skip to content

It is an immense honour and privilege to give the Sixth Sir David Williams Lecture. It is also a formidable challenge, since Sir David's scholarly reputation is so high as to discourage comparison. But the great range of his achievement - as legal scholar, university leader, head of house, public servant and loyal son of Wales - gives the lecturer a broad range of subject matter from which to choose, without straying into fields Sir David has not adorned. In choosing to address the Rule of Law - a big subject for a lecture - my best hope must be that Sir David will himself be provoked into giving us, at greater length, his considered reflections on the subject.

The Constitutional Reform Act 2005 provides, in section 1, that the Act does not adversely affect "the existing constitutional principle of the rule of law" or "the Lord Chancellor's existing constitutional role in relation to that principle". This provision, the Attorney-General has suggested, illustrates the importance attached to the rule of law in the modern age, which is further reflected in the oath to be taken by Lord Chancellors under section 17(1) of the Act, to respect the rule of law and defend the independence of the judiciary. But the Act does not define the existing constitutional principle of the rule of law, or the Lord Chancellor's existing constitutional role in relation to it.

The meaning of this existing constitutional principle may no doubt have been thought to be too clear and well-understood to call for statutory definition, and it is true that the rule of law has been routinely invoked by judges in their judgments. But they have not explained what they meant by the expression, and well-respected authors have thrown doubt on its meaning and value. Thus Joseph Raz has commented on the tendency to use the rule of law as a shorthand description of the positive aspects of any given political system. John Finnis has described the rule of law as "[t]he name commonly given to the state of affairs in which a legal system is legally in good shape". Judith Shklar has suggested that the expression may have become meaningless thanks to ideological abuse and general over-use:

"It may well have become just another one of those self-congratulatory rhetorical devices that grace the public utterances of Anglo-American politicians. No intellectual effort need therefore be wasted on this bit of ruling-class chatter".

Jeremy Waldron, commenting on Bush v Gore in which the rule of law was invoked on both sides, recognised a widespread impression that utterance of those magic words meant little more than "Hooray for our side!". Brian Tamanaha has described the rule of law as "an exceedingly elusive notion" giving rise to a "rampant divergence of understandings" and analogous to the notion of the Good in the sense that "everyone is for it, but have contrasting convictions about what it is". The authors of the 2005 Act may or may not have known of these critical academic opinions. But they can scarcely have been unaware that Dicey's exposition of the rule of law, first propounded in 1885, had attracted considerable controversy over the years which had elapsed since then. So it seems unlikely that the meaning of the existing constitutional principle was thought so clear as to obviate the need for definition.

It is perhaps more likely that the authors of the 2005 Act recognised the extreme difficulty of formulating a succinct and accurate definition suitable for inclusion in a statute, and preferred to leave the task of definition to the courts if and when occasion arose. If so, one has considerable sympathy with that view, the more so since the meaning of the concept has to some extent evolved over time and is no doubt likely to continue to do so. But the statutory affirmation of the rule of law as an existing constitutional principle and of the Lord Chancellor's existing role in relation to it does have an important consequence: that the judges, in their role as journeymen judgment-makers, are not free to dismiss the rule of law as meaningless verbiage, the jurisprudential equivalent of motherhood and apple pie, even if they were inclined to do so. They would be bound to construe a statute so that it did not infringe an existing constitutional principle, if it were reasonably possible to do so. And the Lord Chancellor's conduct in relation to that principle would no doubt be susceptible, in principle, to judicial review. So it is not perhaps premature to attempt to define what, in this country, today, is meant by the existing constitutional principle of the rule of law, recognising of course - as a serving judge necessarily must - that any thoughts he proffers may wilt or die in the light of future adversarial argument in a concrete case.

The core of the existing principle is, I suggest, that all persons and authorities within the state, whether public or private, should be bound by and entitled to the benefit of laws publicly and prospectively promulgated and publicly administered in the courts. I doubt if anyone would suggest that this statement, even if accurate as one of general principle, could be applied without exception or qualification. There are, for instance, some proceedings in which justice can only be done if they are not in public. But it seems to me that any derogation calls for close consideration and clear justification. And I think that this formulation, of course owing much to Dicey, expresses the fundamental truth propounded by John Locke in 1690 that "Where-ever law ends, tyranny begins", and also that famously stated by Thomas Paine in 1776,

"that in America THE LAW IS KING. For as in absolute governments the King is law, so in free countries the law ought to be King; and there ought to be no other."

But I do not think the scope of the existing principle can be adequately understood without examining its implications, which may be conveniently broken down into a series of sub-rules. I have identified eight such rules, which I shall briefly discuss. There is regrettably little to startle in any of them. More ingenious minds could doubtless propound additional and better sub-rules, or economise with fewer.

First, the law must be accessible and so far as possible intelligible, clear and predictable. This seems obvious: if everyone is bound by the law they must be able without undue difficulty to find out what it is, even if that means taking advice (as it usually will), and the answer when given should be sufficiently clear that a course of action can be based on it. There is English authority to this effect, and the European Court of Human Rights has also put the point very explicitly:

"… the law must be adequately accessible: the citizen must be able to have an indication that is adequate in the circumstances of the legal rules applicable to a given case … a norm cannot be regarded as a 'law' unless it is formulated with sufficient precision to enable the citizen to regulate his conduct: he must be able - if need be with appropriate advice - to foresee, to a degree that is reasonable in the circumstances, the consequences which a given action may entail."

Obvious this point is, but not, I think, trivial. Given the legislative hyperactivity which appears to have become a permanent feature of our governance - in 2004, some 3500 pages of primary legislation; in 2003, nearly 9000 pages of statutory instruments - the sheer volume of current legislation raises serious problems of accessibility, despite the internet. And this is compounded by the British tradition of parliamentary draftsmanship which, for all its technical virtuosity, depends so heavily on cross-reference and incorporation as on occasion to baffle.

The accusing finger cannot however be fairly pointed at legislators alone: the length, complexity and sometimes prolixity of modern common law judgments, particularly at the highest level, raise problems of their own. These problems could, at least in theory, be mitigated if the House of Lords were to give a single opinion, a solution advocated from time to time and raised with me by the late Lord Brightman, very shortly before he died, with reference to the lengthy opinions of the House in R (Jackson) v Attorney General. This is a serious argument, recently echoed by Chief Justice Roberts of the United States in an address to the American College of Trial Lawyers, but not one which I would in general accept, for very much the reasons given by Lord Reid, addressing the Society of Public Teachers of Law, in 1971. I agree with Lord Reid that the quality of single Privy Council judgments has on the whole been inferior from the point of view of developing the law to the more diverse opinions of the House. A single lapidary judgment buttressed by four brief concurrences can give rise to continuing problems of interpretation which would have been at least reduced if the other members had summarised, however briefly, their reasons for agreeing. And a well-constituted committee of five or more, can bring to bear a diversity of professional and jurisdictional experience which is valuable in shaping the law.

But I would add three important caveats. First, whatever the diversity of opinion the judges should recognise a duty, not always observed, to try to ensure that there is a clear majority ratio. Without that, no one can know what the law is until Parliament or a later case lays down a clear rule. Secondly, and without challenging the value or legitimacy of judicial development of the law, the sub-rule under consideration does in my view preclude excessive innovation and adventurism by the judges. It is one thing to alter the law's direction of travel by a few degrees, quite another to set it off in a different direction. The one is probably foreseeable and predictable, something a prudent person would allow for, the other not. Thus one can agree with Justice Heydon of the High Court of Australia that judicial activism, taken to extremes, can spell the death of the rule of law. But thirdly, and importantly, all these points apply with redoubled force in the criminal field. The torrent of criminal legislation in recent years has posed very real problems of assimilation. Not all of this legislation is readily intelligible. Whether derived from statute or judicial opinion the law must be stated in terms which a judge can without undue difficulty explain to a jury or an unqualified clerk to a bench of lay justices. And the judges may not develop the law to create new offences or widen existing offences so as to make punishable conduct of a type hitherto not subject to punishment, for that would infringe the fundamental principle that a person should not be criminally punishable for an act not proscribed as criminal when the act was done.

My second sub-rule is that questions of legal right and liability should ordinarily be resolved by application of the law and not the exercise of discretion. Most modern commentators would not share to the full Dicey's hostility to the exercise of official discretions. In the immigration field, for example, judges have routinely and gratefully invited the Secretary of State to exercise his discretion to grant leave to enter or remain to applicants who do not meet the tests for entry laid down in the immigration rules but whose personal history or circumstances demand sympathetic consideration. But the essential truth of Dicey's insight stands. The broader and more loosely-textured a discretion is, whether conferred on an official or a judge, the greater the scope for subjectivity and hence for arbitrariness, which is the antithesis of the rule of law. This sub-rule requires that a discretion should ordinarily be narrowly defined and its exercise capable of reasoned justification. These are requirements which our law, in my opinion, almost always satisfies, because discretion imports a choice between two possible decisions and orders, and usually the scope for choice is very restricted.

There can, first of all, be no discretion as to the facts on which a decision-maker, official or judicial, proceeds. An assessment of the facts may of course be necessary and will depend on the effect made by the evidence on the mind of the decision-maker. The assessment made may be correct or it may not, but if the evidence leads the decision-maker to one conclusion he has no discretion to reach another, any more than a historian has a discretion to conclude that King John did not execute Magna Carta at Runnymede in June 1215 when all the evidence shows that he did. Similarly, most so-called discretions depend on the making of a prior judgment which, once made, effectively determines the course to be followed, and leaves no room for choice. Even the least constrained of judicial discretions - that as to the award of costs - is governed by principles and practice. I take three examples, two judicial and one official.

The grant of a civil injunction, it is always said, is discretionary. But if a clear violation of legal right is shown, and there is a clear risk of repetition injurious to the victim for which damages will not compensate, and there is no undertaking by the lawbreaker to desist, the trial judge ordinarily has no choice. His discretion can only, usually, be exercised one way. A second, very familiar, example is found in section 78(1) of the Police and Criminal Evidence Act 1984, which provides:

"In any proceedings the court may refuse to allow evidence on which the prosecution proposes to rely to be given if it appears to the court that, having regard to all the circumstances, including the circumstances in which the evidence was obtained, the admission of the evidence would have such an adverse effect on the fairness of the proceedings that the court ought not to admit it."

The use of the word "may" is relied on as conferring a discretion. But what the subsection does, I suggest, is to require an exercise of judgment, which may be difficult to make but which will determine the outcome: if the statutory condition is judged to be satisfied, the judge must refuse to allow the evidence to be given; if it is not, the subsection does not authorise the judge to exclude the evidence. For my third illustrative example I return to the immigration field. If an official were to grant leave to enter or remain to a person who did not meet the tests laid down in the immigration rules, but whose case presented no exceptional features whatever suggesting the need for special treatment, such decision would be incapable of rational justification and could not be defended as an exercise of discretion. There is in truth no such thing as an unfettered discretion, judicial or official, and that is what the rule of law requires.

My third sub-rule is that the laws of the land should apply equally to all, save to the extent that objective differences justify differentiation. I doubt if this would strike a modern audience as doubtful. While some special legislative provision can properly be made for some categories of people such as children, prisoners and the mentally ill, based on the peculiar characteristics of such categories, we would regard legislation directed to those with red hair (to adapt Warrington LJ's long-lived example) as incompatible with the rule of law. Even more obviously incompatible would be the statute 22 Henry 8 cap 9 which convicted Richard Rose, the Bishop of Rochester's cook, who had not been tried, of high treason: he had put poison into the porridge in the bishop's kitchen, and the statute ordered that he be boiled to death without having any advantage of his clergy. Other poisoners were to be similarly treated, but the statute was primarily aimed at him. In much more recent times our law not only tolerated but imposed disabilities not rationally based on their religious beliefs on Roman Catholics, Dissenters and Jews, and disabilities not rationally connected with any aspect of their gender on women.

It would be comforting to treat this sub-rule as of antiquarian interest only. But it would be unrealistic, as the treatment of non-nationals here and elsewhere reveals. The position of a non-national with no right of abode in this country differs from that of a national with a right of abode in the obvious and important respect that the one is subject to removal and the other is not. That is the crucial distinction, and differentiation relevant to it is unobjectionable and indeed inevitable. But it does not warrant differentiation irrelevant to that distinction, as Lord Scarman made clear in R v Secretary of State for the Home Department, Ex p Khawaja:

"Habeas corpus protection is often expressed as limited to 'British subjects'. Is it really limited to British nationals? Suffice it to say that the case law has given an emphatic 'no' to the question. Every person within the jurisdiction enjoys the equal protection of our laws. There is no distinction between British nationals and others. He who is subject to English law is entitled to its protection. This principle has been in the law at least since Lord Mansfield freed 'the black' in Sommersett's Case (1772) 20 St. Tr. 1. There is nothing here to encourage in the case of aliens or non-patrials the implication of words excluding the judicial review our law normally accords to those whose liberty is infringed."

This message seems clear enough. But it did not deter Parliament from providing, in Part 4 of the Anti-terrorism, Crime and Security Act 2001, for the indefinite detention without charge or trial of non-nationals suspected of international terrorism while exempting from that liability nationals who were judged qualitatively to present the same threat. The record of the United States in this respect is not better than our own, and arguably worse. As an American academic author has written,

"Virtually every significant government security initiative implicating civil liberties - including penalizing speech, ethnic profiling, guilt by association, the use of administrative measures to avoid the safeguards of the criminal process, and preventive detention - has originated in a measure targeted at noncitizens."

There is, I think, profound truth in the observation of Justice Jackson in the Supreme Court of the United States in 1949:

"I regard it as a salutary doctrine that cities, states and the Federal Government must exercise their powers so as not to discriminate between their inhabitants except upon some reasonable differentiation fairly related to the object of regulation. This equality is not merely abstract justice. The framers of the Constitution knew, and we should not forget today, that there is no more effective practical guaranty against arbitrary and unreasonable government than to require that the principles of law which officials would impose upon a minority must be imposed generally. Conversely, nothing opens the door to arbitrary action so effectively as to allow those officials to pick and choose only a few to whom they will apply legislation and thus to escape the political retribution that might be visited upon them if larger numbers were affected. Courts can take no better measure to assure that laws will be just than to require that laws be equal in operation."

Sixty years on we may say that this is not merely a salutary doctrine but a pillar of the rule of law itself.

I turn to my fourth sub-rule, which is that the law must afford adequate protection of fundamental human rights. This would not be universally accepted as embraced within the rule of law. Dicey, it has been argued, gave no such substantive content to his rule of law concept. Professor Raz has written:

"A non-democratic legal system, based on the denial of human rights, on extensive poverty, on racial segregation, sexual inequalities, and racial persecution may, in principle, conform to the requirements of the rule of law better than any of the legal systems of the more enlightened Western democracies … It will be an immeasurably worse legal system, but it will excel in one respect: in its conformity to the rule of law … The law may … institute slavery without violating the rule of law."

On the other hand, as Geoffrey Marshall has pointed out, chapters V to XII of Dicey's Introduction to the Law of the Constitution in which he discusses what would now be called civil liberties, appear within part II of the book entitled "The Rule of Law", and, as Marshall observes, "the reader could be forgiven for thinking that Dicey intended them to form part of an account of what the rule of law meant for Englishmen." The preamble to the Universal Declaration of Human Rights 1948 recites that "it is essential, if man is not to be compelled to have recourse, as a last resort, to rebellion against tyranny and oppression, that human rights should be protected by the rule of law." The European Court of Human Rights has referred to "the notion of the rule of law from which the whole Convention draws its inspiration." The European Commission has consistently treated democratisation, the rule of law, respect for human rights and good governance as inseparably interlinked.

While, therefore, I recognise the logical force of Professor Raz's contention, I would not myself accept it. A state which savagely repressed or persecuted sections of its people could not in my view be regarded as observing the rule of law, even if the transport of the persecuted minority to the concentration camp or the compulsory exposure of female children on the mountainside were the subject of detailed laws duly enacted and scrupulously observed. So to hold would, I think, be to strip the existing constitutional principle affirmed by section 1 of the 2005 Act of much of its virtue and infringe the fundamental compact which, as I shall suggest at the end, underpins the rule of law. But this is a difficult area, for I would agree with Professor Jowell that the rule of law

"does not, for example, address the full range of freedoms protected by bills of rights in other countries or in international instruments of human rights, or those now protected by our recently enacted Human Rights Act 1998, as set out in the European Convention on Human Rights (such as the right not to suffer torture, or the right to freedom of expression or rights of privacy or sexual freedom)."

There is not, after all, a standard of human rights universally agreed even among civilised nations. We may regret the United States' failure to ratify the UN Convention on the Rights of the Child 1989, which forbids the imposition of capital punishment for offences committed by persons under 18, and the Supreme Court's decision upholding the imposition of capital punishment for a murder committed at the age of 16½, but accession to any international convention is a matter of national choice, and different countries take different views on the morality as well as the efficacy of the death penalty. It is open to a state to acknowledge, as some have, that a penalty is cruel and unusual treatment or punishment within the meaning of its Constitution, and nonetheless to assert that it is authorised by that Constitution as lawful. There is, I would accept, an element of vagueness about the content of this sub-rule, since the outer edges of fundamental human rights are not clear-cut. But within a given state there will ordinarily be a measure of agreement on where the lines are to be drawn, and in the last resort (subject in this country to statute) the courts are there to draw them. The rule of law must, surely, require legal protection of such human rights as, within that society, are seen as fundamental.

My fifth sub-rule is that means must be provided for resolving, without prohibitive cost or inordinate delay, bona fide civil disputes which the parties themselves are unable to resolve. It would seem to be an obvious corollary of the principle that everyone is bound by and entitled to the benefit of the law that people should be able, in the last resort, to go to court to have their rights and liabilities determined. This is not a rule directed against arbitration and more informal means of dispute resolution, all of which, properly resorted to and fairly conducted, have a supremely important contribution to make to the rule of law. Nor is it a rule requiring every claim or defence, however spurious and lacking merit, to be guaranteed full access to the process of the law. What it does is to recognise the right of unimpeded access to a court as a basic right, protected by our own domestic law, and in my view comprised within the principle of the rule of law. If that is accepted, then the question must be faced: how is the poor man or woman to be enabled to assert his or her rights at law? Assuming, as I would certainly wish to do, the existence of a free and independent legal profession, the obtaining of legal advice and representation is bound to have a cost, and since legal services absorb much professional time they are inevitably expensive. The old gibe about the Ritz Hotel is one that cannot be ignored.

For many years, of course, this problem was addressed through the civil Legal Aid scheme established in 1948, a bold, imaginative and somewhat under-celebrated reform of the Attlee post-war government. Although subject to well-known defects, the scheme did bring legal redress within reach of the less well-off. But, as we know, the cost of the scheme rose exponentially, and this led to its curtailment. Whether conditional fees, various pro bono schemes and small claims procedures have filled the gap left by this curtailment I do not myself know. Perhaps they have, and advice and help are still available to those of modest means who deserve it. But I have a fear that tabloid tales of practitioners milking the criminal legal aid fund of millions, and more general distrust of lawyers and their rewards, may have enabled a valuable guarantee of social justice to wither unlamented.

Lurking in the background is another point which, at a certain point, must concern adherents of the rule of law. Successive British governments have insisted that the civil courts, judicial salaries usually aside, should be self-financing: the cost of running the courts should be covered by fees recovered from litigants. The judges for their part have accepted that those using the courts may generally be called on to contribute specifically to the cost of the service - resort to the law being less universal than that to the doctor - but have never, I think, accepted the full recovery principle, regarding the provision of courts as one of the essential functions of a liberal democratic state. The danger again is that the cost of obtaining redress may lead to its being denied to some at least of those who need it. The rule of law plainly requires that legal redress should be an affordable commodity. That it should also be available without excessive delay is so obvious as to make any elaboration unnecessary.

My sixth sub-rule expresses what many would, with reason, regard as the core of the rule of law principle. It is that ministers and public officers at all levels must exercise the powers conferred on them reasonably, in good faith, for the purpose for which the powers were conferred and without exceeding the limits of such powers. This sub-rule reflects the well-established and familiar grounds of judicial review. It is indeed fundamental. For although the citizens of a democracy empower their representative institutions to make laws which, duly made, bind all to whom they apply, and it falls to the executive, the government of the day, to carry those laws into effect, nothing ordinarily authorises the executive to act otherwise than in strict accordance with those laws. (I say "ordinarily" to acknowledge the survival of a shrinking body of unreviewable prerogative powers). The historic role of the courts has of course been to check excesses of executive power, a role greatly expanded in recent years due to the increased complexity of government and the greater willingness of the public to challenge governmental (in the broadest sense) decisions. Even under our constitution the separation of powers is crucial in guaranteeing the integrity of the courts' performance of this role.

The British Government, through one entity or another, is very frequently involved in litigation. It is usually successful, but not invariably so. When unsuccessful it is displeased, being driven like every other litigant by a belief in the rightness of its cause but also no doubt by a belief that the public interest is best served by its succeeding. In the past the convention was that ministers, however critical of a judicial decision, and exercising their right to appeal against it or, in the last resort, legislate to reverse it retrospectively, forebore from public disparagement of it. This convention appears to have worn a little thin in recent times, as I think unfortunately, since if ministers make what are understood to be public attacks on judges, the judges may be provoked to make similar criticisms of ministers, and the rule of law is not, in my view, well served by public dispute between two arms of the state. Some sections of the press, with their gift for understatement, have spoken of open war between the government and the judiciary. This is not in my view an accurate analysis. But there is an inevitable, and in my view entirely proper, tension between the two. There are countries in the world where all judicial decisions find favour with the government, but they are not places where one would wish to live. Such tension exists even in quiet times. But it is greater at times of perceived threats to national security, since governments understandably go to the very limit of what they believe to be their lawful powers to protect the public, and the duty of the judges to require that they go no further must be performed if the rule of law is to be observed. This is a fraught area, since history suggests that in times of crisis governments have tended to overreact and the courts to prove somewhat ineffective watchdogs. In our country and in the United States, decisions have been made of which neither country can be proud. The cautionary words of Justice William Brennan of the United States Supreme Court in 1987 remain pertinent:

"There is considerably less to be proud about, and a good deal to be embarrassed about, when one reflects on the shabby treatment civil liberties have received in the United States during times of war and perceived threats to national security … After each perceived security crisis ended, the United States has remorsefully realized that the abrogation of civil liberties was unnecessary. But it has proven unable to prevent itself from repeating the error when the next crisis came along."

So to my seventh and penultimate sub-rule: that adjudicative procedures provided by the state should be fair. The rule of law would seem to require no less. The general arguments in favour of open hearings are familiar, summed up on this side of the Atlantic by the dictum that justice must manifestly and undoubtedly be seen to be done and on the American side by the observation that "Democracies die behind closed doors."

Application of this sub-rule to ordinary civil processes is largely unproblematical, once it is remembered that not all decisions are purely judicial. As the Chief Justice of Australia has pointed out, "the rule of law does not mean rule by lawyers."

There is more scope for difficulty where a person faces adverse consequences as a result of what he is thought or said to have done or not done, whether in the context of a formal criminal charge or in other contexts such as deportation, precautionary detention, recall to prison or refusal of parole. What in such contexts does fairness ordinarily require? First and foremost, I suggest, that decisions are made by adjudicators who, however described, are independent and impartial: independent in the sense that they are free to decide on the legal and factual merits of a case as they see it, free of any extraneous influence or pressure, and impartial in the sense that they are, so far as humanly possible, open-minded, unbiased by any personal interest or partisan allegiance of any kind. In addition, certain core principles have come to be accepted: that a matter should not be finally decided against any party until he has had an adequate opportunity to be heard; that a person potentially subject to any liability or penalty should be adequately informed of what is said against him; that the accuser should make adequate disclosure of material helpful to the other party or damaging to itself; that where the interests of a party cannot be adequately protected without the benefit of professional help which the party cannot afford, public assistance should so far as practicable be afforded; that a party accused should have an adequate opportunity to prepare his answer to what is said against him; and that the innocence of a defendant charged with criminal conduct should be presumed until guilt is proved.

In the strictly criminal context, two of these principles in particular have raised questions. The first concerns disclosure, since the prosecutor may be in possession of material which he is for public interest reasons unwilling or very reluctant to disclose to the defence. This problem, when it arises, calls for the exercise of very careful judgment by the trial judge. But as the law now stands, material need not be disclosed if in no way helpful to the defence; if helpful to the point where the defence would be significantly prejudiced by non-disclosure, the prosecutor must either disclose or abandon the prosecution. Questions have also arisen concerning statutory offences defined so as to place a reverse burden on the defendant. These are not in themselves objectionable, but may be so if the burden is one which a defendant, even if innocent, may in practice be unable to discharge. I do not think these solutions, even if not ideal, infringe the rule of law.

More disturbing are the growing categories of case outside the strictly criminal sphere in which Parliament has provided that the full case against a person, put before the adjudicator as a basis for decision, should not be disclosed to that person or to any legal representative authorised by that person to represent him. Any process which denies knowledge to a person effectively, if not actually, accused of what is relied on against him, and thus denies him a fair opportunity to rebut it, must arouse acute disquiet. But these categories reflect the undoubted danger of disclosing some kinds of highly sensitive information, and they have been clearly identified and regulated by Parliament, which has judged the departure to be necessary and attempted to limit its extent. Resort to a similar procedure by the Parole Board in the absence of any express statutory authority provoked a division of opinion in the House of Lords: a majority upheld the practice; a minority (which included myself) strongly dissented. There the matter rests. All would probably agree that this is difficult terrain.

My eighth and last sub-rule is that the existing principle of the rule of law requires compliance by the state with its obligations in international law, the law which whether deriving from treaty or international custom and practice governs the conduct of nations. I do not think this proposition is contentious. Addressing a joint session of Congress in September 1990 after the Iraqi invasion of Kuwait, the first President Bush said that a new world was emerging,

"a world where the rule of law supplants the rule of the jungle. A world in which nations recognize the shared responsibility for freedom and justice. A world where the strong respect the rights of the weak … America and the world must support the rule of law. And we will."

President George W Bush, in his State of the Union Address of 29 January 2002, speaking of the international, not the domestic, scene, echoed the same sentiment:

"But America will always stand firm for the non-negotiable demands of human dignity: the rule of law; limits on the power of the state; respect for women; private property; free speech; equal justice; and religious tolerance."

British statesmen today would, I think, share this belief.

It was not always so. On the outbreak of war in 1914 the German Chancellor told the Reichstag:

"Gentlemen, we are in a state of necessity, and necessity knows no law. Our troops have occupied Luxembourg and perhaps are already on Belgian soil. That is contrary to international law. The wrong we thus commit we will endeavour to repair directly our military aim is achieved."

Defending the British blockade of Germany, known to be of doubtful legality, Mr Asquith was not deterred by legal considerations:

"In dealing with an opponent who has openly repudiated all the restraints, both of law and of humanity, we are not going to allow our efforts to be strangled in a network of juridical niceties."

I shall not for obvious reasons touch on the vexed question whether Britain's involvement in the 2003 war on Iraq was in breach of international law and thus, if this sub-rule is sound, of the rule of law. But a revealing comparison may be made between the procedures followed in 2003 and those followed at the time of the Suez invasion of 1956, and the comparison does suggest that over that period the rule of law has indeed gained ground in this country and the law of the jungle lost it. First, Sir Anthony Eden, prime minister in 1956, appears to have treated legal considerations as at best peripheral. Echoing Asquith, but with much less justification, he said:

"We should not allow ourselves to become involved in legal quibbles about the rights of the Egyptian Government to nationalise what is technically an Egyptian company …"

At a later stage of the crisis, the prime minister specifically instructed that Sir Gerald Fitzmaurice, the very distinguished Legal Adviser to the Foreign Office, who had strongly and consistently advised that the British action was unlawful, should not be informed of developments: "Fitz is the last person I want consulted," he said. "The lawyers are always against our doing anything. For God's sake, keep them out of it. This is a political affair." So far as I know, no similar sentiments were ever expressed by Mr Blair.

Secondly, and although in 1956 as in 2003 it was the function of the Law Officers to tender legal advice to the Government, in 1956 they were never formally consulted before the ultimatum to Egypt was delivered. Instead, the Government relied on the advice of the Lord Chancellor, who in turn relied on an ambiguous footnote in an article by Professor Waldock, on which, however, Waldock was never approached. In 2003, so far as is known, the Lord Chancellor did not give a legal opinion on the lawfulness of war but the Attorney General made a brief public statement on the eve of war, and some two years later his more detailed earlier opinion reached the public domain. There seems to me to be room to question whether the ordinary rules of client privilege, appropriate enough in other circumstances, should apply to a law officer's opinion on the lawfulness of war: it is not unrealistic in my view to regard the public, those who are to fight and perhaps die, rather than the government, as the client. If the government is sued for damages in negligence for (say) injuries caused by an army lorry or a mishap in a military hospital, I see no reason why the ordinary rules of client professional privilege should not apply. The government's position as a defendant would be greatly and unfairly weakened if this were not so. An opinion on the lawfulness of war, the ultimate exercise of sovereign power, involving the whole people, seems to me to be quite different. And the case for full, contemporaneous, disclosure seems to me even stronger when the Attorney General is a peer, not susceptible to direct questioning in the elected chamber. But this is not an accepted view, and we know that in 2003 the Attorney General's advice supported the proposed action.

Thirdly, and surprisingly with memories of Nuremberg relatively fresh, it does not appear that the service chiefs in 1956 called for any assurance that the invasion would be lawful. In 2003, as is well known, they did.

The fourth distinction is the most striking of all. Although not formally consulted on the lawfulness of the proposed intervention in 1956, the Law Officers learned what was in the air and expressed the clear view that it could not be justified in law. But despite this they supported the Government's action. Writing to the Prime Minister on 7 November 1956, the Attorney General, Sir Reginald Manningham-Buller QC, said "… I support and have supported the Government's actions though I cannot do so on legal grounds." After a meeting the next day he wrote again, on behalf of himself and the Solicitor General, Sir Harry Hylton-Foster QC: "Although I support what we have done and have said so publicly, we cannot, as you know, agree with the statements made on behalf of the Government that we were legally entitled so to act." Sir Harry also wrote to the Prime Minister saying: "Reflection has convinced me that I was wrong to allow legalistic considerations to weigh so heavily with me." It would no doubt be naïve to suppose that even today major democratic states do not on occasion resort to legal casuistry to justify the use of force in doubtful circumstances. But I do not think that, save perhaps in extremis, the government of such a state would embark on a course which it acknowledged to be blatantly unlawful, or that lawyers advising the government of such a state at a senior level would publicly support action for which they could find no legal justification. To do either would pay scant respect to the existing constitutional principle of the rule of law.

There has been much debate whether the rule of law can exist without democracy. Some have argued that it can. But it seems to me that the rule of law does depend on an unspoken but fundamental bargain between the individual and the state, the governed and the governor, by which both sacrifice a measure of the freedom and power which they would otherwise enjoy. The individual living in society implicitly accepts that he or she cannot exercise the unbridled freedom enjoyed by Adam in the Garden of Eden, before the creation of Eve, and accepts the constraints imposed by laws properly made because of the benefits which, on balance, they confer. The state for its part accepts that it may not do, at home or abroad, all that it has the power to do but only that which laws binding upon it authorise it to do. If correct, this conclusion is reassuring to all of us who, in any capacity, devote our professional lives to the service of the law. For it means that we are not, as we are sometimes seen, mere custodians of a body of arid prescriptive rules but are, with others, the guardians of an all but sacred flame which animates and enlightens the society in which we live.

The Rt. Hon Lord Bingham of Cornhill KG, House of Lords, 16th November 2006