How to Write the UPSC Essay — Practice

Question 1

You are given four Section A topics: (i) 'The empires of the future will be the empires of the mind'; (ii) 'Forests precede civilisations and deserts follow them'; (iii) 'There is no path to happiness; happiness is the path'; (iv) 'Nearly all men can stand adversity, but to test the character, give him power'. Apply the four-part selection test to topic (ii) and decide whether you would write it.

Accepted Answer

Test one — can I name the central tension in one sentence? Yes: the topic claims that civilisation systematically consumes the ecological base that made it possible, so the tension is between civilisation as achievement and civilisation as depletion.

Test two — six workable domains? Historical (Mesopotamia, the Indus, the Mediterranean deforestation that preceded soil loss), environmental (forest as regulator of water, soil and climate rather than as timber stock), economic (why extraction is rewarded and stewardship is not priced), governance (forest rights, common property, and who is treated as an encroacher), technological (whether technique breaks the pattern or accelerates it), ethical (obligation to those not yet born). That is six with material in each — the topic passes comfortably.

Test three — can I argue the opposite honestly? Yes, and this is the topic's strength. The counter-case is genuine: some societies have reversed forest loss while growing richer, restoration is possible, and the deterministic reading treats a tendency as a law. An honest essay has to answer that.

Test four — can I end somewhere? Yes. The essay can conclude that the aphorism describes a default rather than a destiny, and that what breaks the default is whether the costs of depletion are borne by those who cause it.

Decision: write it. All four tests pass, and the counter-case is strong enough to give the essay a real argument rather than a lament.

Question 2

Identify everything wrong, in essay terms, with this opening: 'Since time immemorial, mankind has been fascinated by technology. In this essay, we shall discuss the political, economic, social and ethical dimensions of artificial intelligence. What is AI? According to the dictionary, artificial intelligence means...' Then rewrite it.

Accepted Answer

Four separate faults, each costly. 'Since time immemorial' is a dead opening that signals a candidate with nothing specific to say. 'In this essay, we shall discuss' spends the most valuable position in the essay on administration rather than on thought. Listing the dimensions in advance announces a survey rather than an argument. And the straight dictionary definition is the weakest possible third move, because a definition offered without complication does no work.

A rewrite that does the introduction's three jobs — arrest, interpret, declare:

'In 1997 a computer defeated the world chess champion, and the reaction was less alarm than curiosity, because chess had always looked like the kind of thing machines might eventually do. The unease arrived much later, when systems began writing, diagnosing and deciding — tasks that had been assumed to require judgment rather than calculation. That shift is what makes artificial intelligence a different question from previous automation: earlier machines displaced effort, and these displace discretion. The argument here is that the significant risk is not that such systems will decide badly, but that they will decide adequately enough that we stop asking who is accountable for the decision.'

That version opens on a concrete moment, interprets what is distinctive about the topic, and states a line the essay will defend.

Question 3

Produce a dimension grid and a spine for the Section B topic: 'Social media is triggering fear of missing out amongst the youth, precipitating depression and loneliness.'

Accepted Answer

Grid. Individual: the comparison mechanism — feeds present curated peaks, and the comparison is made against a composite that no single person actually lives. Psychological: intermittent reward and the design of variable reinforcement; the difference between solitude and loneliness. Family and community: the displacement of unstructured, unobserved time in which friendship historically formed. Economy: the attention business model, where engagement rather than satisfaction is the optimised quantity, so the harm is a by-product of the revenue mechanism and not a defect in it. Technology: recommender systems that learn what holds attention, which is not what leaves a user better off. Governance: regulation options — age gating, design standards, transparency of ranking, and why each is hard; the free-expression constraint on content-based remedies. Comparative: jurisdictions experimenting with restrictions in schools and on minors' accounts. Philosophical: what connection actually consists of, and whether a substitute that satisfies its outward form can satisfy its function.

Spine (condition, cause, consequence, response — with a reflective turn to keep it an essay rather than a GS answer): (1) the observed condition, stated without exaggeration; (2) the comparison mechanism — why curated feeds produce a specific kind of inadequacy; (3) the business model, which explains why the mechanism is not incidental; (4) the displacement argument — what the time was previously spent on; (5) the honest counter-case: for the isolated, the disabled, the geographically distant and those in hostile local environments, these platforms supply connection that would not otherwise exist, and a purely negative account is false; (6) why the counter-case does not rescue the design, since the benefit comes from connection and the harm from optimisation for attention, and these are separable; (7) what follows — design-side and disclosure-side remedies rather than content policing; (8) the reflective close on what we are asking a technology to carry.

Connector test: (2) therefore (3), (3) hence (4), (4) but (5), (5) even so (6), (6) therefore (7). No 'also' anywhere — the structure is an argument, not a list.

Question 4

Write the interpretation move only — three to four sentences — for the topic 'The cost of being wrong is less than the cost of doing nothing.' Show what a careless reading would miss.

Accepted Answer

A careless reading turns this into a general topic about decisiveness and produces an essay in praise of bold leadership. That misses the printed claim, which is not that action is good but that a specific **asymmetry** holds between two costs.

The interpretation, as it would appear in the essay: 'The proposition is comparative rather than exhortatory. It does not claim that action is usually right; it claims that the expected cost of an error is smaller than the expected cost of inaction, and therefore that the two should not be weighed as if they were symmetrical. That framing is worth taking seriously because institutions systematically violate it: the cost of a wrong decision has an author and a file, while the cost of inaction is diffuse, delayed and unattributable, so a rational official facing career risk will under-act even when the arithmetic says otherwise. The question this essay pursues is therefore not whether to act, but why the asymmetry is so consistently mispriced — and where the proposition fails, since it plainly does not hold when errors are irreversible.'

That reading gives the essay its whole argument: the asymmetry, why organisations get it wrong, and the boundary condition of irreversibility.

Question 5

A candidate's essay has these paragraph-opening sentences: (1) Education is the backbone of any nation. (2) India has made significant progress in literacy. (3) The New Education Policy has several features. (4) Digital education has grown rapidly. (5) There are many challenges in the education sector. (6) The way forward requires multiple steps. Diagnose the structural failure and show what a repaired sequence looks like.

Accepted Answer

Diagnosis. Every opening is a topic label rather than a claim, so no paragraph commits to anything that the next could follow from. Apply the connector test: between (2) and (3), between (3) and (4), and between (4) and (5), the only honest connector is 'also'. The essay is a sequence of headings with prose underneath — precisely the General Studies answer wearing an essay's word count. Paragraph (5) quarantines every objection into one block and (6) supplies the way-forward reflex that marks the register error.

A repaired sequence, for an essay arguing that access has been solved faster than learning: (1) A country can enrol nearly every child and still fail to teach them, and India's experience over three decades is the clearest available demonstration. (2) The enrolment achievement was real and should not be minimised, because it removed the constraint that had bound the problem for a century. (3) But enrolment measured presence, and what was not measured was not managed. (4) The gap that opened is therefore not a failure of intent but of instrumentation: a system optimised for the metric it could count. (5) Digital delivery was offered as the correction, and it reproduced the same error at higher speed, measuring access to content rather than acquisition of it. (6) The objection deserves an answer: teaching quality is genuinely harder to measure than attendance, and crude measurement of it has done damage elsewhere. (7) Even so, the choice is not between crude measurement and none, and the reforms that have worked have measured learning at the level of the child rather than the school. (8) What follows is a change in what the system is asked to report, because a system will optimise whatever it is made to count.

Every join now carries an honest connector other than 'also', and the objection sits at (6) where it arises rather than in a quarantine block.

Question 6

For an essay arguing that technology has deepened democracy by lowering the cost of political participation, write the counter-argument section at its strongest, and then answer it without dismissing it.

Accepted Answer

The counter-case, stated at full strength: 'The claim that cheaper participation deepens democracy assumes that the binding constraint was cost, and the assumption may be wrong. What has fallen is the cost of expression, not the cost of deliberation — and these are different goods. A citizen can now signal a position in seconds, but signalling is nearly costless and therefore carries almost no information about intensity of preference or willingness to bear consequences, which is what political participation historically conveyed. Worse, the same reduction in cost applies to those who wish to simulate participation: coordinated inauthentic activity is cheap in exactly the proportion that genuine activity is. And the environments in which this expression occurs are optimised for engagement, which rewards the emotionally arousing over the true, so the deliberative quality of the participation may fall as its quantity rises. On this account technology has not deepened democracy but inflated its currency.'

The answer, which concedes rather than deflects: 'Most of this is correct, and an honest defence must accept it. Expression and deliberation are indeed different goods, and the second has not been made cheaper. But the objection proves less than it appears to, because it evaluates the change against an idealised prior state that did not exist. The deliberation that the pre-digital public sphere supposedly protected was available to a narrow stratum with access to editors, party structures and institutional platforms; for most citizens the participation on offer was a vote every few years and no channel in between. The relevant comparison is therefore not between cheap expression and good deliberation, but between cheap expression and near-total silence.

What the objection does establish is the reform agenda. If the problem is that the currency is inflated, the response is to make authenticity legible rather than to make expression expensive again — provenance for coordinated activity, transparency about how ranking is done, and deliberative institutions that deliberately raise the cost of contributing in exchange for genuine influence. The objection is thus better read as a specification of what is still missing than as a refutation of what has been gained.'

Question 7

For the topic 'Nearly all men can stand adversity, but to test the character, give him power', write the introduction and the conclusion, with the conclusion returning to the opening image.

Accepted Answer

Introduction. 'A district officer who has spent a decade being told to wait — for sanction, for posting, for a file to move — is finally given a charge in which nothing waits for anyone else. Within a year, colleagues who knew him well describe a different man, and they are not describing corruption; they are describing a person who has stopped explaining himself, because he no longer has to. The proposition that adversity is the easier test is often read as cynicism about human nature, but it is better read as an observation about **constraint**. Adversity supplies constraint from outside and asks only for endurance. Power removes constraint and asks the person to supply it themselves, from within — and it is that second demand, not the first, that most of us have never had occasion to practise.'

Conclusion. 'This is why the constitutional imagination has always been less interested in selecting good people than in designing offices that behave well when occupied by ordinary ones: fixed tenures, recorded reasons, separated functions, external audit. Such devices are frequently described as expressions of distrust. They are better understood as compassion — they restore, from the outside, the constraint that power removed, and they do so before the officeholder has to discover what he becomes without it. The man in the district was not selected badly. He was placed in the one situation for which nothing in a life of waiting had prepared him, and then left alone in it. The test of character is real, but a system that relies on individuals passing it unaided has misunderstood what it is asking of them.'

Note the mechanics: the conclusion introduces no new argument, is forward-looking, returns to the opening figure with a changed meaning, and ends in the candidate's own voice rather than on a borrowed quotation.

Question 8

The same theme — inequality — appears as a Section A topic ('The rich rule over the poor, and the borrower is servant to the lender') and a Section B topic ('Does economic growth in India reduce inequality?'). Show how the two essays should differ in register, structure and evidence.

Accepted Answer

Section A version. The topic is a claim about **domination**, not about distribution, so the essay must be about power rather than about Gini coefficients. Its correction is concreteness: every abstract claim gets pinned within two sentences. Structure runs on the philosophical spine — what the claim asserts (that economic dependence converts into personal subordination), why it has been true across very different economic systems, what changed when law began treating the borrower as a person with rights rather than an asset (debt bondage, insolvency as discharge rather than servitude, limited liability), and where it persists in forms law has not reached (informal credit, wage advances that bind labour, dependence that is unenforceable but real). Evidence is historical, legal and literary; a small amount of data is admissible but is not the essay's spine. The close asks what freedom means for a person whose choices are formally unconstrained and materially foreclosed.

Section B version. This one is a question about a measurable relationship and its correction is the opposite: it must not become a GS3 answer. It needs the structure of an argument about evidence — that growth's effect on inequality depends on **where the growth occurs**, so the answer differs for growth concentrated in capital-intensive sectors and growth in sectors that absorb labour; that poverty and inequality moved in different directions over the same period, which is the fact most often elided; and that the measure chosen partly determines the finding, since consumption-based and wealth-based measures give different pictures of the same country. Data must be attributed and structural rather than asserted as current point values. What keeps it an essay is the reflective turn: inequality of what, and why we care — outcomes, opportunity, or the political consequence of concentrated economic power — a question no dataset settles.

The summary of the difference: Section A takes an abstract claim and makes it concrete; Section B takes a concrete question and makes it reflective. The theme is shared; almost nothing else is.

Practice — How to Write the UPSC Essay

Sign in to save attempts to your dashboard and track your accuracy over time

Find a verified tutor

Join a live course

Browse coaching centres

About

Categories

Support

Documentation

Get Study Tips & Updates

Download Our App

Download for iOS

Download for Android

Popular Cities