes students to lose ability.
• Empirical generalizations are scoped. A 2023 Stanford study found that… names the source, the year, the limit.
• Counter-acknowledgments are confident in their limits. The objection has force; what it overlooks, however, is…

The pattern: confident where you have grounds, hedged where the evidence is partial, scoped where you are generalizing.

When the prompt asks for to what extent

IELTS Task 2 prompts often ask to what extent do you agree? — a phrasing that explicitly invites a hedged position. A C1 response recognizes the invitation:

• Fully agree: defend with three angles, address the strongest counter.
• Partially agree: name the part you agree with and the part you do not; defend each.
• Disagree: same structure as agree, with the position reversed.

The partial-agreement essay is structurally harder because it requires you to hold two positions in coherent tension. At C1 you should be able to write it. The thesis form: While [opposing claim] holds in [specific scope], it does not extend to [broader claim], where the evidence in fact supports [your position].

Common Russian-speaker writing mistakes

1. Over-formal openings — In the modern world, the question of artificial intelligence in education is very actual. This calques several Russian habits: в современном мире (cliché opening), очень актуальный (false friend — actual means real/existing, not current/relevant). A C1 opening is concrete and specific, not abstract and global.

2. Thesis stated as a question — The question of whether universities should ban AI is a very interesting one. Russian academic writing tolerates this; English C1 does not. The thesis must be a claim, not a meta-comment.

3. Excessive that-clauses with abstract subjects — It is necessary to note that the fact that AI is used widely makes it relevant. English C1 prefers concrete subjects and verbs: AI’s wide adoption makes the question urgent.

4. Calque on в связи с этим / в этой связи — translated as In connection with this or In this connection. The natural English C1 connector is Accordingly, On these grounds, or In light of this.

5. Conclusion phrases that signal exam-essay machinery — To sum up, in conclusion, as I have mentioned above, taking everything into consideration. These were acceptable at B1-B2; at C1 they signal a formula rather than thought. Replace with substantive broadening.

6. Over-stating with intensifiers — AI is absolutely terrible for students. It completely destroys their writing skills. It is totally unacceptable. Russian rhetorical style tolerates strong claims; English C1 academic register prefers hedged confidence: measurably worse, significantly impairs, difficult to defend.

7. Missing hedges in causal claims — AI causes students to lose writing ability is too strong for any single study. AI use appears to be linked to measurable declines in unaided performance is C1-defensible.

Summary

• The C1 opinion essay is 300-350 words: sharper thesis, four-move body paragraphs (claim, evidence, analysis, link), fair counter, broadening conclusion.
• The thesis must be specific, debatable, and one sentence — and risky enough that a smart person could disagree.
• Counter-argument done seriously: name the strongest opposing view, concede or limit, do not strawman.
• Conclusion does not restate — it broadens to a stake, principle, or wider debate.
• Cohesion is built across paragraphs through explicit reference, not just within paragraphs through linkers.
• Russian speakers should especially watch for over-formal openings, calque connectors, and unhedged absolutes.

High-band evidence integration

C1-band essays integrate evidence with three levels of credit-earning specificity:

The high-band form does three things at once: names the source (Stanford), scopes the claim (specific population, specific intervention, specific outcome), and quantifies (n=4,500, 18% lower, twelve weeks). A skeptical reader can engage with this; a believing reader has grounds; a grader scores it as band 7+.

A common B2-to-C1 transition error: students learn to cite but cite generically. According to a study signals citation without earning credit. According to a 2023 Stanford study (Gerlich et al.) earns the credit by naming what is actually being cited.

In exam conditions where you cannot verify a citation, the convention is to use plausible, scoped citations that you have learned during preparation. Examiners do not fact-check citations — they reward the rhetorical form of citation. A 2023 OECD report on remote-work productivity is rewarded even if the specific report is misremembered.

The four C1 essay-type templates

Different prompts call for different essay shapes. Knowing which template the prompt requires saves you the time of inventing structure under exam pressure.

Opinion / agree-disagree template

Prompt form: To what extent do you agree that X?

Structure:

• Intro: hook, context, thesis (your position with three angles).
• Body 1: strongest reason for your position.
• Body 2: second reason, structurally different from body 1.
• Body 3: counter-argument, conceded or limited.
• Conclusion: position restated in fresh language, broadened to wider stake.

Discussion / both-views template

Prompt form: Discuss both views and give your opinion.

Structure:

• Intro: hook, context, thesis (which view you ultimately support and why).
• Body 1: strongest case for view A, fairly represented.
• Body 2: strongest case for view B, fairly represented.
• Body 3: your synthesis — why one view wins, or under what conditions each holds.
• Conclusion: your position broadened.

Problem-solution template

Prompt form: What are the causes of X, and what can be done?

Structure:

• Intro: hook, context, thesis (your diagnosis of the principal cause and the most promising remedy).
• Body 1: principal cause, with mechanism made visible.
• Body 2: most promising remedy, with mechanism made visible.
• Body 3: limitations of the remedy or remaining gaps.
• Conclusion: the deeper question or principle the problem turns on.

Advantages-disadvantages template

Prompt form: Do the advantages outweigh the disadvantages?

Structure:

• Intro: hook, context, thesis (your verdict on the balance).
• Body 1: the strongest advantage.
• Body 2: the strongest disadvantage.
• Body 3: the weighing — why the balance tilts as it does in your judgment.
• Conclusion: implication for policy or practice.

The four templates share the C1 essentials: structured thesis, four-move body paragraphs, broadening conclusion. They differ in how the body angles are organized. Picking the right template at minute 1 saves you the cost of restructuring at minute 30.

Worked example — turning a B2 paragraph into a C1 paragraph

The clearest way to see the difference between bands is to revise a B2 paragraph into C1 shape. Here is a typical B2-band body paragraph:

Firstly, AI is bad for students because they will not learn how to write. For example, if students use AI to write essays, they will not practice writing themselves. According to a study, students who use AI score lower on tests. So AI is harmful for education.

The paragraph has the three B2 moves (point, evidence, link) but no analysis, vague evidence, generic claim. The C1 revision:

The first cost of permissive AI policy is developmental. A 2023 Stanford study (Gerlich et al.) found that students who drafted essays with AI assistance scored measurably lower on subsequent unaided assessments, with the effect persisting at twelve weeks. The finding suggests that the cognitive struggle of drafting is not friction to be reduced but the actual mechanism by which writing skill develops; when AI absorbs the struggle, the skill that would have formed in response to the struggle does not form. This developmental cost is the first of three reasons that justify restriction.

Compare:

• Point: AI is bad for students (vague) → The first cost of permissive AI policy is developmental (specific, scoped).
• Evidence: according to a study (no source) → A 2023 Stanford study (Gerlich et al.) (authored, dated, specific).
• Analysis: missing → present (the finding suggests sentence explains the mechanism).
• Link: So AI is harmful (restating) → This developmental cost is the first of three reasons (forward-pointing to body 2).

The C1 paragraph is 102 words; the B2 is 51. The doubling is not padding — it is the analysis move plus proper attribution.

The 40-minute IELTS workflow

For IELTS Task 2 (or any 40-minute exam essay), the C1 workflow is roughly:

• Minutes 0-3: Parse prompt, identify essay type, sketch thesis. The thesis is the single most important decision; rushing it produces a disorganized essay.
• Minutes 3-6: Outline the three body angles. Decide which is strongest (body 1), which is structurally different (body 2), and what the strongest counter will be (body 3).
• Minutes 6-10: Draft the introduction. Hook, context, thesis.
• Minutes 10-22: Draft body 1, body 2, body 3. Aim for 80-90 seconds per claim-evidence-analysis-link cycle.
• Minutes 22-26: Draft conclusion. Resist restating; broaden.
• Minutes 26-32: Re-read for cohesion. Add cross-paragraph references. Vary connectors. Fix any hedge gaps.
• Minutes 32-38: Re-read for grammar and register. Catch contractions, calques, missing articles, agreement errors.
• Minutes 38-40: Final polish. Check word count is in range.

The workflow assumes the prompt is parseable on first read. If the prompt is ambiguous, spend minute 0-1 deciding which reading you will defend.

Pre-submission checklist

Before submitting any C1 opinion essay, run through:

• Thesis is one sentence, specific, debatable, and confident (not hedged).
• Hook earns the second sentence — no In today’s world openings.
• Each body paragraph has the four-move structure: claim, evidence, analysis, link.
• Counter-paragraph represents the strongest opposing view fairly.
• Conclusion broadens rather than restates.
• Causal claims are hedged where evidence is partial.
• Cross-paragraph cohesion is built through reference, not just linkers.
• No first-person I think in academic registers (unless prompt invites it).
• Word count between 300 and 350 — neither padded nor underdeveloped.

A pre-submission read aloud catches issues that silent reading misses: rhythm problems, calque connectors, sentences that have wandered from the thesis.

Three thesis archetypes that work at C1

Each archetype offers a sentence shape you can adapt. The mechanism-naming form is the safest at IELTS-style prompts; the counterintuitive form earns the most credit when executed well; the trade-off form is the most honest and the hardest to write.

Next lesson: Persuasive essay — ethos, pathos, logos, and the call to action.