Essay technique
Intermediate Conclusions and Signposting
LNAT Section B · Technique guide
The question this answers
How should you signpost an LNAT Section B essay and close each paragraph with an intermediate conclusion, so that an admissions tutor can follow the spine of your argument on a single read rather than reconstructing it?
The stance
A strong Section B essay is not just well-reasoned; it is legible. Two cheap, high-yield techniques make it so: signposting, which tells the reader where each paragraph sits in the argument before they read it, and the intermediate conclusion, which closes each paragraph by stating what it has proved and tying that back to the thesis. Tutors read fast and read many; the essay that makes its own logical skeleton visible is the one that scores for structure and judgement, because it shows a mind that knows what it is doing and why.
Defining the terms
- Signposting: explicit connective phrasing ('My first argument is', 'A distinct objection is', 'It follows that') that announces the function of a sentence or paragraph before its content, so the reader tracks the argument's shape without re-reading.
- Intermediate conclusion: the closing sentence of a body paragraph that states what that paragraph has just established and links it back to the thesis, e.g. 'So, on the ground of liberty, voting should not be compelled'; it is a mini-verdict, not a summary.
- Distinctness marker: an opening signpost that says how this paragraph differs from the previous one ('Beyond liberty, there is a separate problem of legitimacy'), preventing two points from collapsing into one.
- Link-back: the part of an intermediate conclusion that explicitly reconnects the paragraph to the central thesis, so each point is visibly load-bearing rather than a free-floating observation.
- Logical skeleton: the chain of claims (thesis, point 1, point 2, point 3, defeated counter, conclusion) that a reader should be able to recover from your topic and closing sentences alone; signposting and intermediate conclusions are what make that skeleton readable.
- Metadiscourse: writing that talks about the argument rather than advancing it; good signposting is lean metadiscourse, while bad signposting ('In this essay I will now turn to discuss the next point which is also important') is padding that wastes scarce words.
Assumptions to interrogate
- That signposting is filler that pads the word count, when done well it costs three or four words per paragraph and saves the reader the far greater cost of getting lost.
- That good arguments speak for themselves, when under time pressure a tutor who cannot see your structure will not assume it is there; legibility is part of the argument, not decoration on top of it.
- That an intermediate conclusion just repeats the topic sentence, when its job is different: the topic sentence promises a claim, the intermediate conclusion banks it and ties it to the thesis.
- That signposting and original thought are in tension, when in fact clear signposting frees the reader's attention to notice your insight rather than spending it on navigation.
- That only the final conclusion needs to refer to the thesis, when an essay whose paragraphs never link back reads as a list of opinions rather than a single sustained case.
The case for
Open every body paragraph with a signpost that states the paragraph's function and its distinctness from the last.
The first sentence of a paragraph should do two jobs at once: announce what argument is coming and mark how it differs from the previous one. 'A first objection to compulsion is one of liberty' tells the reader this is point one and what it concerns; 'Beyond liberty, there is a separate problem of legitimacy' both opens point two and certifies that it is not a restatement of point one. This distinctness marker is the single most efficient guard against the commonest structural fault, which is three paragraphs that say the same thing in different words. It also gives the tutor the range and control they read as a first-class structure, because they can see at a glance that you are covering genuinely different ground.
Close every body paragraph with an intermediate conclusion that links back to the thesis.
A paragraph that argues hard and then simply stops leaves the reader to work out what it proved and whether it mattered. The intermediate conclusion does that work for them: 'So, on the ground of liberty, voting should not be compelled.' This single sentence banks the point, names which thesis it supports, and signals that the paragraph is finished. Across an essay these closing sentences form a visible chain, the thesis defended three times over, which is exactly the sustained, cumulative argument that distinguishes a top script from a list of loosely related observations. It is also self-disciplining: if you cannot write an honest intermediate conclusion linking a paragraph to your thesis, the paragraph is off-topic and should be cut.
Use a small, varied set of connective signposts to expose the logical relationships between points.
Logic words are the joints of an argument. 'Therefore' and 'it follows that' mark inference; 'however' and 'yet' mark a turn; 'more importantly' and 'decisively' mark the relative weight of points; 'granting that' and 'even if' introduce a concession before a rebuttal. Deploying these deliberately lets the reader see not just what you claim but how the claims relate, which is the difference between a tidy essay and a genuinely argued one. The discipline is variety and restraint: the same connective repeated every sentence reads as a tic, and signposting that announces itself too loudly ('Now I will move on to my next point, which is') wastes the words that lean signposting saves.
Signpost the counter-argument explicitly, so the reader sees you choosing to meet the strongest objection.
Engaging the best objection is the clearest single signal of a balanced mind, but only if the reader can tell that is what you are doing. 'The strongest case for compulsion, fairly stated, is that voting is a civic duty' frames the next sentences as a steel-manned opposing view, not a wobble in your own position. The matching signpost then marks your return: 'But the analogy breaks at the crucial point.' Without these flags a charitable statement of the other side can be misread as a concession you never recovered from; with them, the whole move reads as deliberate intellectual control, which is precisely what the test rewards.
Let the introduction preview the structure and the conclusion echo the signposts, so the essay is framed front and back.
The introduction should state the thesis and name the route the essay will take ('I will argue this on three grounds: liberty, legitimacy and proportionality'), giving the reader a map before they start. The conclusion then synthesises by gesturing back to those same three signposted grounds without introducing new material. This framing means a tutor who reads only your first and last paragraphs already has the spine of the argument, and a tutor who reads the whole thing finds that the body delivered exactly what the introduction promised. Consistency between the promised structure and the delivered structure is itself a mark of control.
The case against
Heavy-handed signposting reads as mechanical and can crowd out the actual argument.
There is a real failure mode where every paragraph opens 'My next point is' and closes 'Therefore this proves my thesis', producing a script that is perfectly navigable and entirely lifeless. If the candidate spends their attention on the scaffolding rather than the reasoning, the essay becomes a filled-in template, and a tutor who has read the same connective phrases fifty times that morning is unmoved by tidiness alone.
Intermediate conclusions can become repetitive padding that inflates length without adding thought.
If each closing sentence merely restates the topic sentence in slightly different words, it doubles the word cost of every paragraph for no analytical gain. In a 600-word essay that waste is expensive: five paragraphs each carrying a redundant closing line can cost forty or fifty words that a stronger candidate spends on an extra argument or a sharper rebuttal.
Over-signalled structure can mask a thin argument and lull the writer into thinking the essay is finished.
Clean signposting makes an essay look complete, which is dangerous when the content underneath is shallow. A candidate can produce a beautifully flagged sequence of three points that are individually weak or that never genuinely engage the prompt's hardest version, and the very clarity of the signposting can stop both writer and reader from noticing that the reasoning has not actually done the work.
The argument, step by step
- Fix your thesis and your three or four distinct points first; signposting cannot rescue an essay that has no skeleton to expose, so the structure must exist before you label it.
- Write each body paragraph's opening sentence as a signpost that both states the point and marks its distinctness from the previous paragraph, using phrases such as 'Beyond X, there is a separate problem of Y'.
- Inside the paragraph, use lean connectives ('therefore', 'however', 'even if', 'decisively') to show how each sentence relates to the last, so the reasoning is visible and not merely asserted.
- Signpost the counter-argument when you reach it ('The strongest objection, fairly stated, is...') and signpost your return to it ('But this fails because...'), so the reader sees the steel-man and the rebuttal as a deliberate, controlled move.
- Close each body paragraph with an intermediate conclusion that banks the point and links it back to the thesis ('So, on this ground, X should be supported'); if you cannot write that sentence honestly, cut the paragraph.
- Frame the essay front and back: preview the signposted route in the introduction and echo those same grounds in the conclusion, then check that a reader of only your topic and closing sentences could recover the whole argument.
The model plan
Worked example on a paraphrased prompt: 'Should the law require people to give reasons before they are allowed to protest in public spaces?' Stance: no, a reasons requirement inverts the burden the right to protest is meant to carry and chills exactly the spontaneous, unpopular speech the right exists to protect; legitimate concerns about disruption are better met by content-neutral conditions on time, place and manner. Time map: 0-3 interrogate the prompt (does 'reasons' mean justification to the state, or mere notice?); 3-8 skeleton, word budget, and draft the five signposts and five intermediate conclusions before writing prose; 8-33 write; 33-40 proof, checking each paragraph opens with a distinctness marker and closes with a link-back. Intro (80-100w): state thesis, name the three signposted grounds (burden of justification, chilling effect, proportionality of alternatives). P1 signpost = 'The first objection is one of burden'; argue a free society does not make citizens justify the exercise of a right; Article 11 ECHR protects peaceful assembly and the state must justify restricting it, not the citizen justify using it; intermediate conclusion = 'So, on the ground of burden, no reasons requirement should be imposed.' P2 signpost = 'Beyond the formal question of burden lies a separate, practical harm: the chilling effect'; argue a reasons test deters the spontaneous and the unpopular, the very speech most in need of protection; intermediate conclusion = 'So, on the ground of chilling effect, the requirement should be rejected.' P3 signpost = 'The decisive ground, distinct again, is proportionality'; even granting genuine concerns about disruption, content-neutral time-place-manner conditions are a less restrictive means than a justification test, so the requirement fails the least-restrictive-means limb; intermediate conclusion = 'So, on the ground of proportionality, the requirement cannot be defended.' Counter-paragraph signpost = 'The strongest case for a reasons requirement, fairly stated, is public order'; concede force, then rebut: order is protected by regulating conduct, not by vetting motive. Conclusion (60-80w): echo the three signposted grounds, restate thesis once, no new material.
The model essay
Public protest is disruptive by design, and the wish to manage it is understandable. I will argue, however, that requiring people to give reasons before they may protest is the wrong instrument: it inverts the burden a fundamental right is meant to carry, chills the spontaneous and unpopular speech the right exists to protect, and fails the test of proportionality because gentler tools are available. I defend this on three grounds, taken in turn.
The first objection is one of burden. A free society does not, as a rule, make its citizens justify the exercise of a right before they may exercise it; we do not ask people to explain why they wish to worship, or to speak, before permitting them to do so. The right to peaceful assembly protected by Article 11 of the European Convention places the burden of justification on the state that would restrict the protest, not on the citizen who would hold it. A reasons requirement quietly reverses this: it treats protest as a privilege to be argued for rather than a liberty already held. One might reply that giving a reason is a trivial administrative step. But the triviality is illusory, because the moment a reason can be demanded it can be judged, and a liberty that the state may withhold until satisfied is no longer a liberty. So, on the ground of burden, no reasons requirement should be imposed.
Beyond the formal question of burden lies a separate and more practical harm, which is a distinct argument rather than a restatement of the first: the chilling effect. The protests that most need protection are precisely the spontaneous, the inconvenient and the unpopular, the response to yesterday's news or the cause the majority dislikes. A requirement to articulate and submit reasons in advance falls hardest on exactly these, because it adds friction, delay and the fear of refusal to the very speech that is least able to bear them. Comfortable, well-organised, uncontroversial protest will clear the hurdle easily; it is the dissenting voice that is deterred. A right that bites only on the powerful and not on the marginal has been hollowed out while appearing intact. So, on the ground of chilling effect, the requirement should be rejected.
The decisive ground, distinct again, is proportionality. Suppose we grant that the state has a real and legitimate concern about serious disruption to ordinary life. The question a serious legal mind must then ask is whether a less restrictive means is available, because proportionality forbids the heavier tool where a lighter one will do. Here the lighter tools are obvious: conditions on the time, place and manner of a protest regulate its impact without interrogating its purpose, and notice requirements let the authorities prepare without demanding that anyone justify their cause. A justification test is both more intrusive and less precise than these, since it polices motive rather than conduct. A measure that fails the least-restrictive-means limb is hard to defend whatever its aim. So, on the ground of proportionality, the requirement cannot be defended.
The strongest case for a reasons requirement, fairly stated, is public order: the state must protect those not protesting, and knowing why a crowd has gathered helps it anticipate violence and prevent harm. I accept the concern is genuine and the aim legitimate. But the argument proves something narrower than it claims. Public order is protected by regulating what protesters do, the route they take, the hours they keep, the noise they make, not by vetting why they have come. The link between a protester's stated reason and the risk of disorder is weak, while the cost to liberty of demanding that reason is high, and a measure that buys so little order at so great a price to freedom cannot be the proportionate one.
A reasons requirement, then, answers a real problem in the wrong register. It mistakes managing the impact of protest for licensing its purpose, and it spends the coin of a fundamental freedom to buy an assurance that conduct-based rules already supply. We should not require people to justify their protests before they may hold them; we should regulate how protests happen and leave why they happen where a free society keeps it, beyond the state's permission.
Authorities worth knowing
DPP v Ziegler
[2021] UKSC 23; [2022] AC 408
Used here to model how a single authority earns its place rather than to be name-dropped: the Supreme Court treated the right to protest under Articles 10 and 11 ECHR through a structured proportionality assessment, which is the same analytical move the model essay signposts in its third paragraph, showing that the burden of justifying a restriction sits with the state.
R (Daly) v Secretary of State for the Home Department
[2001] UKHL 26; [2001] 2 AC 532
Lord Steyn's restatement of structured proportionality (legitimate aim, rational connection, least restrictive means, fair balance) is the template behind the signposted 'is there a less restrictive alternative?' move that gives the model essay its decisive third intermediate conclusion.
How the law frames it
United Kingdom
Section B is read by UK law schools (Oxford, UCL, KCL, Bristol, Durham, LSE, Nottingham, Glasgow and others) as part of admissions; tutors use it to judge written reasoning, structure and judgement rather than legal knowledge. Signposting and intermediate conclusions map directly onto the 'clarity and structure' and 'critical development' an examiner is looking for, and they mirror how UK public-law reasoning itself proceeds, by stating a test, applying it, and concluding before moving on.
Canada
Although the LNAT is a UK test, the legibility discipline it rewards transfers directly to North American admissions and personal-statement writing and to the IRAC method taught in Canadian and US law schools, where each paragraph is expected to state an issue, apply a rule, and close with a conclusion that the reader can lift out cleanly.
ECHR
Public-policy prompts in Section B frequently turn on rights reasoning under the European Convention, and the proportionality structure (legitimate aim, rational connection, least restrictive means, fair balance) is itself a sequence of signposted intermediate conclusions, which is why practising explicit link-backs prepares a candidate for the kind of analysis the degree will demand.
Counter-arguments and how to defeat them
Counter. heavy signposting reads as mechanical and kills originality.
Rebuttal. that is a charge against bad signposting, not signposting as such; lean, varied connectives free the reader's attention for your insight, and the cure for a robotic 'my next point is' is better phrasing, not silence.
Counter. intermediate conclusions just repeat the topic sentence and waste words.
Rebuttal. the two sentences do different jobs; the topic sentence promises a claim and the intermediate conclusion banks it and links it to the thesis, and if they really are identical the paragraph has not advanced and needs rewriting, not the closing line removed.
Counter. clean signposting can disguise a thin argument.
Rebuttal. true, which is why the technique is a presentation discipline layered on top of genuinely distinct points, never a substitute for them; the honesty test of the intermediate conclusion ('can I truthfully link this back to my thesis?') is itself a guard against padding.
Counter. in only 600 words you cannot afford the words signposting costs.
Rebuttal. lean signposting costs three or four words per paragraph and saves the far larger cost of a reader who has to reconstruct your structure, so the trade is strongly positive; it is verbose signposting, not signposting, that is unaffordable.
Counter. only the final conclusion needs to reference the thesis.
Rebuttal. an essay whose paragraphs never link back reads as a list of opinions that happen to share a topic; the recurring link-back is what turns several points into one sustained, cumulative case, which is the structure examiners actually reward.
Conclusion
Signposting and intermediate conclusions are the cheapest upgrade available to a Section B essay, and among the most consequential. They cost almost nothing in words and they change how the whole script reads: a tutor who can see your skeleton on a single pass credits you with the control and judgement the test is built to measure, while a tutor forced to reconstruct your structure credits you with neither. Open each paragraph by saying what it argues and how it differs from the last; close each one by banking the point and tying it back to your thesis; flag your counter-argument so the steel-man reads as a choice; and frame the whole essay so its first and last paragraphs already carry the spine. Make the argument legible, and you have made half of it persuasive before the reader has weighed a single claim.
Evidence you can cite
- Section B of the LNAT gives candidates 40 minutes to write one essay chosen from three prompts, and the essay is not scored by LNAT but is sent to and read by the law schools the candidate applies to, who judge written reasoning and structure rather than legal knowledge.LNAT (National Admissions Test for Law), test format overview — source
- Section A of the LNAT comprises 42 multiple-choice questions across twelve passages with 95 minutes allowed, so the full test runs 135 minutes; because Section B has the tighter clock relative to its difficulty, legibility techniques that save the reader and writer time carry an outsized return.LNAT (National Admissions Test for Law), test format overview — source
- LNAT's own guidance recommends an essay of roughly 500 to 600 words, with around 750 treated as a practical ceiling; at that length every word of signposting must earn its place, which is why lean intermediate conclusions beat verbose ones.LNAT, 'Preparing for the LNAT' guidance (official preparation materials) — source
Further reading
- LNAT, 'Preparing for the LNAT' and the official 'Your Guide to Preparing for the LNAT' PDF (essay format, length and assessment guidance)
- LNAT, sample essay questions and example essays with examiner commentary (for how structure reads to a marker)
- DPP v Ziegler [2021] UKSC 23 (structured proportionality applied to the right to protest)
- R (Daly) v Secretary of State for the Home Department [2001] UKHL 26 (Lord Steyn on the proportionality sequence)
- George Orwell, 'Politics and the English Language' (on cutting words and writing plain, legible prose)
- Joseph Williams, 'Style: Lessons in Clarity and Grace' (on cohesion, old-to-new information flow and metadiscourse)