How Laws Discriminate

Stephen Sedley

‘One law for the Lion & Ox,’ wrote Blake, ‘is oppression.’ He was describing in his oblique way what Anatole France a century later described more brutally as ‘the majestic even-handedness of the law, which forbids rich and poor alike to sleep under bridges, to beg in the streets and to steal bread.’ France’s English contemporary Lord Justice Mathew made the point in more genteel terms: ‘In England,’ he said, ‘justice is open to all, like the Ritz.’ The Early Victorian poet Thomas Love Peacock had noted the unequal impact of the Sunday observance laws:

The poor man’s sins are glaring;
In the face of ghostly warning
He is caught in the fact
Of an overt act –
Buying greens on a Sunday morning.

The rich man is invisible
In the crowd of his gay society;
But the poor man’s delight
Is a sore in the sight
And a stench in the nose of piety.

The rich man goes out yachting
Where sanctity can’t pursue him;
The poor goes afloat
In a fourpenny boat
Where the bishop groans to view him.

But the truth is that all laws discriminate. They discriminate between the virtuous and the wicked, between the permitted and the prohibited, between the taxable and the duty-free. They discriminate, too, on grounds which from era to era are taken to be so obvious that they do not even require justification. It was obvious that the right of Athenian citizens to vote did not include women or slaves. Among the American founding fathers who proclaimed the self-evident truth that all men are born equal were several slave-owners. In this country until the 20th century the unsuitability of women to vote, sit on juries or join the professions was regarded – at least by men – as too obvious for argument. We continue to regard it as self-evident that the freedoms which we now regard as the birthright of all men and women without distinction do not apply to children. Our law, both common law and statute, permits acts against children which, if committed against adults, would amount to criminal assaults. It was only in this generation, and only by the narrowest of margins, that the House of Lords in Victoria Gillick’s unsuccessful case abandoned the notion of parental rights as a form of proprietorship and put in their place a balance between the child’s evolving autonomy and the parent’s role as carer.

As to crime, it is not law – the argument goes – that criminalises some people and not others, but social conditions or personal choice that lead wrongdoers to do wrong. The law may be able to mitigate the consequences for those who offend through misfortune, but it cannot treat them as free of blame without forfeiting the very claim to even-handedness which its detractors mock. But Blake, too, was right to claim that one law for all is ‘oppression’. His was the age of large-scale enclosures and of the Game Laws when, as the jingle went:

The fault is great in man or woman
Who steals a goose from off a common;
But what can plead that man’s excuse
Who steals the common from the goose?

Enclosure in England was the work of the law, but few poor people benefited from it. The rich never found themselves trespassing in search of game: they could pursue it on their own or their friends’ land. The law which in form governed the powerful and the submissive – the lion and the ox – without distinction, was in substance a means by which the one could oppress the other, and was meant to be so. There is little doubt that the sole reason Georgian and Regency judges, who were otherwise active in developing new crimes, did not criminalise trespass by itself was that it would have made foxhunting impossible. The dilemma has plagued the law to the present day, resulting in the creation of statutory constructs like ‘trespassory assembly’. So undisguised an intention to discriminate by law between classes, genders or races may be a thing of the past, but the unequal effects of equal laws remain a living – indeed a growing – issue.

It has been one of the great achievements of the last generation to recognise and begin to grapple with this snake in the legal grass. Responding to the pressure of the Civil Rights movement, the United States’s Civil Rights Act 1964 set out a simple prohibition on all discrimination on grounds of sex or race. (The inclusion of sex was a brilliant accident: the measure was designed originally only to prohibit race discrimination; an amendment to include sex discrimination, intended to make the Bill a laughing stock, was moved and passed; and so was the Bill as amended.) It was left to the courts to make this bald provision work, and they did so by recognising that discrimination could occur where, although the same condition was being applied to everyone concerned, it had a disproportionate adverse impact – perhaps deliberate, perhaps accidental – on one sex or racial group. Early in the now chequered existence of the Civil Rights Act the Supreme Court ruled: ‘Congress has now provided that tests or criteria for employment or promotion may not provide equality of opportunity merely in the sense of the fabled offer of milk to the stork and the fox ... It has ... provided that the vessel in which the milk is proffered be one all seekers can use.’ The Sex Discrimination Act 1975 and the Race Relations Act 1976, leaving nothing to chance in the manner of British drafting, spelt out the analytical process by which indirect discrimination was to be identified and, if identified, justified. The formula can be problematical to apply, but over time it has worked, and not only in courts and tribunals: far more importantly, it has been used to change good practice, replacing subjective and undisciplined recruitment procedures with relevant criteria designed to find the best candidate.

But neither law nor practice has succeeded in eliminating inequality of opportunity for ethnic minorities and women; nor will they, in some fields at least, unless entrenched forms of élitism, most of them based on inherited and unarticulated assumptions, are directly challenged. The powerful case for affirmative action in some of these areas – that is to say, the argument for loading the criteria of choice in favour of those who have historically been disadvantaged – encounters the equally powerful argument that if discrimination on grounds of race or sex is wrong, then practising it for benign reasons cannot make it right. It also faces the practical argument that, while affirmative action to encourage disadvantaged groups to come forward is acceptable, positive discrimination by definition involves a selective dilution of standards. These arguments, built as they are on the very legislation which sets out to combat discrimination, seem to have a monopoly of justice: how, they ask, can you combat discrimination by discriminating? And yet history argues differently. It shows us centuries of positive discrimination in favour of white men, of jobs and advantages going to incompetents and mediocrities whose faces happened to fit or who had the right connections. It helps to explain why forms of inequality remain embedded in our ways of thinking and operating. Women and members of ethnic minorities still face problems of self-confidence even when they are merely deciding to try to enter fields of activity where the white male image dominates. For those who do enter, experience still suggests that to succeed they have to do better than their white male counterparts. As a society we continue to set our faces against practices which will turn these tables. We accept the legitimacy of target numbers against which to monitor performance; but we do not allow the use of quotas to redress performance which is proving inadequate. The consequent near-stalemate is a different but no less real form of injustice. We have legislated against individual acts of discrimination, for each of which the law can try to provide a remedy; but we have no legal means of dealing with the kinds of systemic disadvantage that the legislation has so far failed to reverse. One marginal solution, within the present law, is to give priority to those relevant criteria of choice which members of disadvantaged minorities are more likely to be able to satisfy. Another, now adopted in such countries as Canada and New Zealand, is to keep a quota of places for historically disadvantaged minorities in institutions – chiefly universities – which provide the passkeys to the problem areas of employment, housing and so on.

These are entirely defensible, albeit contested, initiatives. But they throw up a succession of further questions. If the number of university places is finite, for everybody who is admitted somebody is excluded. It may be law’s task to define the permitted and the forbidden criteria of choice, but in making the distinction legislatures, and judges in their wake, have to make fundamental decisions about justice. They must choose, in particular, between immediate justice to individuals and long-term justice to segments of society – not because of some inherent superiority of group over individual but precisely because the typecasting of many individuals as members of a group may make it necessary to take remedial steps on behalf of that group in order to liberate the individuals who compose it.

There is a powerful argument that judges, if they are to do justice, ought to resort to principle rather than to precedent in deciding difficult cases. But the argument only begins here. The question it repeatedly begs is: what principle? There are issues, of which positive discrimination is one, on which fundamentally different positions of principle can legitimately be taken. Ronald Dworkin has persuasively defended reverse discrimination both on the grounds of concern and respect which inform his theory of justice and on utilitarian grounds. But, as he acknowledges, principles can be misappropriated. How would we react if the principle on which Dworkin defends selection criteria designed to redress the under-representation of blacks in American law schools were used to reduce a real or supposed over-representation of Jews? The issue – which Dworkin himself recognises – illustrates something that triumphalist talk about justice ignores: even from a single broad standpoint there may well be more than one just outcome, and the choice between these outcomes may itself raise questions of fundamental principle which cannot be resolved without resort to philosophical and political premises that ordinarily lie beyond the judicial remit. In the kind of case I have mentioned, for example, the underlying choices that have to be made are between social and moral evils which have somehow to be first quantified and then compared. In other cases the law finds itself trapped between inherited assumptions and a developing social morality: can there be rape within marriage; what beneficial interest does a non-earning spouse have in the matrimonial home; do parents have a right to beat their children; does sovereign immunity extend to perpetrators of crimes against humanity? In such situations the courts have choices of principle to make between the injustice of deciding that the law is not what it was thought to be and the injustice of tolerating something that has become or is becoming intolerable. Although there are principles of fairness which limit the choice – for example, a change in the law may not criminalise what was an innocent act when it was committed, nor destroy vested rights – there remains a large area governed in the last resort only by the courts’ receptivity or resistance to change. And because the courts decide only real cases, not abstract questions, there is superimposed the problem that the just solution of a hard case may make law which produces fresh and unexpected injustices in other cases. It is not a sufficient response to this familiar problem to counsel endless judicial caution by reciting the adage that hard cases make bad law; bad law makes hard cases too.

The rhetoric of the law should not obscure the fact that justice – of process as much as of outcome – is in large part a matter of personal perception. What justice a legal system achieves is more likely to be a negotiated outcome with rough edges and loose ends than a triumphal result pronounced to universal acclaim. And the complication of justice is added to by disparities of access. With the coming of the Human Rights Act one can venture some predictions, based on experience in this country as well as in Canada under its 1982 Charter of Rights and Freedoms and New Zealand under its Bill of Rights Act 1990. One is that when – at a date which is not likely to be before the middle of the year 2000 – the human rights shop opens, that endlessly indignant litigant, the drinking driver, will be waiting there. He will be hoping to follow the trail blazed under the Charter by his Canadian counterpart, who was stopped by the police at three in the morning, in sub-zero conditions, well inside the Arctic Circle and went to court to assert that the right to counsel entitled him not to have to blow into the bag until his lawyer had been brought several hundred miles from Whitehorse or Dawson City. Another – I don’t mean to put the two on a par – is that litigation about freedom of expression under Article 10 of the Convention will be near or at the head of the queue. Why should it be so?

Since anybody who even asks such a question risks being branded an enemy of free speech, let me first reiterate that the right to utter criticism or heresy without fear of suppression or reprisal from those who may be angered or embarrassed by it is fundamental in any free society. But other rights are equally important. Why does experience suggest that they tend to be less frequently and less stridently claimed? Does the disparity have to do with disparities of power? Is it the lion who litigates while the ox puts up with things? Access to the courts is expensive, but for those organisations which possess and depend for their prosperity on audibility, inhibitions on what they can say are important enough to justify investment in rights litigation. We are likely to see commercial interests prominent among those claiming the right of free expression in relation to the advertising of things that government wants to regulate. Differences in financial muscle may also have a considerable influence on what is made in our courts of a particular scheduled right. Article 8’s provision that ‘everyone has the right to respect for his private and family life, his home and his correspondence’, for example, may quite rapidly establish a rich jurisprudence of privacy rights for the famous. But this is far less likely to be the case for the larger number of humble people whose autonomy is endlessly compromised by a variety of processes to which public administration subjects them as claimants, clients or patients. For them, legal advice is not readily available and legal aid, limited now by means-testing to the poorest, is likely to be the only route to court. For the middling sort real problems are likely to be encountered in finding lawyers who are prepared – or indeed financially able – to work for contingent fees on human rights issues. It may be for this reason, rather than as a reflection of the real incidence of Convention issues, that criminal law – where legal aid remains available as of right – will initially at least see the largest number of human rights arguments.

Why should any of this matter? Why is the market in litigation not itself a sufficient indicator of need? Law and economics theory, at least in its early versions, might well claim that it is. Communitarian theories, on the other hand, begin by positing fairness in the distribution of individual rights or freedoms as the essence of justice. To speak of justice, of course, is not necessarily to speak of law: the judicial job is to achieve the one in applying the other, but the rigidity of law and the elusiveness of justice continually conspire to keep the two things in tension. If, as I have argued, there is frequently no single just outcome to a particular conflict, justice itself requires the possibility of non-confrontational forms of conflict resolution; and the courts themselves are starting to recognise and promote mediation as a sometimes superior route to justice. But to the extent that courts of law, with their confrontational processes, remain the forum in which justice is ordinarily sought, it is (or so it seems to me) from a common ethical sense rather than from any prescriptive or functional source that justice as a value embodying fairness and equity has to be derived. Neither Rawls nor Dworkin can prove why it is justice in this sense that matters; nor need they so long as we endorse the moral sensibility which says that it does. It is perhaps significant that in an era when equality in other fields has either imploded or been exploded as a guiding ideal, equality before the law remains an uncontested good and an unchallenged right. Is there an element of human sensibility which, because it is as much aesthetic as moral, is unsatisfied with an outcome loaded by extraneous factors which skew the creative acts of debate and judgment by which justice is done and displayed? Whatever the reason, we continue to want the forensic playing-field to be level.

This is more easily said than done. It is not simply that some witnesses give a poorer account of themselves than others although they may well be the more truthful; nor that the quality of legal representation may unfairly favour one party, whether by chance or by purchase. It is that the legal process is itself based on assumptions which can turn procedures designed to achieve even-handedness into engines of oppression.

It is a central principle of our criminal procedure that the defence must be allowed to explore the weaknesses in the Crown’s case and to put its own case without inhibition. A rape case is in principle a criminal trial like any other, but in practice it has been known for a long time that unless the process is controlled a rape trial will as often as not be turned into a trial of the complainant. It is thanks largely to the fact that women have made themselves heard over the last thirty years or so that Parliament and the courts have tried in some measure to reverse this process. But the consequent statutory presumption against questioning a complainant about her sexual past, which is necessarily rebuttable if justice requires it in a particular case, has turned out to provide uneven and sometimes illusory protection. Such cross-examination is allowed by the courts where it ‘might reasonably lead the jury ... to take a different view of the complainant’s evidence’.

What is more, the adversarial process itself can obstruct rather than promote justice. This is recognised by the family courts, which have long since taken charge of the evidential process in order to ensure that when marriages break down the children are not fought over along with the house and the car, and that in abuse cases the court, and not whoever wins, decides what is safest for the child. Other branches of civil process lag behind; and criminal cases remain trapped in a body of rules which at times seem more apt to a game of snakes and ladders than to a system of justice. Many of the problems are highlighted in rape cases, but they are not special: they simply tend to show in particularly acute form some of the anomalies of the adversarial system in the form which history and culture have given it in England and Wales.

A rape case will typically involve a young woman whose lifestyle is somewhere between the casual and the chaotic, and a young man who has found her in this situation and has forced himself on her. A disturbing proportion of these defendants turn out to have been acquitted of one or more rapes in similar circumstances in the past. With the coming of DNA testing, the ‘it wasn’t me’ defence has almost disappeared. The defence now tends to be that the complainant consented, and the story will more often than not be cast in pornographic detail which presents the woman as the initiator. The first the victim is likely to learn of these counter-allegations is when, having told the jury her story, the defendant’s counsel gets up to cross-examine her. Although attention has been focused on defendants who dismiss their counsel and cross-examine in person, the experience of crucifixion by a skilled and remorseless advocate can be even worse. The very vulnerabilities which made the complainant prey to the defendant in the first place now make her a prey to the barrister and to the more censorious of the jurors: why was she not living at home? Why was she out at 2 a.m.? Why did she allow the defendant to be alone with her? We have not been able to reconcile the need to ensure that an accused person is not wrongly convicted with the equally important need to protect the complainant from becoming the accused in a counter-trial about her lifestyle. Judges become case-hardened: many may of course have heard this kind of defence so often that they could script it themselves. We know, too, that a high proportion of these men have access to offender networks (especially but not exclusively in prison, where it is made easier by the Home Office’s need to segregate sex offenders, including those on remand, for their own protection) where consent defences are circulated. The jury is an essential protection against the consequent risk of typecasting defendants; but because each jury is hearing the story for the first and only time there is an opposite risk that, having no objective evidence such as injuries (and these are the exception in rape cases), and not knowing how consistently consent defences are manufactured, they will accept that the defence might be true and will acquit. Nobody can say that in any one case they were wrong to do so; but experience strongly suggests that the low and still falling rate of conviction in rape cases reflects a systemic injustice to those women who fall victim in disturbing numbers to predatory men.

The task of achieving fairness between accuser and accused in such cases depends not only on the trial judge but on the trial advocate. There are counsel – many of them – who can put their client’s case and test the Crown’s evidence without compromise but equally without insult or injury. There are others – a small but prominent minority – who either cannot or will not do so, and who breach the Bar’s Code of Conduct, which forbids the asking of questions that merely vilify or annoy. With these the trial judge has a Herculean task in preventing the humiliation and bullying of the complainant without provoking a successful appeal on the ground that the defence was prevented from putting its case. The initiative and the responsibility at present rest with the Bar: there is no longer a place for a forensic culture which tolerates and even promotes something closer to bare-knuckle fighting than to a fair trial and a just verdict, and which fails to insist on advocacy that treats with respect individuals – the good, the bad and the ugly alike – who find themselves in the temporary but overwhelming power of an interrogator in a wig. There is no worthwhile evidence that inquisitorial systems are intrinsically better or fairer; but it does not follow that they are without virtues, or that an adversarial system has to demean and oppress those who get caught up in it. Fairness in a trial does not affect only the accused.

It is perhaps in the criminal sentencing process that justice comes under the cruellest spotlight. It is not just that almost every sentence except one prescribed by law is a compromise between justice to the wrongdoer and justice to the victim, or between such incommensurable imperatives as reform and retribution; nor simply that the law which continues to prescribe a life sentence for every murder is itself an impossible compromise between justice and vengeance. Nor is it simply that public comprehension of sentencing has been so damaged by media presentation that the public simultaneously believe that judges sentence too leniently and, when asked what they would do, turn out to favour sentences markedly lighter than those the judges impose. It is that we continue to be both fascinated and baffled by the interplay between compulsion and free will.

The law starts from an assumption that people are responsible for what they do in the sense that they can choose not to do it. It has for centuries made it a principle that ‘a man cannot stultify himself’ by excusing what he has done on grounds of his own fecklessness or incapacity. It abandons this position when dealing with those who are incapable of knowing what they are doing, or that what they are doing is wrong, or whose acts or mind are for one reason or another – intoxication excepted – not their own. For those who, though criminally responsible, are mentally disturbed it provides specialised disposals at the point of sentencing. But this leaves a vast number of cases – the majority, indeed – in which a defendant may have been driven by something inside or outside him or herself to offend. For thieves it may be need, genuine or perceived; for sexual abusers it is commonly a history of having themselves been abused; frequently in mugging and burglary cases it is the compulsion to feed a drug addiction; and as frequently in cases of violence it is the short fuse of anger. The law is on the whole uncompromising in holding people responsible for what they do under such pressures (with the notable exception of the law of homicide, which uniquely allows provocation to afford a defence, reducing murder to manslaughter with a quantum drop in the penalty). Justice for its part, in the form of sentence, can sometimes show mercy but cannot dilute the principle of responsibility.

For a long time this moral, even moralistic, view has sat awkwardly beside the belief that criminality, far from being a matter of choice, is an aspect of personality – in its cruder forms a function of heredity discernible in the shape of the skull. While deductions from this have ranged from eugenic final solutions to arguments for the abolition of imprisonment, the one thing that theories of entrenched criminality cannot sustain is the belief that prison works, except by way of temporary containment. If prison is intended to reform (and our Prison Rules have since 1899 announced this as the primary purpose of imprisonment) it can only be because offenders are capable of changing their behaviour. Yet day after day realistic and thoughtful pre-sentence reports confront sentencing courts with portraits of individuals irreparably damaged by their formative experiences. The court knows, as it sends them down, that it is in a sense punishing them again for what life has already done to them; yet it also knows, or believes, that without a condign response society risks losing its already contested grip on civil order and the courts their toehold in public confidence; and so we continue to hold them responsible for what they have done.

This constant negotiation of the meaning of justice is going to find itself under new pressures. The kickback of the reductionist Darwinism most strongly associated in this country with ‘selfish gene’ theory has not just been a (possibly unintended) endorsement of a particular strain of individualist ideology. By including a fresh encouragement of determinist ideas about human behaviour it has reinvigorated the notion that anti-social conduct can be both appropriate and desirable – more appropriate and desirable than the artificial construct of human society. It also challenges received notions of freedom and free will, not so much by presenting the human being as a genetically programmed automaton as by postulating selfish ends as the ‘natural’ objectives of the free individual.

You cannot meet a scientific case of this kind by resisting its conclusions on moral or political grounds. What matters is the scientific critique of it as a monist grand theory which can make its case only by denying the demonstrable complexity and indeterminacy of the natural world, starting with the dependence of the genetic material itself on the functioning of the cell which holds it and extending to the nature of matter and of life. The human reality that the law both encounters and is part of has at least the virtue of mimicking this infinitely complicated and labile universe, ultimately explicable but lacking any single determinant. The presently attractive evidence that there is a genetic component in certain forms of behaviour does not spell an end to responsibility or free will; but it does have implications both for the symbiosis of justice and mercy and for the dilemma of reform and containment.

In quite different ways, too, changes in society’s material potentiality are likely to impact on legal values. Thanks chiefly to medical advances we have a rapidly ageing population in the developed countries: people are living longer and having fewer children. The pattern of dependency in a generation’s time is going to place new and alarming pressures not only on ethical but on legal assumptions about something as fundamental – and as entrenched in our law – as the right to life. The courts have already determined that the allocation of interventionist medical resources, even where life is at stake, is ordinarily not justiciable; but how long will it be before a foundering medical service asks whether it can withdraw first medical attention and then nourishment from incurably demented geriatric patients? If it happens, will the courts do what they did (with distinction) in Tony Bland’s case and decide something because Parliament would decide nothing? If they do, by what legal and ethical standards will they answer the question? It is one of history’s ironies that, having now put the judicial taking of life behind us, the law’s ability to sanction the taking, or more urgently the non-prolongation, of life by others is likely to come dramatically to the fore. And less dramatically too, for example in relation to the enforcement of living wills: who is to know whether any one of us, signing away our continued existence in the event of incurable degenerative illness at a time when we are fit and confident, would have been saying the same when that point was reached? In matters of life and death the law has a fraught journey ahead which is going to jolt our notions of justice.

And not just ahead: homelessness is here and now. One of the things which is not to be found in the European Convention on Human Rights is a right to shelter. The ideal of negative liberty has little to offer the homeless: they are as unconstrained by law as the next person, and so long as nobody is interfering with them they are free. If, however, you ask the question which the common law – and almost any model of justice I can think of – would ask: free to do what? the answer is almost nothing. Possibly they are encouraged by the fact that an Act of Parliament now protects their right to hold up a piece of card saying ‘Hungry and Homeless’, for this is the right of free expression protected by Article 10 of the Convention. But beyond this the homeless are, as Jeremy Waldron has said, ‘comprehensively unfree’. Their legal liberty to seek work and to rent accommodation has reached a dead end, and they are free only to sleep rough. Why should it matter to them that they live under the rule of law? One of the enduring images from my early days in metropolitan magistrates’ courts is the down-and-out who has thrown a brick through a shop window on Christmas Eve so that he can at least be housed and fed in custody on Christmas Day.

This kind of Dickensian sentiment, true, is easy enough to conjure up. It has to do, no doubt, with social justice, but does it have anything to do with the justice administered by the courts? In my view it does, for neither justice nor law necessarily requires a formal foundation in tabulated rights. The pure economic liberal view, indeed, is that rights and wrongs do not come into the picture at all: Hayek, for example, argues that unless the creation of misery is deliberate, poverty involves no injustice. And from a communitarian standpoint it must equally be the case that a practical limit exists to what rights the courts can enforce against the state, especially where the right to litigate belongs only to individuals. A successful legal claim – for recognition of special educational or physical needs, for instance – may simply mean reallocating already inadequate funds within a ring-fenced local budget, so that the silent pay for the gains of the assertive.

If injustice there is, then, it lies deeper than the allocation or denial of rights – dependence on which, as my colleague John Laws has argued, is a sign of an immature society. It has to do with a common sense of equity, an ethic of kindness, a morality of feeling, which does not and cannot be expected to stop at a desire for legal justice, even though that is necessarily where the law itself must stop. But within the law’s necessary limits I see nothing wrong with the reaction of a good judge of recent years, Mr Justice McKenna, when in my early years at the Bar he was asked to grant injunctions ordering some travellers to leave a roadside verge to which they had been forcibly removed from other land. ‘Where are they to go?’ he asked. ‘That’s not our concern,’ said the local authority’s counsel. ‘We’re entitled to an order.’ ‘These are human beings,’ McKenna said. ‘And you’re not getting any order until you can tell me where they are to go.’ It may be beyond the power of the courts to change a world in which privation in the midst of plenty is possible; but that does not mean that they are obliged to endorse all the consequent injustices.