LNAT Section A practice: 42 questions on 12 passages

Every few years, after a long fraud trial collapses or ends in an acquittal that surprises the prosecution, someone proposes that juries be dropped from serious fraud cases. The argument is familiar. The evidence runs to thousands of pages, the trials last for months, and ordinary people cannot be expected to follow the workings of complex financial products. Better, it is said, to leave such cases to a judge, perhaps sitting with expert assessors, who will understand what he or she is looking at.

The proposal has the appeal of common sense, and that is precisely what makes it dangerous. It assumes that the difficulty of a fraud trial lies in the arithmetic. It rarely does. The question a jury must answer in almost every fraud case is not whether money moved from one account to another — that is usually agreed — but whether the defendant was dishonest when he moved it. Dishonesty is judged by the standards of ordinary decent people. It is hard to think of anyone better placed to apply the standards of ordinary decent people than a group of them.

Nor is it obvious that judges would cope better with the complexity itself. A judge is a lawyer, not an accountant. A former commercial barrister may have read more balance sheets than the average juror, but the gap is smaller than reformers imagine, and in any case it is the prosecution's job to make its case intelligible. If a case cannot be explained to twelve attentive adults, that is often a sign that the prosecution has charged too much, too vaguely, rather than that the audience is at fault. Long trials are frequently the product of overloaded indictments, and the cure for an overloaded indictment is a shorter one.

There is also the question of who is in the dock. Defendants in serious fraud cases are disproportionately wealthy, well connected and articulate. It would be an uncomfortable irony if the one category of crime in which defendants most resemble the judges themselves became the category in which ordinary citizens were removed from the judging. Confidence in the criminal law depends partly on the belief that the powerful face the same tribunal as everyone else.

None of this means that fraud trials work well. They are too long, too expensive and too often fail. But the remedies lie elsewhere: firmer case management, simpler charges, clearer presentation of documentary evidence, and proper support for jurors who give up months of their lives. Abolishing the jury would treat a symptom as if it were the disease, and in the process discard a safeguard whose value is greatest exactly where reformers find it least convenient.

The debate about banning smartphones in schools has settled into a predictable shape. Supporters point to anxious, distracted pupils and to teachers who spend their lessons confiscating devices. Opponents reply that the research is mixed: some studies find that bans improve results, others find no effect, and very few can separate the effect of a ban from everything else a school was doing at the time. Since the evidence is inconclusive, they say, schools should not impose a restriction that interferes with family life and deprives pupils of a useful tool.

The opponents are right about the research and wrong about what follows from it. Inconclusive evidence is a reason for caution about large claims; it is not a reason to favour whatever arrangement happens to exist already. Schools routinely make rules for which there is no rigorous evidence at all — rules about uniform, about running in corridors, about where pupils may eat their lunch. Nobody demands a controlled trial before forbidding pupils to play football in the library. We ask instead whether a rule is reasonable, whether it is cheap to enforce, and whether it protects something worth protecting.

By those tests, a ban on phones during the school day does well. The cost is small. Pupils can still carry a phone to and from school, and parents can still reach their children in an emergency through the school office, as they did for generations before anyone owned a smartphone. What the ban protects — several hours a day in which a child's attention is not competing with a device designed to capture it — is valuable whether or not it shows up in examination results. A school is not only a machine for producing grades. It is also one of the few places left in which young people are required to be present with one another.

There is a fair objection that bans simply move the problem: pupils who cannot use their phones at school will use them all the more in the evening. Perhaps they will. But a school cannot be held responsible for every hour of a child's day, and the fact that a measure does not solve a problem everywhere is no argument against its solving the problem somewhere. The case for the ban is a modest one. It does not depend on believing that phones are the cause of every difficulty young people face, or that banning them will raise a single grade. It depends only on thinking that the school day is better without them.

The great museums of London, Paris and Berlin hold thousands of objects acquired during the colonial period, some by purchase, some as gifts and some as plunder taken by soldiers. When the countries from which those objects came ask for their return, the museums usually answer with a single, elegant argument. They are, they say, 'universal museums': institutions in which the art of every culture can be seen side by side, so that a visitor may walk from Egypt to China to West Africa in an afternoon and grasp the common humanity behind them all. To break up the collections would be to lose something that belongs to everyone.

It is a fine ideal, and I do not doubt that many curators hold it sincerely. But notice what the argument does. It takes a question about ownership and answers it with a claim about benefit. Whether an object should be returned depends, first of all, on how it was obtained. If a bronze was carried off from a royal palace by an invading army, the fact that it is now well displayed and seen by millions does not alter how it came to be there. We would not accept that a stolen painting should stay on the wall of the thief's house because his guests enjoy looking at it.

The universal museum also universalises rather selectively. It is striking how rarely its logic runs in the other direction. No one proposes that the crown jewels be sent on permanent loan to Lagos so that Nigerian schoolchildren can grasp the common humanity of European goldsmiths. The 'everyone' to whom these collections are said to belong turns out, in practice, to be those who can afford a trip to a European capital.

Some objections deserve a hearing. Museums worry, reasonably, that returned objects may be poorly conserved, or that a precedent will empty their galleries. But the first worry is an argument for helping to build conservation capacity, not for keeping things indefinitely; and the second is a strong argument only if much of the collection was wrongly acquired — in which case the precedent is precisely the point. Nor need restitution mean the end of shared display. An object that has been returned can be lent back, on terms its owners set.

What the universal museum cannot do is make the history of its collections irrelevant. The question 'Who should have this?' must be preceded by the question 'Whose is it?' Until museums are willing to answer the second question honestly, object by object, their talk of shared heritage will sound to many like a polite way of saying "ours".

Opponents of lowering the voting age to sixteen tend to rest their case on a single claim: that sixteen-year-olds are not mature enough to vote. It is worth asking what 'mature enough' is supposed to mean, because on almost any test we might devise, a large number of adults would fail it. We do not require voters to understand fiscal policy, to have read a manifesto or even to know the name of their local MP. We do not take the vote away from people who are impulsive, badly informed or easily swayed, and we do not ask whether a person's views are likely to change, though a voter of fifty may change their mind as readily as a voter of sixteen. If maturity were really the test, we would have to examine every voter, not just the youngest. The vote is not a reward for good judgement. It is a recognition that the people affected by a government's decisions should have a say in choosing it.

On that principle, the case for sixteen is strong. Sixteen-year-olds can work and pay income tax. They can be prosecuted for any criminal offence. They will live with the consequences of today's decisions about housing, climate and public debt for longer than anyone else at the ballot box. If being affected is what grounds the right to vote, they have at least as good a claim as their grandparents.

There is a practical argument too. People who vote in the first election for which they are eligible are more likely to keep voting for the rest of their lives, while those who miss it are more likely to drift away from politics altogether. At eighteen, many young people have just left home, moved to a new town for work or study, and are not yet registered to vote anywhere. At sixteen, most are still living with their families and attending school or college, where registration can be organised and the mechanics of voting explained. A first vote cast while lessons about Parliament are still fresh would also turn an abstract subject into a practical one. Lowering the voting age would allow the habit of voting to form at the point in life when it is easiest to form.

None of this requires us to pretend that sixteen-year-olds are especially wise. The argument is not that they would vote well. It is that they have as good a right as anyone else to vote badly.

The precautionary principle sounds like the most reasonable idea in the world. Where a new technology might cause serious harm, the burden should lie on those who want to introduce it to show that it is safe, and regulators should not wait for proof of damage before acting. Applied to crops, it has produced in many places a regime in which a wheat variety edited to resist a fungal disease can spend a decade and a fortune seeking approval, while a variety with similar traits produced by bombarding seeds with radiation — a technique in wide use since the 1950s — may be sown after little more than routine checks.

The anomaly is instructive. Radiation breeding produces thousands of random mutations, most of them unknown and unexamined; gene editing makes a small number of deliberate changes whose location is known exactly. If the aim of regulation were to manage risk, it would scrutinise the first technique at least as closely as the second. That it does the reverse suggests that what is being regulated is not risk but novelty. The older method is treated as safe not because it has been shown to be so but because people are used to it.

The deeper trouble with the precautionary principle, as it is usually applied, is that it counts the risks of acting and ignores the risks of not acting. A crop that cannot be grown is not risk-free. Farmers who are denied a disease-resistant variety will spray more fungicide; those denied a drought-tolerant one will lose harvests in dry years; consumers will pay more for food. These harms are real, but they are diffuse and invisible, because they consist of things that did not happen. A regulator who approves a product that later causes damage will be blamed. A regulator who blocks one that would have done good will not be noticed.

None of this is an argument for abolishing regulation. New crop varieties should be tested, whatever the method used to produce them, and some traits — a plant engineered to produce its own toxin against pests, say — raise questions that deserve careful examination. The point is that the questions should be about the plant, not about the technique. A regime that asked what a new variety does, rather than how it was made, would be both safer and fairer.

Precaution is a virtue only when it is applied in both directions. Applied in one, it is not caution at all but a preference for the familiar, dressed up as prudence.

When rents rise faster than wages, the call for rent controls follows as surely as night follows day. It is easy to see why. A cap on rents promises immediate relief to tenants, costs the government nothing and falls on landlords, who are rarely popular. Economists, almost alone, object, and their objection is usually dismissed as the voice of theory raised against the experience of people struggling to pay.

Yet the economists' case is not really theoretical. It rests on an observation about who benefits. A rent cap helps those who already have a tenancy and intend to keep it. It does nothing for those looking for somewhere to live, and it may harm them. If letting a property becomes less profitable, some landlords will sell to owner-occupiers, turn flats into short-term holiday lets, or simply stop maintaining what they let. The number of homes available to rent shrinks, and the newcomer — the young worker, the family moving for a job, the person leaving a relationship — finds the door closed. Those who already hold a controlled tenancy, meanwhile, have every reason to stay put, even when a new job or a growing family would otherwise lead them to move, so fewer homes come free at all. Controls turn a high price into a long queue, and in a queue, those who are well connected tend to reach the front.

Defenders of controls reply that tenants with secure, affordable homes are better off, and that this matters more than the abstractions of supply and demand. They are right that it matters. But a policy should be judged by its effects on everyone it touches, not only on the people it was designed to help. A measure that protects today's tenants by making tomorrow's search harder has not made housing more affordable; it has changed who pays.

The underlying problem in most expensive cities is simple: there are not enough homes where people want to live. High rents are merely its most visible sign. Any serious response has to involve building more, which means confronting the planning rules, local objections and infrastructure costs that make building slow. That is harder and slower than passing a cap, which is exactly why caps appeal to politicians. But capping the price of something scarce does not make it less scarce. Every flat that is never built means one more household competing for a flat that already exists. At best a cap postpones the reckoning; at worst it disguises the shortage while allowing it to grow.

Courts in several countries now use computer programs to estimate how likely a defendant is to reoffend, or to fail to turn up for trial, if released on bail. The programs are built from the records of past cases and produce a score that judges may consult when deciding whether to grant bail. Critics have attacked them on two grounds: that they are opaque, since their workings are often protected as commercial secrets, and that they are biased, since the records from which they are built reflect decades of unequal policing.

Both criticisms have force, but the second is often made in a way that proves less than its authors think. It is true that a program built on biased records will reproduce some of that bias. But the relevant comparison is not between a biased program and a perfectly fair judge. It is between a biased program and the judges we actually have, who have learned from the same history, carry many of the same assumptions, and often make their decisions in a few minutes at the end of a crowded list. Studies have found that bail outcomes can vary markedly with the judge assigned, and one much-cited study of parole hearings reported that decisions shifted with the time of day and even with whether a hearing fell just before or just after a break. A program is at least consistent: two identical defendants will receive identical scores.

Consistency, however, is not the same as fairness, and this is where the first criticism becomes decisive. A judge's decision can be challenged: she must give reasons, and those reasons can be examined on appeal. A score produced by a secret formula cannot be challenged in the same way. A defendant told that he is 'high risk' has no means of discovering whether the program has given weight to his postcode, his family's criminal record or his employment history, still less of arguing that it should not have. A system that cannot be questioned does not deserve the name of justice, however accurate it may be on average.

The right conclusion is therefore neither that such tools should be banned nor that they should be welcomed, but that they should be permitted only on conditions. Their formulas should be published. The factors they use should be open to challenge in the individual case. Their performance should be independently audited, including for differences between groups. And their scores should inform judges, never replace them. A tool that meets these conditions might make bail decisions both more consistent and more fair. A tool that does not should have no place in a courtroom.

All four nations of the United Kingdom, beginning with Wales, have moved to an 'opt-out' system of organ donation, under which most adults are treated as willing donors unless they have recorded an objection. Supporters often defend the change by saying that anyone who does not opt out has, in effect, consented. This is the weakest argument available to them, and they would do better to abandon it.

Consent is not a matter of failing to object. A person who has never heard of a scheme cannot be said to have agreed to it, and however well publicised a change in the law, some people will never hear of it: those who do not follow the news, those who do not read English well, those who simply never think about their own death. To call their silence consent is to stretch the word until it means nothing. If we accepted this reasoning elsewhere — if we held people to a contract because they had not written to refuse it — we would rightly be alarmed. Silence may be many things — indifference, ignorance, procrastination — but it is not agreement.

Yet the case for opt-out does not need the language of consent. Every system has a default: something must happen to a person's organs if they have expressed no wish. Under opt-in, the default is that they are not donated. Opt-in is often described as neutral, as though it made no assumption at all; but declining to use a person's organs is as much a decision as using them. The choice is not between presuming and not presuming, but between two presumptions. The opt-in default too is a presumption about what people want, and it is a less accurate one, since surveys consistently find that most people say they would be willing to donate. If we must guess what the silent would have wanted, it is better to guess in the direction that most of them would have chosen — particularly when the cost of guessing wrongly in one direction is the death of patients waiting for transplants.

Framing the matter honestly has practical consequences. If the justification is a presumption about wishes rather than a fiction of consent, then the presumption should give way whenever there is evidence that it is mistaken — for example, when relatives report that the person had objected but never got round to registering. A system built on presumed wishes should be eager to discover the actual ones. A system built on presumed consent has no reason to look.

Every year, thousands of people are sent to prison for terms of six months or less. Most of them are not dangerous. They are shoplifters, petty thieves and people who have breached a court order, and many have problems with drink, drugs or mental health that prison does little to address. Prison rarely protects the public from them for long, and the evidence suggests that it makes matters worse: those released from short sentences go on to reoffend at higher rates than similar offenders given community sentences.

The reason is not hard to find. A short sentence is long enough to do damage and too short to do good. Within a few weeks a person can lose a tenancy, a job and contact with their children; yet no course of education or treatment can be completed in the time available, especially since prisoners are usually released part-way through their term. They come out with fewer of the things that keep people out of crime than they had when they went in, and with new acquaintances who know more about it. Before long many are back inside, which is why people who work in the system speak of a revolving door.

It is sometimes said that short sentences are justified even if they do not reduce crime, because they express society's condemnation of the offence. But condemnation can be expressed in other ways. A demanding community order — unpaid work, a curfew, compulsory treatment for addiction — is not a soft option for those who must comply with it, and breaching it can itself be punished. If the purpose is to mark disapproval, the question is whether prison marks it more effectively than the alternatives; and at a cost of tens of thousands of pounds a year for each prisoner, it would need to mark it very effectively indeed.

Scotland has adopted a presumption against sentences of twelve months or less, under which a court may still send someone to prison but must explain why no other sentence is appropriate. Something similar should be adopted here. A presumption is not a ban: it leaves room for the persistent offender who has failed every community order, or for the case in which a victim needs protection. What it changes is the starting point. Prison would become a sentence the court must justify, rather than the one it reaches for when nothing else comes to mind.

The objection that this is 'soft on crime' gets matters backwards. A policy that increases reoffending is soft on crime, however tough it sounds.

Advocates of the four-day working week have found a winning formula. Firms that try it, they report, discover that staff produce as much in four days as they used to in five, while being healthier, happier and less likely to leave. Since nobody loses, the argument runs, the only obstacle to change is habit.

I am sceptical of this claim, not because it is impossible but because of how the evidence is gathered. The firms that volunteer for trials are those whose managers already expect the change to work, and their employees know that the trial's success may decide whether they keep their extra day off. Under those conditions a burst of effort is hardly surprising. Changes to working practices often bring a temporary lift simply because they are new and people know they are being watched. What we need to know is whether it lasts once the arrangement has become routine, and whether it works in jobs — nursing, teaching, driving a bus — where much of the work consists simply of being there for a set number of hours. A ward still needs nurses on the fifth day.

But the productivity argument, even if it fails, is not the strongest case for a shorter week. Over the last century, rich countries have become several times more productive, and they have taken most of that gain as higher incomes rather than as more free time. Working hours did fall over the century, but far more slowly than output rose. That was a choice, even if it was rarely made consciously. There is no law of economics that says the next round of productivity growth must also be spent on more goods rather than on more hours of life. A society that chose to work four days and accept slower growth in what it could buy would not be irrational. It would simply be valuing time more highly than its predecessors did.

The honest version of the case, then, is not that a shorter week is free. It is that it has a price, and that the price may be worth paying. That argument is harder to win, because it asks people to give something up rather than promising them everything at once. But it has the merit of being true, and an argument that rests on a false promise will collapse the first time the promise is broken. Campaigners who want the four-day week to last should stop selling it as a free lunch and start making the case that lunch is worth what it costs.

Whenever a public figure is abused online, there are calls to end anonymity on social media. If everyone had to post under their real name, the argument goes, the cowards who hide behind pseudonyms would be exposed, and the knowledge that they could be identified would make them behave. The idea also seems to cost the law-abiding nothing: if you have nothing to hide, why mind being named? It is a tidy theory. Unfortunately, the people it is meant to restrain are among the least likely to be restrained by it.

Much of the worst abuse on large platforms is already posted under real names. People who are convinced that their targets deserve what they get feel little shame in saying so openly; some regard their abuse as a public service. Meanwhile, the people who depend most on pseudonyms are not trolls. They are the teenager exploring their sexuality in a hostile household, the employee describing wrongdoing at their firm, the woman who has fled a violent partner and does not wish to be found, the citizen of an authoritarian state who wants to criticise its government. For them a pseudonym is not a mask for cruelty but a condition of speaking at all. A rule that removed their cover would silence them long before it silenced anyone else.

Supporters of verification sometimes propose a compromise: users would prove their identity to the platform but could still display a pseudonym to others. This is better, but it is not as safe as it sounds. A database linking millions of pseudonyms to real identities is exactly the kind of thing that gets hacked, leaked or demanded by governments. Users would be trusting not only the platform's present intentions but the security of its systems and the good conduct of every future owner and every state in which it operates. Information of that kind, once gathered, rarely stays unused for long.

None of this means that online abuse should be tolerated. Platforms already know a great deal about their users — their devices, their networks, their patterns of activity — and the police can, with proper authority, obtain that information when a crime has been committed. The difficulty is usually not that abusers cannot be identified but that reports are ignored, investigations are under-resourced and platforms are slow to act. Those are problems of enforcement, and they will not be solved by a measure that burdens the innocent in the hope of catching the guilty.

When a government apologises for a wrong committed generations ago — a massacre, a policy of forced removal, a part in the slave trade — a familiar objection is raised. No one now alive committed the wrong, the critics say, so no one now alive has anything to apologise for. An apology from people who were not there is empty theatre: it costs nothing, means nothing, and may even cheapen the suffering of those who were wronged. On this view the decent course is silence, or at most the kind of regret one might express about a natural disaster.

The objection treats an apology as a confession of personal guilt. But a state's apology is not a statement by the individuals who happen to hold office; it is a statement by an institution that persists across generations. The state that apologises today is, in a meaningful sense, the same state that acted a century ago. It has the same name, the same constitution or its successor, the same treaties and the same debts. We do not think a government can disown a loan because the ministers who borrowed the money have died. It is hard to see why it should be able to disown its wrongs on that ground, especially when it continues to enjoy the advantages its past conduct secured.

Nor is it true that such apologies cost nothing. A sincere apology commits the one who makes it to an account of what happened, and that account can be held against later conduct. A government that has acknowledged that a policy of forced removal was wrong has made it harder for itself, and for its successors, to defend similar policies in future, and harder to resist claims for the return of records, land or remains. Some apologies have been followed by the opening of archives, support for survivors or changes to what children are taught. That is why apologies are so often resisted, delayed or worded with lawyerly care: those who make them understand perfectly well that words have consequences.

The objection does contain one truth. An apology that is followed by no change in conduct is indeed empty, and those who were wronged are entitled to treat it with suspicion. An apology issued while the policy it regrets carries on in another form may be worse than none at all. But that is an argument for following apologies with action, not for refusing to make them. Words alone are not enough; but deeds without words leave the wrong unnamed.