Melbourne Journal of International Law
The following brief is published as it was submitted to the Fiji Court of Appeal. The only changes made to it are formatting changes to match the style of this journal. Any errors or omissions remain the authors’ own.
THE ATTORNEY GENERAL OF FIJI
OF SUVA, FIJI
CHANDRIKA PRASAD (F/N PRABHU)
 This action was brought by Chandrika Prasad, citizen of the republic of Fiji, because he had been deprived of his fundamental civil and political right to live at liberty under a constitution. By no act of law has the constitution and its attendant guarantees, vouchsafed to him in 1998 by a fully representative democratic process, been ‘removed’ or ‘abrogated’. He is not a member, much less a representative of any political party. He is an ordinary citizen, who has suffered greatly as his affidavit attests. But he seeks no redress in these proceedings for his physical suffering, but rather for his loss of dignity and humanity by being unlawfully stripped of the rights the constitution gives to live under the rule of law in a democracy, with guarantees for equality and fair treatment. Against the armed might of this ‘interim’ Goliath, he stood with but one slingshot - the integrity of justice in the court of Gates J. His victory in that first round is now appealed, on the basis of an argument that shouts ‘MIGHT MAKES RIGHT’ (and sotto voce ‘Don’t do justice, or the heavens will fall’). In these submissions, which will be further developed in oral argument on his behalf, he invites this appeal court to do justice, by affirming that NEITHER MIGHT, NOR TWO WRONGS, MAKE RIGHT.
 The Constitution of the Republic of the Fiji Islands was adopted on 25 July 1997 by Act No. 13 of 1997: it entered into force on 27 July 1988. It was the product of exhaustive and extensive consultation and deliberation involving all the people of the Fiji Islands, and it represents their sovereign will to be governed in accordance with the rules of this supreme law of their democratic state (see its Preamble, and Chapter 1). Its democratic and popular genesis was:
(i) Unanimous resolutions by the House of Representatives and the Senate in September and October 1993 to establish a constitutional review .
(ii) In 1994 both Houses established a Joint Parliamentary Select Committee on the Constitution. It recommended a Constitutional Review Commission, which was duly appointed on 15 March 1995 chaired by Sir Paul Reeves and with a distinguished government and opposition member.
(iii) The Reeves Commission held extensive public hearings throughout the islands and received submissions from all political parties and significant interest groups. Debate was vigorous and fully covered by the media. The report, ‘The Fiji Islands: Towards a United Future’ was tabled in Parliament in September 1996. (Parliamentary Paper No. 34).
(iv) No less than 5 Parliamentary sub-committees deliberated on the Report over the following 6 months, and the Joint Parliamentary Select Committee published its own conclusions in May 1997. (Parliamentary Paper No. 17).
(v) The Constitution Reform Bill, based on this paper, was debated in detail in June and on 3 July 1997 was passed by unanimous vote in the House of Representatives. It was approved by the Senate and assented to by the President on 25 July. (See C F Powles, The Constitution of 1997).
 In May 1999 elections were held under electoral boundaries drawn up by the Constituency Boundaries Commission, appointed by the previous Government. There were appropriate public awareness and education programmes and media advertising about the new system (based on the Australian preferential system) throughout the islands. (Vol.2, Rigamoto Affidavit, para. 7). The overall percentage of invalid votes was 8.6%, with no discernible differences between locations and classes of citizens. The preferential voting system is widely thought to be the fairest method of polling, and there is no suggestion that the election tun-out was poor or that the level of informal vote was statistically abnormal (indeed Australia, which has had the voting system for a century, has a comparable rate of 6%). See ‘Likely Causes’, Vol.II p16. The elections were characterised as ‘peaceful and democratic’ by the US State Department’s 1999 Country Report on Fiji.
 The election resulted in the Peoples Coalition forming Government with 58 out of 71 seats. The Prime Minister, Mahendra Chaudhry, and his cabinet governed for 11 months effectively and without serious incident. There was no decision by the courts that it had acted unconstitutionally, nor a single challenge to the constitutionality of any legislation. The Respondents allege (in their chronology but not by way of affidavit evidence) that there were three protest marches in April/May 2000 by the Taukei Movement - the first in Lautoka, handing in a petition to the President (April 20) then a peaceful march in Suva on 28 April and then a march of 15,000 (Respondent says 5,000) indigenous Fijians to hand a petition to the President, on 19 May. Such protests and petitions by interest groups are a normal incident of democracy. But it is alleged, by both the caretaking Prime Minister Qarase (affidavit, para. 4) and by the military Commander Bainimarama (affidavit, p.4-5) that many indigenous Fijians perceived the new Constitution;
(a) weakening positive discrimination in their favour;
(b) weakening basic Fijian legal rights; and
(c) creating an incomprehensible and unfair electoral system.
 The existence of such ‘perceptions’ is alleged without supporting evidence: in particular, there is no evidence that they were ‘widely’ shared, and the language of the Respondents is careful not to contend that the perceptions were accurate. So far as (c) is concerned, there can be no suggestion that preferential voting is ‘unfair’ or that it produced a democratically skewed result or that it disenfranchised indigenous Fijians. (See para. 2, above). So far as (a) is concerned, any changes wrought by the 1997 Constitution were the result of the democratic process described in para. 1 above, and were capable of amendment in the manner prescribed by the Constitution. It is not even clear that the provisions vaguely referred to by the Appellants have the interpretation which was ‘perceived’: the courts, if asked to construe them, would have applied Section 6 principle a), h), i) and j). So far as b) is concerned, it is notable that the deponents do not refer to any particular Act passed by parliament, or explain why such Acts were not the subject of lawful constitutional challenge in the courts.
 It will be necessary to revisit these ‘perceptions’ in so far as they are advanced by the military Commander to legitimise his purported ‘removal’ of the Constitution. However, it may be noted at this point that none of those assumed grievances were irremediable: they reflected political interests which were capable of resolution through the democratic process, either by Parliament or by electing a new Government or by action in the courts. The Appellants’ fresh evidence shows merely that there was normal political opposition to the Chaudhry government, and there was a faction of indeterminate size which had come to regret support for the 1997 Constitution, but it cannot possibly begin to found an argument for the ‘necessity’ to abrogate the Constitution.
 On 19 May 2000 a group of armed insurgents led by George Speight entered the Parliament, assaulted the prime minister and held him and other cabinet ministers and MP’s as hostages. They demanded ‘abolition of the democratic provisions of the constitution as presently framed and assertion of indigenous Fijian supremacy’ (the salient facts, found by Gates J at pp.17-20 of his judgment, are virtually uncontested by the Appellants’ evidence). On 20 May, Speight purported to abrogate the Constitution and thereafter pretended to take over the reins of Government. The President, who had immediately declared a state of emergency under the Public Safety Act, on 27 May appointed an Acting Prime Minister (who immediately resigned) and purported (so it is alleged - the evidence is still lacking) to dismiss the government. He prorogued parliament for six months, (i.e. until 27 November 2000).
 On 28 May, the President was taken on board a naval vessel for his own safety. It would appear (from the Qarese letter re pension options for the former President) that the President may have ‘retired’ on 28 May. On 29 May the Police Commissioner wrote to the President seeking military assistance, and on that all-important day the Commander of the Fiji military forces took control of the country. His affidavit merely records (para. 2) that:
Between 29 May and 13 July 2000 I served also as Head of State … while acting in that capacity I abrogated the Constitution.
In another affidavit, relied on by Gates J the Commander said that on 29 May he ‘assumed executive authority from the President as Commander and Head of the Interim Military Government of Fiji’, because of ‘the absence of any other viable alternative’. In this self-styled capacity, the Military Commander purported to issue a decree (IMG Decree No. 1) that:
Notwithstanding the provisions of any law existing … the Fiji Constitution Amendment Act 1997 is with effect from 29th Day of May 2000, wholly removed.
IMG Decree No. 2 of the same day purported to save all existing laws, and Decree No. 3 (which notably came after the purported abrogation) purported to establish his authority for the IMG. Gates J found that the Commander had issued these decrees, as he had said in an affidavit, ‘in an effort to restore normality and to ensure the safety of the hostages’. In his affidavit in these proceedings, the Commander has said that he issued decree No. 1 because the continuance of the 1997 constitution threatened the peace through its hostile perception by ‘many indigenous Fijians’ who were ‘unsophisticated’ and ‘not equipped to adequately comprehend the niceties and technicalities of the Constitution’. He states (para. 5) that:
I abrogated the 1997 constitution only (our italics) because I had become satisfied that it was these perceptions which animated those engaging in the events of May 19, 2000 [the ‘events’ being the criminal actions of the Speight group].
 The executive actions on 19 and 27 - 29 May were carefully and correctly analysed by Gates J. The antics of the Speight group, which aped constitutional process, were plainly unlawful and of no effect. The constitutional provisions for declaration of a state of emergency were impossible to fulfil (because the cabinet and a majority of MP’s were being held hostage) but the doctrine of necessity plainly legitimised the declaration of a state of emergency (Gates J, p.23). Thereafter, the doctrine legitimised such actions taken by the civil authorities as were ‘transient and proportionate response to the crisis’ (ibid. p.25). The President, as Head of State and Commander-in-Chief of the armed forces had a duty to act so as to restore order and uphold the supreme law. As sole executive authority, his duty was to direct the army and police to secure the release of the hostages, and his assumption of emergency powers to deal with a situation unforeseen by the provisions of section 187 of the Constitution were valid to the extent (and no more) that they were calculated to secure the release of the hostages and the restoration of constitutional government.
 The actions of the President on 27 May remain obscure. The Appellants’ chronology account is not accepted. The crisis might well have justified the appointment of a free cabinet member as an ‘acting’ Prime Minister, although this was apparently a device which was thought to permit the 6 month prorogation of Parliament – an act (like any ‘dismissal’ of the Government) which could not rationally be related to ending the hostage crisis, then in its eighth day. Gates J held a) that these acts were not properly evidenced and b) would in any event have been within the President’s ‘necessity’ powers but c) were not relevant to the determination of the issues in the case. The Respondent endorses a) but is doubtful of b). It contends that c) is plainly correct: even if these actions were lawful - and they were the last acts of the lawful President prior to his ‘retirement’ - they should have led to the summoning of Parliament on 27 November 2000, convening under the 1997 Constitution. These presidential actions, if done, in no way abrogated the Constitution, but were designed to restore representative constitutional government after a period within which it must have been assumed that the hostage crisis would have been resolved.
 The evidence indicates that on 28 or 29 May the President ‘relinquished’ his positions as Head of State and Commander-in-Chief of the military forces. On the office becoming vacant, there is no power vacuum or ‘absence of any viable alternative’: the office descends to the Vice-President and then to the Speaker of the House of Representatives under section 88 of the Constitution. It was not therefore possible, under the Supreme law extant at that time, for Commander Bainimarama to become the Head of State, let alone to issue subsequently a decree ‘removing’ a constitution which had barred him from occupying this position (and its the only position) which he claims entitled him to issue the decree. He was not, on 29 May, the Head of State under Chapter 7 of the Constitution, and never became such on that day or any other. The Appellants’ case founders on this simple fact.
 It is astonishing that despite the indulgence extended to the Appellants to file further evidence for appeal, they have produced no evidence at all in respect of their claim that Bainimarama became Head of State on 29 May, prior to promulgating IMG 1-3. There is no affidavit from the ex-President or his staff, and no record, formal or informal, of any purported transfer of power, despite the critical comments on this evidential deficiency by Gates J. In a craftily worded paragraph (37) in their skeleton, the Appellants state:
There is no issue as to who was the President from 19 to 29 May. He then resigned. (Vol.1 Tab 16 p. 176(h)). He has confirmed his resignation in recent correspondence (Vol.2 Tab 1 Annex A p.4-6).
This is not the full story. The correspondence amounts to a letter from the ex-President on 15 December 2000 stating, ‘This is to confirm that I have retired …’ No date is given. The resignation reference is to the Bainimarama affidavit filed in separate proceedings. (Vol.1 p. 176). Gates J found that the doctrine of necessity could aid the Commander in dealing with the hostage crisis by emergency measures, but these he would take in his capacity as military Commander, not as Head of State. Neither the constitution itself nor the necessity doctrine would permit a soldier in operational command of a country in crisis to abrogate or remove its supreme law. His sworn duty was to uphold and protect the very constitution under which (section 112(2)) he had, in 1999, been appointed.
 The Appellants try to skate over the fundamental problem by a mixture of bluff about ‘sovereignty’ and distortions of the facts. After para. 37 (set out in 12 above) their skeleton continues:
38. There is then no dispute for the office of President or the title of sovereign power. The previous Government held office under the sovereignty of Ratu Sir Kamisese Mara. That sovereign authority dismissed the government and dissolved parliament. No other sovereign has appointed them a government or restored Parliament. (our italics)
This passage embodies the Appellants’ crucial and obvious error about the nature of sovereignty – i.e. that it resides entirely in the person of the President for the time being, rather than in the people through their representatives in parliament and office-holders under the Constitution: in the case of Britain since 1688 through ‘the Queen in Parliament’. (See below). However, as Gates J found (p.18) there was no reliable evidence that the President dismissed this Government. There remains no such evidence (and if he had done so, his actions would have been unconstitutional). There is certainly no evidence that he dissolved parliament – on the contrary, there exists a gazette notice that he prorogued (i.e. suspended) Parliament for 6 months - an action which confirms its continuing existence.
 So what happened, as a matter of law, on 29 May 2000? If the President ‘retired’ on that evening, then the functions of the Head of State devolved under the Constitution, prior to its ‘abrogation’, to the Vice President or to the Speaker of the House of Representatives. The military commander could not lawfully make himself the head of State. There is no evidence that the Vice President was a hostage. Bainimarama had no sovereign power, and there was no legal way by which he could acquire any. What he could do, of course (and did, in effect if not in form) was to attempt to restore peace and good order and to resolve the hostage crisis. His actions to this end were legitimated by the doctrine of necessity. But he was on 29 May, and he remained, no more than the Commander of the military forces of Fiji. He had no ‘sovereignty’ transferred to him, nor did he partake of any by declaring himself the Head of State, which was an idle statement. He was always the military Commander, acting in a national crisis with a duty to uphold the Constitution and not to abrogate it, however much its abrogation may have seemed to him a good idea at the time. The very source of his power as military Commander sprang from his appointment under the Constitution, and he was and remained powerless to magic it away by any formal pronouncement. Such necessary laws as he promulgated were martial laws, not sovereign decrees.
 The Appellants’ arguments for non-justiciability of decisions by ‘the sovereign’ are misconceived in law, but cannot apply in any event to bar the justiciability of actions taken by the military Commander. Quite simply, 1) he had no power to ‘remove’ the Constitution, and 2) could acquire none through the doctrine of necessity, which by definition legitimates only actions which are necessary to protect the constitution. You do not ‘protect’ a Constitution by abolishing it. Even if he could, then his action would be reviewable on grounds of reasonableness and proportionality, and the evidence filed by the Appellants fails these tests inter alia because:
(i) the only objections to the Constitution came from one ‘unsophisticated’ section of society who merely ‘perceived’ it to be inadequate and who had lawful avenues for securing its amendment;
(ii) the idea that its continuance was a threat came entirely as the result of the treasonable actions of the Speight group – i.e. it was their unlawful demand which created the conditions under which the abrogation was believed to be helpful to the restoration of order. The purported abrogation was therefore undertaken under duress as to circumstances: most dangerously, it was calculated to reward the Speight group by offering them a sop for ending their criminal activities. Two wrongs do not make a right; and
(iii) on proportionality principles, the Commander could have suspended any law ‘perceived’ to be adverse to indigenous ambitions and might have offered support for the GCC in its 25 May request for an amendment to constitutional provisions causing concern. But no further. Chitty, when he famously stated, ‘necessity knows no law’, was not quite right. The law of necessity, like any other legal doctrine, must be bound by the supreme law, i.e. the Constitution.
The ‘removal’ of the Constitution was beyond the power of the military authorities, but even had it been within their competence the decision to do so was unlawful because it was irrational, made in bad faith (to reward crime) and disproportionate.
 Gates J was indubitably correct in finding (at p.34);
In view of the factual situation in this case, the doctrine of effectiveness does not apply. Commodore Bainimarama is clearly no usurper. Having acted as he thought best in a temporary but dire hostage crisis, he handed over power to a civilian caretaker administration. Necessity would permit him to suspend the Constitution just for so long as to allow him to restore law and order. That concluded his role. An examination of the regime for effectiveness does not arise.
There is nothing in the fresh evidence filed by the Appellants to dispute this conclusion. The military Commander and the head of the caretaker Government do not pretend to be revolutionaries, or to have acted to establish a new legal order. Although they talk of amending the constitution prior to an election, they have not done so and they disclaim support for the actions of Speight and his group, whom they have charged with treason in terms which belie their legal submissions on this appeal. Speight et al are charged with a conspiracy ‘to overthrow the Government of the Republic of the Fiji Islands as established by law’ (i.e. the law of the 1997 Constitution) ‘between the 1st day of May 2000 and the 27th day of July 2000 inclusive’. This charge admits to the fact that at all times up to and including the surrender date the Government and Constitution continued in existence.
 Since the Appellants cannot show a lawful transfer of sovereignty from the President to the military Commander, they seek retrospectively to characterise the latter’s actions as a ‘revolution’ or a ‘coup d’etât’. But this amounts to a re‑writing of history in the interests of legal argument: there is nothing in the evidence to suggest that Fiji has gone through a revolutionary process since 19 May. On the contrary, it has been the victim of an attempted coup by the Speight group, which eventually (on 27 July) failed but which in its course entailed the application of emergency powers, including the appointment of persons to act in caretaker capacities to exercise executive powers. Most of these applications were legitimated by the doctrine of necessity, but not the abrogation of the Constitution or any other decisions which operate to undermine it. The Appellants’ new affidavits serve only to emphasise the non-revolutionary nature of the caretaker regime, appointed ‘to return Fiji to democratic parliamentary rule at the earliest possible time’ – (Qarase, para 2) – hardly a revolutionary objective. The only political change it proposes is by ‘ordering of a review of the Constitution’ i.e. of the 1997 constitution. This process – of amending the constitution – was provided for prior to 29 May, by the Constitution itself. Whoever heard of a revolution in order to pass a constitutional amendment?
 The fresh evidence demonstrates that the State, far from falling by a blow, i.e. a ‘coup’, remains standing just as before, save for dislocation caused by Speight et al. This is verified by the caretaker Prime Minister (paras. 9-10) and by the military Commander, who annexes a ‘press statement’ beginning, ‘When the RFMF had declared martial law on the 29th May the RFMF had assured the nation that it was acting in what we perceived as the best interests of the nation …’ This is an important admission that RFMF had no intention to carry out a revolution, but merely to impose a martial law while the constitutional power was paralysed. The other affidavits all emphasise the conduct of the following offices has been the same after, as before, the declaration of martial law on 19 May (save for Speight-related disruption):
(1) The public service: Jale affidavit (although note change in Judicial Services Commission, as to which see later).
(2) The Reserve Bank (Narube: ‘the routine conduct of business in Fiji since 19 May proceeded in precisely the same fashion as before 19 May’).
(3) Justice Department (Rabuka: ‘the administration of justice has proceeded as before’).
(4) Foreign Affairs (Tagicakibau: ‘the ministry has continued to conduct itself subsequently to 19 May in precisely the same fashion as before’ but note unexplained admission of Fiji’s suspension from the Commonwealth of Nations, the imposition of sanctions and suspension of aid other than humanitarian aid).
(5) Public prosecutions: (Naigulevu).
(6) Tax collection: (Veretayaco).
(7) Policing (Konrote: ‘the police force is today wholly answerable to government in precisely the same way, and to the same extent, as was the case prior to May 19th’).
(8) Local court (Ram: ‘the administration of local government has continued virtually unaffected…’).
(9) Public enterprises: (Korodrau – operating the same as before).
 The above affidavits were filed in an attempt to prove that the ICG is an ‘effective’ government (skeleton, paras. 81-83) although they show no such thing since ‘efficacy’ is a test directed to the degree of public acceptability and civil society respect of political institutions (see later). However, what they do serve to show is the non-revolutionary nature of the ICG across all governmental departments: its accession to power has produced no revolutionary change and no new constitution, albeit that it has resulted in numerous violations of domestic and international human rights norms. The government has issued a ‘Blueprint’ (see later) and set up a Commission to consider amendments to the Constitution - matters which could have been attended to through the democratic process established by the 1997 Constitution.
 The Respondents contend that the 1997 Constitution remains in force. The actions of the caretaker government are summarized by Brian Opeskin (an Australian Law Reform Commissioner) in  Public Law 622:
In October 2000, the Interim Administration established a Constitutional Commission comprised of 12 members, four of whom are Indo-Fijian. The terms of reference of the Commission are to ‘review previous Constitutions of the Republic of Fiji and examine Constitutions of other countries and after consulting widely through Fiji … report and make recommendations on a new Constitutional arrangement for Fiji’. The Commission is specifically instructed to take full cognizance of the principles of the 1997 Constitution and the relevant international laws on human rights. However, while purporting to be independent, the Commission’s terms of reference appear to presage its final result by requiring recognition of the paramount interests of indigenous Fijians, affirmative action for indigenous Fijians, and so on. The Commission is required to report by March 2001 so that a new constitution can be drafted in time to honour the Interim Administration’s commitment to return to a constitutional democracy within two years. It remains to be seen to what extent the Commission takes up the invitation, extended by its terms of reference, to model a new constitution on those of other countries and on norms of international law.
 The submissions which follow aim to support the decision of Gates J. They deal to some extent with the Appellants’ skeleton, although lack of time precludes a comprehensive rebuttal. Its fatal flaw is its failure to understand the nature of legal sovereignty (‘the Queen in Parliament’ not the Queen) and its serious misconception of the judicial role in resolving a constitutional crisis. It further confuses ‘efficacy’ with forcible control of territory, ignoring the requirements both of respect from civil society and legitimacy under norms of international law.
 The Appellants’ submissions as to justiciability are fundamentally misleading, in that they elide two entirely distinct issues, namely:
(i) the question of whether a power exists in law; and
(ii) the question of whether the exercise of a lawful power is reviewable by the judicial branch of government;
 The first question is logically prior to the second. By way of example, if a lunatic committed an act of arson and at his trial asserted that he was the King of Fiji and had acted pursuant to a bona fide opinion as to the requirements of national security, the court would commit a profound error if it decided that this assertion rendered his case non-justiciable. Similarly, if an Englishman arrived in the country and declared that he was the duly appointed Governor-General, any actions he purported to take on Fiji’s behalf in the field of foreign affairs would not be immune from judicial scrutiny in this country’s courts merely because of their subject matter. In both cases, the courts would seek first to establish whether the person before them was the person whom he claimed to be, wielding the power that he claimed to have; just as in the present case the Court would have to decide whether the present regime constitutes the lawful government before questions of justiciability could become even potentially relevant.
 These propositions are so self-evident that it is not surprising that the Appellants are unable to cite any relevant authorities to support this part of their case. Two which ‘raise different issues’ are however quoted (see paragraphs 12‑13):
(i) Government of the Republic of Spain v Arantzazu Mendi is clearly not to the point, given the clear authority establishing that the questions of a regime’s ‘necessity’ and ‘effectiveness’ are distinct from whether it ought to be given international recognition by other nations. See Madzimbamuto v Lardner-Burke  UKPC 2;  1 AC 645 at 723 per Lord Reid; Mangope v Van der Walt 1994 (3) SA 850 at 863 per Comrie J; Uganda v Commissioner of Prisons  EA 514 at 540 per Sir Udo Udoma CJ.
(ii) CCSU v Minister for Civil Service also provides no support for the Respondent’s case. It nowhere suggests that the existence of a power might itself be a non-justiciable question. Indeed, the case is authority for the proposition that all powers are judicially reviewable regardless of their source (see Lord Diplock at  1 AC 374), and that the existence and extent of a power are matters into which the courts will inquire (see Lord Fraser at 398E, Lord Scarman at 407D; Lord Roskill at 417E). (As Lord Scarman noted, this has been an unquestioned part of the common law since the early 17th century.) The Respondent also invites the Court to take judicial notice of the fact that over the twenty years since this case, the supervisory role of the common law over the executive has widened rather than narrowed: see by way of example, R v Home Secretary ex parte Bentley  QB 349 and Lewis v Attorney‑General of Jamaica 
3 WLR 1785 where England’s High Court and the Privy Council respectively asserted the reviewability of the executive discretion to withhold pardons and mercy from convicted persons.
 Since the Appellants concede that the two cases to which they refer ‘raise different issues’ (see para.13), their desired function is apparently to encourage this Court to adopt a deferential attitude to the present regime, to ‘avoid conflicts’ with it (para.13) implicitly because to do justice would undermine ‘the defence of the realm, the disposition of troops and the deployment of the police, [which] are all intimate aspects of order’ (para.14). These are not proper considerations for any court of law which upholds the rule of law. The mere fact that an action is an ‘intimate [aspect] of order’ could never of itself render an action non-justiciable, even if it were carried out by a lawfully constituted authority; and a fortiori could not render it immune from judicial scrutiny if it were performed by an unlawful one. It is submitted that any proposition to the contrary would be astonishing; and if such is being made by the Appellants, they are invited to make this explicit, and to submit authority in its support.
 The Appellants themselves assert that they have destroyed a lawfully promulgated constitution, overthrown a democratically elected government and unilaterally abolished the nation’s Supreme Court. They assert that this is proper and lawful; while the Respondents of course assert that it is not. It is submitted that it is inherent within the concept of an ordered legal system that the conflicting submissions fall to be resolved by the judicial organs of the state. This basic proposition, unlike the Respondent’s circular submissions about self‑creating sovereignty, does not merely depend on itself for validity. The very fact that this court is called upon to pronounce on the law applicable to this case requires that it decide the merits, for even if other bodies or individuals might wish to make this decision, a law court cannot abdicate its decision-making power unless it is to deny the rule of law and thus its very reason for existence. Even in a case where the lawful constitution of the executive is not in doubt, a court is bound by the history of the common law to assert its supremacy over that executive. To paraphrase Lord Templeman’s words in M v Home Office  UKHL 5;  1 AC 377 (at 395), the contrary argument:
would, if upheld, establish the proposition that the executive obey the law as a matter of grace and not as a matter of necessity, a proposition which would reverse the result of the [English] Civil War.
 In the United States, the matter has been put in the following way:
No one, no matter how exalted his public office or how righteous his private motive, can be judge in his own case. That is what courts are for. (United States v United Mine Workers,  USSC 40; 330 US 258 at 307-309 (1947)).
 This requires steadfastness on the part of the judiciary, but few principles are more fundamental to the professional obligations of a judge. In their joint report on ‘Justice in Jeopardy: Malaysia 2000’, this imperative was spelled out by representatives of the International Bar Association, ICJ Centre for the Independence of Judges and Lawyers, the Commonwealth Lawyers’ Association and the Union Internationale des Avocats, when they stated (at Part III, para.6) that:
The judiciary should do all in its power, in the wider interests of justice, to counter the harshness of repressive legislation and overbearing actions on the part of the executive. That is the role of the judiciary when faced with oppression, no matter where it comes from. The burden will fall mainly on senior members of the judiciary, who must take the lead.
 It cannot be right that a greater degree of deference is due when the very legitimacy of the executive is questionable. As was memorably stated by United States Chief Justice Marshall in Marbury v Madison,  USSC 16; 5 US 137 (1803):
It is emphatically the province and duty of the judicial department to say what the law is … This is of the very essence of judicial duty.
 The supremacy of the judicial branch of government was reaffirmed by the unanimous Supreme Court decision in the landmark desegregation case of Cooper v Aaron,  USSC 160; 358 US 1 (1958), where it was said (at 25) that:
Our kind of society cannot endure if the controlling authority of the Law … is not to be the tribunal specially charged with the duty of ascertaining and declaring what is ‘the supreme law of the land.
 In that case there was no assertion by either of the parties that the Constitution itself was invalid. It is however submitted that the existence of such an assertion in the present case does not alter this ‘essen[tial] judicial duty’. It stems not from the formal source of the law to be applied by the court, but from the nature of democracy and the rule of law to which this court must hold itself committed – from the fact that, ‘The historic phrase ‘a government of laws and not of men’ epitomises the distinguishing character of our political society’ (c.f. Cooper v Aaron at 23). The mere assertion by a military-backed regime that Fiji is not this kind of society can have no legal force with this court. Marbury v Madison established the ground rules for legal debate and political contention within the United States; and insofar as such rules are said to be lacking in Fiji, it is no less ‘emphatically’ this court’s duty to state what they are. It cannot, without denying its own commitment to the rule of law, defer to the assertions of an undemocratically installed regime and refuse to decide the merits of this case. Its most basic duty under the common law is to assert its supremacy over all other actors in the body politic.
 The reasons for this are not obscure, academic or theoretical. Only too often during the twentieth century have individuals and unpopular minorities suffered exploitation, or worse, at the hands of over-powerful governments and police states. The notion of entrenched human rights has developed on the national and international level in response to this abuse of power, but such rights are an empty promise unless a court is ready, willing and able to adjudicate on whether or not they have been breached, and a fortiori, whether they have been abrogated. It is possible that the rights to equal treatment and democratic governance hitherto protected by Fiji’s 1997 Constitution will cease to be observed in Fiji as a matter of fact, but it is respectfully submitted that this court cannot as a matter of law properly refuse to decide whether they exist.
 By way of conclusion, the Respondents submissions on justiciability in the present case are the following:
(i) judicial deference may, in a narrow and specific range of circumstances, apply to limit the level of scrutiny of a duly constituted authority; but where the very legitimacy of that authority is in question, it can have no application;
(ii) the lawfulness of the abrogation of the 1997 Constitution is a question that can be resolved by this Court;
(iii) this matter does not fall into any of the categories which have traditionally been held by the courts to be non-justiciable. It does not, for example, raise questions of foreign policy or deal with the internal workings of the Parliament;
(iv) it is a legal, as opposed to a political, question which can be resolved by the application of the relevant legal doctrines. In other like circumstances, domestic courts have consistently held that such questions are justiciable: see all the cases on effectiveness and necessity cited below, and in particular see Madzimbamuto  UKPC 2;  3 All ER 561 at 573 (‘[The court] must decide.)
(v) Each of the orders made by Gates J lay within the power of a judge of the High Court in the sense of dealing with justiciable matters. See Bavadra v Attorney-General  SPLR 95.
 The Appellants’ submissions contain a number of assumptions and prescriptions as to the appropriate conduct of judges when the constitutional order is unsettled. According to their skeleton argument, judges must make no value judgments on the moral merits of the new regime (5 xv; 17) they must either resign or remain ‘to discharge the judicial function within its proper province’ (5 xxiii; 18; 68). That ‘proper province’ is defined by the Appellants to exclude any determination of the legitimacy of the administration, other than by enquiring whether it claims to act under a new legal order and if so whether it has sufficient de facto control of the country to enforce obedience to its dictates. This prescription is a logical development of the ‘might makes right’ tenor of the Appellants case: it now becomes ‘might has made right, and will not moreover suffer any court to say it has done wrong’. The judge who (without hearing argument, other than his own) concludes that a power grab is unlawful (and hence immoral) must resign say the Appellants, so that his decision is never given in court; and the judge who stays to do whatever justice is on offer cannot entertain a case which raises any point other than whether the regime is in ‘sufficient’ control of the country. By this route, the Appellants require judges to abandon their duty, either by leaving their office or demeaning it, by refusing to hear argument which impeaches the legitimacy of the executive.
 These are not true alternatives. Generalisations are difficult: history shows how political crises have caused honourable judges to resign, and less honourable judges to do the bidding of the new executive. Some courageous judges have remained in place, only to be murdered when their decisions displease that executive (e.g. the Court of Appeal judges in Uganda under Idi Amin). Judges last year in Pakistan, and this year in Zimbabwe, face similar decisions. But if upholding the rule of law is the essential judicial role, then at times of political disturbance a judge is perfectly entitled - and may indeed be bound - to remain in place, hearing argument and making such rulings on the legitimacy of any new order as he or she decides to be right. Such duty would obviously be limited in practice by considerations of personal safety, but not of personal conscience: judges who remain and decide (after hearing argument) that the new order is illegal may find themselves removed from office or put on circuit while more pliant fresh appointees overrule them on appeal. But they have stood for justice, and that virtue, recorded in the law reports, will have its own reward. The choice is not between resigning or being a lickspittle of the new regime: the third way, of remaining to pass public judgment on that regime, is preferable when it is practicable. The first alternative may in some circumstances be appropriate; the second is, in any circumstance other then duress, unthinkable.
 This third way comports with the shared legal history of the Commonwealth of nations, ever since Chief Justice Coke faced down the raging James I with the words:
The law is the golden met-wand and measure to try the causes of your majesties subjects, and it is by that law that your majesty is protected in safety and peace.
 When James accused Coke of treason for suggesting his sovereign power might be under the law, the Chief Justice replied ‘Thus wrote Bracton: ‘the king is under no man but God and the law’. As Lord Denning points out, this saying has ‘reverberated down the centuries’ to make judges the guardians of the constitution, (What Next in the Law, p311-318). The Appellants’ claim to the contrary – in effect that judges must ‘get out, or get under’ - denies this heritage, and the very raison d’être of judicial conscience and independence.
 It is submitted that the Appellants’ case is flawed at its very root by its submissions on sovereignty, which are (i) based on archaic sources, and (ii) irrational and self-contradictory.
 The Appellants cite two ‘classical writers’ at the outset of their arguments. These citations are clearly intended to lend authority to their later submissions on the law, and for this reason it is submitted that it is appropriate for the court to have some regard to the historical context and intellectual traditions in which they were writing.
Jean Bodin’s Six Books of the Commonwealth are regarded as a classic statement of absolutist power, written in a medieval European tradition which was a precursor of the legal doctrines and jurisprudential apologias for Nazi Germany and Soviet Russia. JM Kelly in A Short History of Western Legal Theory (OUP, 1992) characterised Bodin as an ‘evangelist of absolutism’ and his doctrines are contrasted with those which would eventually coalesce into the ‘rule of law’ (pp.168-79).
Bodin was celebrated among his 16th century contemporaries for his contributions to another legal theory, that of witchhunting. His Demonomanie of 1580 contains an authoritative guide as to how to prosecute witches, based on his belief that ‘proof of such evil [is] so obscure and difficult that not one out of a million witches would be accused or punished if regular legal procedures were followed’ (quoted in G.R. Quaife, Godly Zeal and Furious Rage (Croom Helm, 1987 at p.140; see also p.26). To erect a modern theory of sovereignty with Bodin as the first intellectual building-block is to invite ridicule.
The jurisprudence of John Austin is of some historical interest in the development of positivist thought, but during the 20th century was progressively criticised. ‘Much of the keenest criticism [was] directed against Austin’s rather rigid views as to the nature of sovereignty … in a revolutionary setting, a military takeover, a revolution, a secession, Austin’s test of habitual obedience is too open-textured to be useful as a criterion’ (see Lloyd’s Introduction to Jurisprudence (6th ed.) at p.216-17). Even Austin’s own editor, in the edition from which the Appellants cite, notes that his notions of sovereignty were disproved by Maine and undermined by others. It is a generalised descriptive theory, now abandoned, and of no assistance in consideration of a written constitution.
Professor Ronald Dworkin in Law’s Empire did make note of Austin’s lingering influence on jurisprudential thought but the court is invited to consider the context in which the single sentence cited by the Appellants appeared. Professor Dworkin’s entire career has been concerned to refute this ‘definitional’ type of jurisprudence, which he has at various times ridiculed as ‘semantic’ and ‘conventionalist’. The court’s attention is drawn to the full text in which the cited quotation appears (pp.31-35): note that Dworkin’s exposition of Austin’s views on sovereignty is made within the context of an exposition of ‘semantic theories of law’, which he later describes as a ‘caricature of legal practice’ (p.87). Far from adopting Austin’s views, Dworkin regards his type of definitional jurisprudence as fundamentally wrongheaded.
 The Appellants assert that, ‘In a republic the legal personality of the sovereign is the holder of the office of the President’ (para.37). This in fact contradicts the Appellants’ reliance on Austin, who located sovereignty in the empirical fact of habitual obedience (see Lloyd’s Introduction to Jurisprudence at p.219) and who himself drew no distinction between ‘sovereignty’ and its ‘legal personality’: indeed such a distinction would have made no sense within his own terms. This confusion is symptomatic of a more fundamental logical unsoundness in the Appellants’ submissions. Even if it were correct (which it isn’t) that on 29 May 2000, President Ratu Mara was the personification of Fijian sovereignty, there is no evidence (and it is barely asserted in the submissions) that he purported to transfer that sovereignty. The only clues provided by the Appellants as to how a new sovereign supposedly emerged are contained at p.4(viii), and paras. 23 and 24. These paragraphs describe the means by which the present regime usurped power, but the Court is invited to note that they can do no more than this; and certainly do not set out a lawful transfer of authority.
 The Appellants state in their submissions, para 5(viii) ‘that it is inherent in the nature of sovereignty that a sovereign is the source of authority and is capable of creating a new legal order irrespective of the terms of the old one’. This conception of sovereignty is consistent only with the power held by Henry VIII in 16th century England, or a ruler in a modern dictatorship. It is fundamentally at odds with the rule of law and the constitutional structure of a democratic nation such as Fiji, where the State is constituted by the people of Fiji under the 1997 Constitution, and not by the President or the Military.
 In a constitutional monarchy (viz., the United Kingdom), it is universally accepted that sovereignty resides not in the Queen, but in the Queen in Parliament, i.e. the head of state and the peoples’ representatives as bound by constitutional conventions. Similarly in a constitutional republic, sovereignty, while ultimately derived from the people, is formally divided between the people’s representatives and officers appointed under the constitution, bound by the rules of that constitution.
 Section 2 of the 1997 Constitution states that ‘(1) This Constitution is the supreme law of the State; (2) Any law inconsistent with this Constitution is invalid to the extent of the inconsistency’, while the Preamble states, among other things, that ‘We the people of the Fiji Islands … with God as our Witness, Give Ourselves This Constitution’, and section 21 provides that the Bill of Rights binds ‘the legislative, executive and judicial branches of government …’. This constitutive statement of intention is irreversible, because:
(i) the rights of political participation, racial equality, self-determination and democracy are now a part of customary international law. Thomas Franck, in his comprehensive survey of the law in this field (‘The Emerging Right to Democratic Governance’, (1992) 86 Am. J. Int’l L. 46.) notes at 85 that:
It is no longer arguable that the United Nations cannot exert pressure against governments that oppress their own peoples by egregious racism, denials of self-determination, and suppression of freedom of expression.
and at 78 that:
... state sovereignty, by operation of technological advances as much as of heightened humanistic sensitivity, is not what it used to be. Even those who defend the vitality of the principle [of state sovereignty] would probably concede that genocide does not fall ‘essentially’ within the ambit of protected ‘domestic’ government activity. Nor, certainly, do egregious racism and, at least since the 1960s, denials of self-determination.
The UN Security Council Resolution on Haiti, and its intervention to restore constitutional democratic government in former Yugoslavia both before and after the 1995 Dayton Accords, pursuant to the principles of the UN Charter, have provided dramatic evidence of the extent to which these norms fall nowadays to be enforced on the international sphere. It is submitted that this development necessarily means that these norms must now be regarded as human rights entrenched by virtue of international law, even if (which is denied) they might otherwise be removable at the domestic level. In this context it is noted that section 43(2) of the Bill of Rights chapter of the 1997 Fiji Constitution states that:
In interpreting the provisions of this Chapter the courts must promote the values that underlie a democratic society based on freedom and equality and must, if relevant, have regard to public international law applicable to the protection of rights set out in this Chapter.’
(ii) the last several hundred years of common law and constitutional tradition, within which heritage this Court itself is of course situated, has recognised and given immediately irreversible effect to a number of constitutional shifts where these shifts have tended to advance popular participation over oligarchic or autocratic rule. (By way of example, the Court is invited to take judicial notice of the fact that it would be unthinkable for the British Crown to reassert its sovereignty over its former colonies – but if the Appellants were correct, there would be nothing to stop it from doing so, and such recolonisation would have the immediate force of law.)
 Sovereignty in a constitutional state is divided according to the terms of that constitution, and a judicial tribunal committed to the rule of law must give effect to the division as enumerated therein. To suggest that either the President of Fiji or the Bose Levu Vakaturaga (Great Council of Chiefs), both of whose powers are carefully delineated by the 1997 Constitution (sections 85 and 116) possess a power to abrogate the source of their own power and set themselves above it is inconsistent with basic tenets of the rule of law, as classically expounded by A.V. Dicey in Introduction to the Study of the Law of the Constitution (Macmillan, 1st ed 1885, 10th ed 1959) at 202-203:
It [the ‘rule of law’] means, in the first place, the absolute supremacy or predominance of regular law as opposed to the influence of arbitrary power, and excludes the existence of arbitrariness, of prerogative, or even of wide discretionary authority on the part of the government. … It means, again, equality before the law, or the equal subjection of all classes to the ordinary law of the land administered by the ordinary law courts; the ‘rule of law’ in this sense excludes the idea of any exemption of officials or others from the duty of obedience to the law which governs other citizens or from the jurisdiction of the ordinary tribunals.
See Re Manitoba Language Rights  1 SCR 721 at 748-749
 If the Appellants were correct in their explicit assertion that ‘in a republic the legal personality of the sovereign is the … President’ (para.37), and their implicit assertion that he could therefore lawfully hand on absolute power as he willed, there would be no reason why in, for example, the United States, the President could not legally dispense with the courts, penal codes and Congress, and begin to institute a policy of murdering his opponents. Such an act would of course be possible in fact, and if it occurred the courts of the United States would have a duty to withhold recognition of such action. The absence of authority can hardly be a reason for this court to lend its imprimatur to the Appellants’ novel and regressive propositions of law in this regard.
Has Not Been Followed
 It is the duty of a court of law to determine whether purported government action complies with the Constitution.
Marbury v Madison, 5 US  USSC 16; (1 Cranch) 137 (1803)
 The procedures set down for the alteration of a constitution must be complied with if an amendment is to be legally effective. This is a fundamental tenet of the rule of law.
 The 1997 Constitution establishes such a mechanism in Chapter 15. Furthermore, the Constitution establishes that this is the only way of amending the Constitution. Section 190 states:
This Constitution may be altered in the way set out in this Chapter and may not be altered in any other way.
 This means that any other method of amending the Constitution is unlawful and of no legal effect. This is reinforced by section 2 of the Constitution (set out above).
 The lawful amendment process is provided by section 191 in Chapter 15. The parliamentary procedure set out there must be followed and a Bill incorporating the changes must gain the support of:
(1) at least two-thirds of the members of each House of the Fiji Parliament; or
(2) if the Prime Minister certifies that a particular amendment is an urgent measure, then a majority of at least 53 of the 71 members (three quarters) of the House of Representatives and a majority of the members of the Senate.
 Section 192 limits this power in one respect. The number of seats in the House of Representatives allocated to each ethnic group cannot be altered unless a specified level of support is received from the representatives of that group. For example, 23 of the 71 members of the House of Representatives are elected by persons on the Fijian electoral roll. Section 192 states that this level of representation cannot be changed unless the amendment receives the support of at least 15 of those 23 members.
 As a consequence of these provisions, there is only one lawful way to repeal the Constitution entirely. A Bill must be passed by the required majorities in the Fiji Parliament, including with the specified level of support from each of the groups of ethnic representatives.
 This has not occurred. Interim Military Government Decree No 1, Fiji Constitution Revocation Decree 2000, which provides that ‘the Fiji Constitution Amendment Act 1997 is with effect from 29th day of May 2000, wholly removed’, does not comply with Chapter 15 of the Fiji Constitution and is therefore legally ineffective.
 As a consequence, the 1997 Constitution remains in force.
 Courts have recognised a principle of ‘necessity’ that dictates that, in times of extreme crisis, emergency action (including by a president or the armed forces) may validly be taken that would otherwise be illegal. Such action, such as a declaration of martial law, must be a transient and proportionate response to the crisis.
de Smith and Brazier, Constitutional and Administrative Law (7th ed 1994) at 73‑74
Mitchell v Director of Public Prosecutions  Law Reports of the Commonwealth (Const) 35 at 88-89 per Haynes P
Attorney-General v Ibrahim  Cyrpus Law Reports 195
 Hence, the suspension of the ordinary operation of the law after 19 May by the President of Fiji, Ratu Sir Kamisese Mara, and then by the Military might be upheld by this court on the basis that martial law was required by the seriousness of the events then occurring due to the attempted coup d’etât of the Speight Group. On 27 May 2000 it is said that Kamisese Mara dismissed the government and prorogued the Parliament. Significantly, it is not asserted that he dissolved the Parliament (Appellants’ Chronology of Events, 27 May 2000).
 The onus of proof in showing that the necessity doctrine could be invoked and that the actions purportedly taken under it were justified lies with the party defending the use of the doctrine.
Mitchell v Director of Public Prosecutions  Law Reports of the Commonwealth (Const) 35 at 89 per Haynes P
Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 132 per Cullinan CJ
Makenete v Lekhanya  Law Reports of the Commonwealth 13 at 65 per Ackerman JA
 The principle of necessity is limited in one critical respect. The doctrine extends only to action that is designed to uphold the rule of law and the existing constitution. Necessity can never justify or support the abrogation of the existing legal order, for that would amount to the exactly sort of action that the doctrine is designed to protect against. As Brookfield has summarised the principle in ‘The Fiji Revolutions of 1987’  New Zealand Law Journal 250:
the power of a Head of State under a written Constitution extends by implication to executive acts, and also to legislative acts taken temporarily (that is, until confirmed, varied or disallowed by the lawful Legislature) to preserve or restore the Constitution, even though the Constitution itself contains no express warrant for them. (emphasis added)
See also Brookfield, Waitangi & Indigenous Rights: Revolution, Law and Legitimation (1999) at 20.
 Emergency action cannot be undertaken under the principle of necessity as a means of subverting the existing constitutional structure. The necessity doctrine cannot authorise the abrogation of the existing legal order, only its temporary suspension.
Bhutto v Chief of Army Staff 1977 PLD SC 657 at 723, 728, 753
Zafar Ali Shah v Pervez Musharraf, Chief Executive of Pakistan (Pakistan Petitions Case) (2000) 33 Supreme Court Monthly Review 1137 at 1160-1161 per Irshad Hasan Khan CJ
 According to Lord Pearce of the Privy Council in Madzimbamuto v Lardner-Burke  UKPC 2;  1 AC 645 at 732:
I accept the existence of the principle that acts done by those actually in control without lawful validity may be recognised as valid or acted upon by the courts, with certain limitations namely (a) so far as they are directed to and reasonably required for ordinary orderly running of the State, and (b) so far as they do not impair the rights of citizens under the lawful (1961) Constitution, and (c) so far as they are not intended to and do not in fact directly help the usurpation and do not run contrary to the policy of the lawful Sovereign.
 Similarly, in Madzimbamuto v Lardner-Burke 1968 2 SA 284 at 441, Fieldsend AJA (who subsequently became the first Chief Justice of Zimbabwe) stated:
the act must not be intended to, or in fact in its operation directly, further or entrench the usurpation.
 Similarly, in Texas v White, 74 US  USSC 97; (7 Wall) 700 at 733 (1862), the Supreme Court of the United States held that limited validity might be accorded to the actions of the individual insurrectionist Southern States during the United States Civil War. Chase CJ stated for the Court at 733:
It is not necessary to attempt any exact definitions within which the acts of such a state must be treated as valid or invalid. It may be said perhaps with sufficient accuracy, that acts necessary to peace and good order among citizens, such, for example, as acts sanctioning and protecting marriage and the domestic relations, governing the course of descents, regulating the conveyance and transfer of property, real and personal, and providing remedies for injuries to person and estate, and other similar acts, which would be valid if emanating from a lawful government, must be regarded in general as valid when proceeding from an actual, though unlawful, government; and that acts in furtherance or support of rebellion against the United States, or intended to defeat the just rights of citizens, and other acts of like nature, must, in general, be regarded as invalid and void. (emphasis added)
See Baldy v Hunter,  USSC 114; 171 US 388 at 400-401 (1898) (United States Supreme Court)
 These authorities demonstrate that emergency action taken from 19 May onwards by the President or the Military can be upheld by this Court only to the extent that it was necessary to as a temporary measure to:
• resolve the crisis brought on by the coup attempt of the Speight Group;
• maintain the 1997 Constitution; and
• uphold the rule of law.
 Such action should not be upheld by this Court insofar as it would assist in the usurpation, whether of the Speight Group or the Military.
 The necessity doctrine cannot be applied to uphold the legality of a new regime, only to recognise the legality of certain acts taken to support the existing arrangements.
Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 122 per Cullinan CJ
 The lawful limits of the emergency action are strictly limited in this way, or otherwise the exercise of the power would not be valid under the principle of necessity, but would be in effect an illegal seizure of power which for its legality must be judged according to the doctrine of ‘effectiveness’ (see below).
See Paris Minimum Principles in a State of Emergency; Barcelona Traction 1970 ICJ Reports at 32
 On this basis, the actions of the President and the Military may be upheld by this Court in so far as they do not undermine or abrogate the existing legal order, including the 1997 Constitution. To the extent that the actions of the President and the Military go beyond this, they must be read down to their proper lawful scope.
For example, Interim Military Government Decree No 7, Fundamental Rights and Freedoms Decree 2000, which limits the right of equality in section 38 of the 1997 Constitution by stating in section 19(7)(g) of the Decree that the right does not extend to laws that provide ‘for the prosecution of unnatural offences or indecent practices’.
 Interim Military Government Decree No 1, Fiji Constitution Revocation Decree 2000, which provided that ‘the Fiji Constitution Amendment Act 1997 is with effect from 29th day of May 2000, wholly removed’, is accordingly unlawful. Each of the other Interim Military Government Decrees (including Interim Military Government Decree No 10 of 2000, Interim Civilian Government (Establishment) Decree 2000) that are based upon an abrogation of the 1997 Constitution are also unlawful.
Must Uphold the 1997 Constitution
 This Court cannot declare the purported Interim Civilian Government (sought to be created by Interim Military Government Decree No 10 of 2000, Interim Civilian Government (Establishment) Decree 2000, and then by Interim Military Government Decree No 19 of 2000, Interim Civilian Government (Transfer of Executive Authority) Decree 2000) to be the lawful government of Fiji.
 This Court is legally bound to declare that the elected government of Fiji remains the lawful government and that the 1997 Constitution remains the supreme law of Fiji. It lacks the jurisdiction to decide otherwise.
 The powers and functions of the High Court and Court of Appeal of Fiji are created by Chapter 9 of the 1997 Constitution. Section 117(1) in Chapter 9 provides:
(1) The judicial power of the State vests in the High Court, the Court of Appeal and the Supreme Court and in such other courts as are created by law.
 The jurisdiction of the High Court is determined by section 120, while its composition is established by section 126. The jurisdiction of the Court of Appeal is similarly determined by section 121, and its composition is established by section 127.
 The purported re-establishment of the Fiji judicial system by Interim Military Government Decree No 5 of 2000, Administration of Justice Decree 2000, is unlawful and ineffective. As submitted above, it is unlawful to the extent that it subverts the existing Constitution. This Court can only be re-established in accordance with the terms of the 1997 Constitution.
Adams v Adams (Attorney-General intervening)  3 All ER 572 at per Sir Jocelyn Simon P.
 So long as this Court remains the Court of Appeal as established by the 1997 Constitution, it lacks the power to declare other than that the 1997 Constitution remains in force and cannot be abrogated except by the mechanism set down therein. This Court cannot deny the source of its own power, and must uphold the effectiveness of the 1997 Constitution.
 According to Lord Pearce of the Privy Council in Madzimbamuto v Lardner-Burke  UKPC 2;  1 AC 645 at 732 (in dissent on this point): ‘The judges under the 1961 Constitution therefore cannot acknowledge the validity of an illegal government set up in defiance of it.’
 Similarly, as stated by Fieldsend AJA of the High Court of Southern Rhodesia in Madzimbamuto v Lardner-Burke 1968 2 SA 284 at 432: ‘The law to be administered by a municipal court is … determined solely by the set of norms prescribed by the legal order upon which the court considering the case is founded’ (see at 429). And at 430:
A court created by a written constitution can have no independent existence apart from that constitution; it does not receive its powers from the common law and declare what its own powers are; it is not a creature of Frankenstein which once created can turn and destroy its maker.
 Similarly, as stated by Chief Justice Taney for the United States Supreme Court in Luther v Borden, 48 US (7 How) 1 at 40 (1849):
Judicial power presupposes an established government capable of enacting laws and enforcing their execution, and of appointing judges to expound and administer them. The acceptance of the judicial office is a recognition of the authority of the government from which it is derived.
 To deny the ongoing force of the 1997 Constitution (other than in accordance with the amendment mechanism provided for in the instrument) would also be a breach of a judge’s oath or affirmation of office as made under section 135 of the 1997 Constitution and as set out in Part D of the Schedule:
Oath or Affirmation for due Execution of Judicial Office
Oath: I, A.B. , do swear that I will well and truly serve the Republic of the Fiji Islands, in the office of [ ]. I will in all things uphold the Constitution; and I will do right to all manner of people in accordance with the laws and usages of the Republic, without fear or favour, affection or ill will. So help me God!
Affirmation: I, A.B. , do solemnly, sincerely and truly declare and affirm that I will well and truly serve the Republic of the Fiji Islands, in the office of [ ]. I will in all things uphold the Constitution; and I will do right to all manner of people in accordance with the laws and usages of the Republic, without fear or favour, affection or ill will. (emphasis added)
See Madzimbamuto v Lardner-Burke 1968 2 SA 284 at 425, 430 per Fieldsend AJA
 Furthermore, it is the special duty of all courts to strive to uphold the human rights of the citizens of a nation, especially against governmental or purported governmental action that breaches basic constitutional standards.
Sir Gerard Brennan, ‘The Impact of a Bill of Rights on the Role of a Judiciary: An Australian Perspective’ in Alston (ed), Promoting Human Rights Through Bills of Rights (1999) at 458-464
Sir Robin Cooke, ‘Fundamentals’ (1988) New Zealand Law Journal 158.
Justice Michael McHugh, ‘The Law-making Function of the Judicial Process’ (1988) 62 Australian Law Journal 15 at 24, 27, 28, 123-124
Should Not Uphold the Lawfulness of the Revolution
under the ‘Effectiveness’ Doctrine
 Alternatively, if the arguments above are not accepted and this Court is willing to exercise a ‘supra-constitutional’ jurisdiction, it should only uphold the lawfulness of the purported Interim Civilian Government and the abrogation of the Fiji Constitution to the extent that this is permitted by the ‘effectiveness’ doctrine as recognised by the common law or by a fundamental unwritten principle of Fijian constitutionalism. The authorities on this doctrine are reviewed extensively in Mitchell v Director of Public Prosecutions  Law Reports of the Commonwealth (Const) 35 and Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24.
 The onus of proof in showing that the effectiveness doctrine is applicable lies with the party seeking to invoke the doctrine in order to overturn the established legal order. According to Cullinan CJ of the High Court of Lesotho (formerly of the High Court of Fiji at Lautoka) in Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 132:
the burden of proof of legitimacy must always rest upon the new regime. No presumption of regularity can operate in the regime’s favour: indeed there must be a presumption of irregularity.
See Mitchell v Director of Public Prosecutions  Law Reports of the Commonwealth (Const) 35 at 72 per Haynes P; Makenete v Lekhanya  Law Reports of the Commonwealth 13 at 65 per Ackerman JA
 Even if this Court holds, contrary to our primary submissions, that the law of Fiji recognises a limited exception to the rule that a court must uphold the legality of a constitution under which it is created, then we submit that this exception is not applicable in this case.
 Furthermore, even if this Court holds that it was re-established by Interim Military Government Decree No 5 of 2000, Administration of Justice Decree 2000, that does not preclude this Court from holding that the purported Interim Civilian Government is not the lawful government of Fiji. This would be correct even if the Judges of this Court had taken new judicial oaths as part of the re‑establishment.
Uganda v Commissioner of Prisons  EA 514 at 530 per Sir Udo Udoma CJ
Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 139 per Cullinan CJ
Zafar Ali Shah v Pervez Musharraf, Chief Executive of Pakistan (Pakistan Petitions Case) (2000) 33 Supreme Court Monthly Review 1137 at 1156 per Irshad Hasan Khan CJ
 A leading authority on the ‘effectiveness’ doctrine is the decision of the Privy Council in Madzimbamuto v Lardner-Burke  UKPC 2;  1 AC 645. That case concerned the legality of a unilateral declaration of independence (UDI) by Southern Rhodesia from Britain in 1965. The regime led by Ian Smith repealed the existing Constitution and announced a new Constitution. The Privy Council was asked to determine whether legal recognition could be given to the new regime.
 Lord Reid for the majority held that the Privy Council could rule upon whether a new regime had become the lawful government of a nation. In narrow circumstances, the Court possessed the jurisdiction to hold that a revolutionary regime was lawful (despite the fact that the acts that led to its ascent to political power were themselves unlawful), and ‘must decide’ upon the ‘status of a new regime which has usurped power and acquired control of that territory’ (at 724). According to Lord Reid at 724:
It is an historical fact that in many countries - and indeed in many countries which are or have been under British sovereignty - there are now regimes which are universally recognised as lawful but which derive their origins from revolution or coups d’etat. The law must take account of that fact. So there may be a question how or at what stage the new regime became lawful.
 The only legal question in determining whether recognition may be given to a new regime is that of ‘effectiveness’, that is, whether the revolution has been practically successful. As stated by the Chief Justice of Pakistan in State v Dosso 1958 PLD SC 533 at 539 (and quoted by Lord Reid in Madzimbamuto at 725):
Thus the essential condition to determine whether a Constitution has been annulled is the efficacy of the change.
 In applying this test, the validity of the pre-existing legal order must be upheld until it is clear that the new regime has succeeded in ousting rival claims from the preceding regime, such that there could not be said to be two rival regimes contending for power. All attempts by the predecessor government to resume control must have effectively ceased. Until that time, the revolutionary regime must be regarded as illegal.
Madzimbamuto v Lardner-Burke  UKPC 2;  1 AC 645 at 725 per Lord Reid
 This inquiry is a mixed question of fact and law.
 Primarily, it depends on a purely empirical inquiry as to the degree of acceptance and effective control established by the new regime. It is submitted that no case establishes as a general rule that a habit of obedience amounts to legitimacy: see Madzimbamuto at p.726C. As de Smith points out in Constitutional and Administrative Law (8th ed.) at 71, ‘might’ may become recognised as ‘right’, but this ‘offers a description, not a prescription. It does not dictate what attitude judges and officials ought to adopt when the purported breach of legal continuity takes place.’ To put the matter another way, ‘might’ can never ipso facto equal ‘right’ under the common law. Whether people are as a matter of fact behaving in conformity to the government is one of the factors which must be considered in deciding whether an effective change has occurred: see Mokotso at 125, 133, Makenete at 43c. But although ‘submission’ of the populace may be a necessary condition for a new regime’s lawfulness, it can never be a sufficient one. Many other factors are also relevant, as is made clear by the other authorities in this field (see e.g. Liverpool JA’s judgment in Mitchell at 115h – 116e; Mokotso at 164). In this case, it is submitted that these factors include:
(i) whether the former ruler purports still to be a rival for power. In the present case, the elected government has mounted a legal challenge to its usurpation; the elected Prime Minister (see his affidavit) has on numerous occasions asserted his legitimacy and has indeed been received as such by several foreign leaders; and more generally, numerous political parties have refused to accept the validity of the present dispensation;
(ii) the stance of the army and police force. It is of importance in this case that both bodies have not indicated an unequivocal stance that the constitution has been abrogated. Instead, representatives from each have affirmed their commitment to the rule of law, to the ultimate jurisdiction of this court, and their willingness to abide by its decisions;
(iii) the stance of NGOs and other bodies which form part of Fiji’s body politic – trade unions, religious bodies, public associations, the country’s ‘civil society’;
(iv) the attitude of the country’s judiciary and legal profession: see Madzimbamuto at p.725. In the present case, the Law Society maintains its opposition to the purported abrogation of the constitution, and at least two High Court judges have explicitly ruled that the 1997 Constitution remains the supreme law of the land: see Ved Prakash v Native Land Trust Board & Filipe Kubuyawa (Civil Action No. HBC0409D) and Julie Doyle v Phyllis Doyle Trustee Corporation Ltd. (Civil Appeal No.ABU0001/97S & ABU0003/97S). It is also noted that the usurping regime’s own purported decrees do not enforce the taking of a new oath by a judge ‘if that person has taken such oaths within Fiji on any previous occasion’ (see IMG Decree No.5, s.14(2); ICG Decree No.22, s.13(2)), and it is submitted on the Respondent’s behalf that as a matter of fact the majority of Fiji’s judiciary are still bound by their oaths to the 1997 Constitution; and
(v) the extent to which the usurping regime has become regarded as having become the settled, unquestioned and legitimate successor of its predecessor. In this case, it is noted that organisations as diverse as the European Parliament, the Commonwealth, the CERD and numerous foreign governments have questioned the legitimacy of the usurping regime.
 Having regard to all these factors, the Appellants are unable to discharge the heavy burden they bear of establishing that the usurping regime has established itself to the requisite degree, such that all its acts and laws can be regarded as necessarily legitimate.
 Certain ‘reasons’ are given in the Appellants’ affidavit evidence for the abrogation of the 1997 Constitution and the assumption of power by the Interim Civilian Government with the support of the Military, such as concerns about the workability of the electoral system and that sections of the community were dissatisfied with the policies of the former Government. The answer to such purported grievances and to any ‘discontent’ with the 1997 Constitution (Appellants’ Submissions, paras 29-31) lies in the ballot box and in lawful amendment of the Constitution. These are in any event irrelevant to the question of effectiveness. They are helpful only to the extent that the latter ‘reason’ strengthens the argument that the Interim Civilian Government is not effective because it is supported by one section only of the population rather than by the wider Fijian community.
 The test of effectiveness might be satisfied by, for example, the fact that a long period of time has elapsed since the revolution during which the people of the nation have acquiesced in the new government or by the fact that the new government has been supported by the people at a democratic, and truly representative, election.
Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 125, 128
 Lord Reid applied the effectiveness test in Madzimbamuto to hold that the Smith regime in Southern Rhodesia was unlawful. The United Kingdom Parliament had provided in the Southern Rhodesia Act 1965 (and in the Southern Rhodesia (Constitution) Order 1965) that Southern Rhodesia continued to be part of Her Majesty’s Dominions. Hence (at 725): ‘The British Government acting for the lawful Sovereign is taking steps to regain control and it is impossible to predict with certainty whether or not it will succeed.’ It could not be said that the Smith regime was the lawful government as it was not the ‘effective’ ruler of the colony.
 The decision in Mokotso was influenced by the work of the legal theorist Hans Kelsen, who in General Theory of Law and State portrayed the legal order of a nation as a pyramid of norms derived (through successive levels of derivation) from the norm-creating powers conferred by the Constitution, which itself derived its validity from any earlier constitutional arrangements pursuant to which it was adopted, and ultimately from a Grundnorm or ‘basic norm’ embodying the axiomatic assumption that the legal order was to be obeyed.
 Kelsen argued in Pure Theory of Law at 212, that a legal system is based upon the presupposition or basic norm that ‘One ought to behave according to the actually established and effective constitution.’ Hence, where there has been a revolution or other event that has overthrown the existing constitution and put another in its place, this presupposition may allow the new constitution to become lawful. The new constitution becomes the basic norm of the legal system, thereby replacing the old.
 In Uganda v Commissioner of Prisons  EA 514, the High Court of Uganda considered a resolution passed by the National Parliament of Uganda in April 1966 repealing the 1962 Constitution of Uganda and replacing it with another Constitution. The resolution was not in accordance with the procedures prescribed by the 1962 Constitution for its repeal. Despite this, Sir Udo Udoma CJ on behalf of the High Court, held at 538-539 that the 1962 Constitution had been repealed:
A victorious revolution is an internationally recognised legal method of changing a constitution. Such a revolution constitutes a new law creating organ, by virtue of having become a basic law creating fact ... Our deliberate and considered view is that the 1966 Constitution is a legally valid constitution and the supreme law of Uganda; and that the 1962 Constitution having been abolished as a result of a victorious revolution in law does no longer exist nor does it now form part of the Laws of Uganda, it having been deprived of its de facto and de jure validity. The 1966 Constitution, we hold, is a new legal order and has been effective since April 14, 1966, when it first came into force.
 The Kelsenian reasoning expressed in Mokotso and Uganda v Commissioner of Prisons has been the only way that courts have upheld an otherwise unlawful regime. It is therefore relied upon by the Appellants. But the Supreme Court of Pakistan has found that this is not a means by which an otherwise unlawful revolution might be given legal recognition.
Jilani v Government of the Punjab 1972 PLD SC 139 at 179 per Hamoodur Rahman CJ
Bhutto v Chief of Army Staff 1977 PLD SC 657 at 681, 692 per Anwarul Haq CJ
 The role played by Kelsen’s theory in the Pakistan, Rhodesian and Lesotho cases has been analysed by Justice Cyrus Das (as he now is) in his book ‘Governments and Crisis Powers’ (1996). The thesis that new orders can pull themselves up by the straps of their own jackboots first surfaced in Dosso after President Mirza abrogated the Constitution and took power to himself. (He was overthrown a few days after the court declared his rule efficacious). Kelsen was rejected as a guide by the later Pakistan case of Jilani, and Lord Reid, speaking for the Board in Madzimbamuto, was careful not to mention him by name, or to indicate agreement with his approach, which was not adopted in Mitchell. Justice Das concludes:
Kelsen’s theory has an unacceptable face to it. It endorses unconstitutional behaviour and legalises the actions of usurpers and mutineers. Before long it was rejected in some jurisdictions … (which) it is submitted … had rightly given preference to the necessity doctrine over Kelsen’s thesis. It (the ‘necessity’ doctrine) reflects a more discerning approach to the illegal state of affairs produced by the unconstitutional behaviour of a military dictator. It enables the court to recognise only those acts necessary for the maintenance of law and order and discard actions taken in promotion of the rebellion. Lastly, it enables the judges to remain true to their oath.
 In any event, Kelsenian theory does not apply to the Fiji situation. The basic norm of Fiji has not shifted from the 1997 Constitution as a result of events since 19 May. These events are clearly distinguishable from those that are necessary to lead to a shift in lawful authority as considered in the cases listed above, including Madzimbamuto and Uganda v Commissioner of Prisons. The Appellants have not shown that the Interim Civilian Government is based upon a democratic process, such as national elections, or upon the overwhelming support of the Fijian people.
 The situation in Fiji is unlike that described by Irshad Hasan Khan CJ of the Supreme Court of Pakistan in Zafar Ali Shah v Pervez Musharraf, Chief Executive of Pakistan (Pakistan Petitions Case) (2000) 33 Supreme Court Monthly Review 1137 at 1158-1159 or that by Cullinan CJ in Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 167. In the latter case, the prior regime in Lesotho was characterised by ‘corruption, misappropriation of funds, bloodshed, secret murders made and perpetrated by secret murder squads (Koeeoko), illegal arrests and prison detentions as well as general mass suffering’. The coup that overthrew that regime was ‘greeted with jubilation by the people in the streets’ and apparently was also ‘internationally popular’. By contrast, the purported revolution in Fiji is not a ‘victorious’ or ‘glorious revolution’ to which this Court should accord lawful authority.
 In any event, the Kelsenian efficacy test is not satisfied here for three reasons.
 First, Fiji has not gone through a revolutionary process since 19 May. The correct characterisation is that Fiji has been the subject of a failed coup by the Speight Group, which was dealt with by applying the emergency powers discussed above under the doctrine of necessity. There has been no revolution capable of leading to a transfer of power, only a temporary application of emergency power by the President and the Military that must be limited in operation such that the exercise of the power does not subvert the Constitution that it exists to protect.
 Second, if the actions of the President and the Military after 19 May lie outside the legitimate use of the emergency powers, then this amounts to an attempted revolution that cannot be seen as lawful by this Court because, ‘by and large’, Fijians (including Indian Fijians) have not acquiesced in or behaved in conformity with the mandates of the Interim Civilian Government. The continuing civil unrest in Fiji as well as the discontent evident in large sections of the Fijian community with the events since 19 May demonstrate this. The Interim Civilian Government has been condemned by large sections of the community as well as by many of the most important Fijian non-governmental organisations, such as trade unions and employer organisations. All but one of the major Fijian political parties have refused to co-operate with the Interim Civilian Government and have instead asserted their intention to form a government under the 1997 Constitution.
 The Appellants’ own affidavit evidence suggests that the authority of the Interim Civilian Government does not rest upon a broad foundation of popular acquiescence and lacks support among the wider Fijian community. The Interim Civilian Government depends for support from only one section of the community, comprising some indigenous Fijians.
 This conclusion can be supported, where necessary, by this Court taking judicial notice of ‘notorious facts’ (including those noted by Gates J) as have other courts in like situations. See, for example, Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 136 per Cullinan CJ.
 This second basis allows decisions such as Uganda v Commissioner of Prisons to be distinguished, as that case rests in part on the view of the court at 539 that the new regime had been ‘accepted by the people’ (see also Mitchell v Director of Public Prosecutions  Law Reports of the Commonwealth (Const) 35 at 116 per Liverpool JA).
 Third, the effectiveness doctrine can only be applied when the new regime exists as a matter of historical fact and is no longer subject to competing claims of sovereignty.
 The only case that is arguably to the contrary is the decision of the Supreme Court of Pakistan in State v Dosso 1958 PLD SC 533. In that case the Court upheld a new regime after only 21 days. However, State v Dosso, which was decided before the decision of the Privy Council in Madzimbamuto, was a decision that primarily concerned the effect of a change in the legal order upon certain laws, rather than whether the legal order itself ought to be recognised as lawful. In Jilani v Government of the Punjab 1972 PLD SC 139 at 162 it was suggested by Hamoodur Rahman CJ of the Supreme Court of Pakistan that ‘it does not appear that the question of the validity of the abrogation of the Constitution … was directly put in issue’ (see also at 165, 175). State v Dosso was in any event overruled in Jilani v Government of the Punjab at 183, 248, 257 (which was reaffirmed in Bhutto v Chief of Army Staff 1977 PLD SC 657) and has been doubted elsewhere (Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 109 per Cullinan CJ). In Jilani v Government of the Punjab at 177, Hamoodur Rahman CJ stated that the result in State v Dosso ‘was, to say the least, premature’.
 It cannot be said that the purported Interim Civilian Government has an undisputed claim to power. It has not ousted rival claims from the preceding regime, which in fact continues to claim that it is the legal government of Fiji.
 The ongoing destabilisation campaign waged by the Speight Group suggests that it is also a potential rival to the Interim Civilian Government.
 It is clear that all attempts by the predecessor government to have their lawful power recognised have not effectively ceased. The elected government, in the words of Lord Reid at 725 in Madzimbamuto, is ‘taking steps to regain control’. Thus, as stated by Lord Reid at 725:
If the legitimate Government had been driven out but was trying to regain control it would be impossible to hold that the usurper who is in control is the lawful ruler, because that would mean that by striving to assert its lawful right the ousted legitimate Government was opposing the lawful ruler.
 In these circumstances, this Court cannot apply the approach in Madzimbamuto (for exactly the same reason that it was not applied in that case) and must declare that the purported Interim Civilian Government is unlawful. This is one basis upon which the decision in Uganda v Commissioner of Prisons can also be distinguished. As stated by the Chief Justice of Uganda: ‘The Government of Uganda is well established and has no rival.’ To which Lord Reid in Madzimbamuto at 725 added after quoting this passage: ‘It would be very different if there had been still two rivals contending for power.’
 Similarly, in Mangope v Van der Walt 1994 (3) SA 850, which also upheld the legality of a revolutionary regime, Comrie J at 860 noted that the former President ‘does not ask for the withdrawal of the [new regime]. He does not seek his own reinstatement as President, nor does he seek the re-instatement (expect finally) of his dismissed Ministers’. He was hence ‘not a serious rival for power’ (at 867). See also Mitchell v Director of Public Prosecutions  Law Reports of the Commonwealth (Const) 35 at 116 per Liverpool JA (‘no other contesting legal authority’) and Mokotso v HM King Moshoeshow II  Law Reports of the Commonwealth (Const) 24 at 139 per Cullinan CJ (‘no rival government’). The unusual but significant fact in the Fiji situation is that the lawful government does not seek restoration by force of arms, but by force of law.
of Democracy and Fundamental Human Rights
 The common law by its very nature imposes additional inherent limits on the types of usurping regime which a court ought properly to recognise. This is merely to note that, as indicated by de Smith in the extract quoted above, even if there had been a successful revolution in this case, this would provide no necessary answer to what a judge ought to do. Insofar as Kelsen might be thought to stand against this proposition, his approach can be criticised as confusing ‘is’ and ‘ought’ (see Bhutto v Chief of Army Staff 1977 PLD SC 657 at 688, 721), but it is noted that Kelsen himself in his answer to the criticisms of Julius Stone did not view his theoretical insights as establishing a morality that ought to be followed by judges.
Brookfield, ‘The Courts, Kelsen and the Rhodesian Revolution’ (1969) University of Toronto Law Journal 326 at 342-344
Brookfield, Waitangi & Indigenous Rights: Revolution, Law and Legitimation (1999) at 25
 It is submitted that as stated by Haynes P in Mitchell v DPP at 71i –72f, in order to be accorded legitimacy by the courts, a usurping government must not merely have no rivals and be in effective control, but that:
(c) such conformity and obedience [must be] due to popular acceptance and support and … not mere tacit submission to coercion or fear of force; and (d) it must not appear that the regime was oppressive and undemocratic.
 Similar principles were taken into account by the Supreme Court of Canada in its decision on the right of Quebec to achieve secession from Canada unilaterally (Reference re Secession of Quebec  2 SCR 217 at 247 per the Court). The unanimous decisions of the Court reasoned that the question was to be decided by reference to principles such as ‘adherence to the rule of law, respect for democratic institutions, the accommodation of minorities, insistence that governments adhere to constitutional conduct and a desire for continuity and stability’.
 It is noted that the two latter conditions laid down by Haynes P. in Mitchell were not adopted by the Lesotho High Court and Court of Appeal in Mokotso and Makenete respectively, on the basis that ‘throughout the course of history there have been regimes, indeed dynasties, holding sway for many years, indeed centuries, whose rule could not be said by any manner of means to be popular and could even be described as oppressive: but who is there to say that a new legal order was not created with their coming and going?’ (per Cullinan CJ in Mokotso at p.130a; Makenete at p.55c).
 The appropriate contemporary response to Cullinan CJ’s question is that the courts of the common law exist precisely to say such a thing. It is undoubtedly true that humanity has lived under ‘centuries of oppressive regimes and dynasties’; it can also be said in hindsight that the common law has at times lent aid and comfort to oppressive regimes; but it is submitted that there are principles inherent in the common law which render it inappropriate that it should take a deliberately neutral stance as between oppressive and non‑oppressive regimes. Such principles would include the law’s preference for individual liberty over the absence of rights, its regard for openness over secrecy, and its preference of consistency and equality to arbitrariness, unreasonableness and irrational discrimination.
 It is for this reason that Haynes P’s statement is a correct summary of the law. The common law has only rarely yielded to regimes which have contravened its underlying principles, and when it has done so it has become the agent of its own corruption, as in Uganda following ex parte Matovu and in Pakistan following State v Dosso. A judge’s overriding duty to uphold the common law renders it imperative that he or she conduct an inquiry, once a party to an action has asserted that a contender for power has contravened principles which are fundamental to its very existence.
 In the present case, it is submitted that the court ought properly to take account of the fact that the Interim Civilian Government breaches basic democratic principles accepted in Fiji and internationally by seeking to displace an elected government chosen by the people of Fiji in accordance with the democratically established 1997 Constitution. The 1997 Constitution was developed in consultation with all Fijian communities and is underpinned by a respect for Fijian democracy that is inclusive of all sections of the community and operates for the benefit of the rights and aspirations of all segments of Fijian society. Section 1 of the 1997 Constitution establishes that ‘The Republic of the Fiji Islands is a sovereign, democratic state.’ This is also reflected in the preamble to the Constitution, which states in part:
REAFFIRMING our recognition of the human rights and fundamental freedoms of all individuals and groups, safeguarded by adherence to the rule of law, and our respect for human dignity and for the importance of the family.
 Section 6 of the Constitution also establishes a non-justiciable ‘Compact’, under which:
‘The people of the Fiji Islands recognise that, within the framework of this Constitution and the other laws of the State, the conduct of government is based on following principles: …
(f) the rights of a citizen include the right to form and join political parties, to take part in political campaigns, and to vote and to be a candidate in free and fair elections of members of the House of Representatives held by secret ballot and ultimately on the basis of equal suffrage.
 Contrary to the Appellants’ Submissions, para 69, the Interim Civilian Government has not faithfully ‘re-enacted the human rights provisions of the 1997 Constitution’. Of course, the voting rights and responsibilities set out in the 1997 Constitution (Sections 36, 55 and 56) have been denied because the validity of the Parliament thereby chosen has been frustrated. In other significant respects, there have been further attempts to derogate from the rights listed in the 1997 Constitution. In particular, it is noted that the right to equality has been purportedly limited by Interim Military Government Decree No 7, Fundamental Rights and Freedoms Decree 2000, which restates this right but in section 24(6) of the Decree ‘Subject to any law providing for or protecting the enhancement of Fijian or Rotuman interests’.
 More specifically, the explicitly racist nature of the political changes canvassed by the caretaker regime contravenes customary international law. The Appellants intend to safeguard the ‘paramountcy’ of Fijian and Rotuman interests, and to ensure ‘that the national leadership positions of Head of State and Head of Government’ should never again be held by someone who is not a Fijian or a Rotuman. (See the Blueprint for the Protection of Fijian and Rotuman Rights and Interests, and the Advancement of their Development, presented by the purported Interim Prime Minister, Mr Laisenia Qarase, on 13 July 2000.) It is submitted that the relative disenfranchisement and subordination of any racial group, let alone one which comprises over 40% of Fiji’s population, amounts to degrading treatment contrary to recognised international standards (c.f. East African Asians v United Kingdom, (1981) 3 EHRR 76 at paras. 207-08) and places the country in clear breach of customary international law. As Kirby J of the High Court of Australia stated in Kartinyeri v Commonwealth  HCA 22; (1998) 195 CLR 337 at 419-420:
‘there is one subject upon which the international law of fundamental rights resonates with a single voice it is the prohibition of detrimental distinctions on the basis of race. I consider that Judge Tanaka was correct, in the International Court of Justice [South West Africa Cases (Second Phase)  ICJR 3 at 293], when he declared that: ‘[T]he norm of non-discrimination or non-separation on the basis of race has become a rule of customary international law’.
See also Oppenheimer v Cattermole  AC 249 at 278.
 The purported abrogation and the limitation of certain rights listed in the Bill of Rights in the 1997 Constitution also breaches international human rights norms as expressed in international instruments to which Fiji is a party. Such norms are relevant in this case to guide the court in the exercise of its discretionary powers, and to resolve areas which would otherwise be ambiguous or incapable of resolution by the common law: see e.g. Chung Chi Cheung v The King  UKPC 75;  AC 160 at 168; Trendtex v Central Bank of Nigeria  QB 529 at 553B. (As noted above, Fiji’s 1997 Constitution expressly incorporates such norms into domestic law.) It is also noted that a state’s assumption of international treaty obligations are capable of giving rise to a legitimate expectation in the citizens of that state that those obligations will be observed (see Minister of State for Immigration and Ethnic Affairs v Teoh  HCA 20; (1995) 128 ALR 353 at paras. 28-29), and it is submitted that common law courts are beginning to recognise that, whatever may be the answer to the vexed question of ‘direct enforceability’ of treaty norms, this principle applies even more strongly where the obligations involved concern the protection of human rights (see ibid. at para.28; see also Lewis v Attorney-General of Jamaica  UKPC 35;  3 WLR 1785 at 1810F – 1811A). Even if the Appellants were correct to assert that these norms were ousted by their domestic ‘decrees’, which is of course denied, it is additionally noted that under clearly established international law, ‘A party may not invoke the provisions of its internal law as justification for its failure to perform a treaty’ (see the Vienna Convention on the Law of Treaties 1969, Art.27).
 For all these reasons, it is submitted that the Court should have regard to the following international obligations which have been assumed by Fiji:
a) the International Convention on the Elimination of All Forms of Racial Discrimination A/RES/2106A (XX), 21 Dec. 1965, to which Fiji became a signatory in 1973, and which provides inter alia that, ‘States Parties condemn racial discrimination and undertake to pursue by all appropriate means and without delay a policy of eliminating racial discrimination in all its forms and promoting understanding among all races’ (Art.2), and makes plain that the rights protected from discrimination include ‘political rights, in particular the rights to participate in elections – to vote and to stand for election – on the basis of universal and equal suffrage, to take part in the Government as well as in the conduct of public affairs at any level and to have equal access to public service.’ As the Committee has made clear, this is an integral aspect of the right to self-determination, which is ‘a fundamental principle of international law’ (General Comment, 15/3/96). It is noted that the Appellants purport to rely on their reservation to this Convention to justify their discriminatory actions, but the Court is invited to recognise that this is of no effect, in that such reservations are of effect only insofar as they are not ‘incompatible with the object and purpose of [the] Convention’ (Art. 20), and the Committee itself, in its 1165th Report dated 24/10/96, noted that ‘[Fiji’s] reservations were incompatible with the goal and purpose of the Convention [albeit that] at the time no other State party had made an objection to them.’ It is further noted that insofar as the Appellants seek to rely to justify their Blueprint and other racist measures as permissible affirmative action, they fall foul of Article 4 of the ILO Indigenous and Tribal Peoples Convention 1989 (C169), ratified by Fiji on 3 March 1998, which states among other things that: ‘(3) Enjoyment of the general rights of citizenship, without discrimination, shall not be prejudiced in any way by such special measures.’ It is respectfully submitted that expressing an intention to deny non-Fijians the right ever again to hold political power is about as clear an example as can be imagined of ‘general rights of citizenship’ being ‘prejudiced’.
b) the Universal Declaration of Human Rights, esp. Art. 2 (universal application), Art. 5 (degrading treatment), Art. 7 (equal protection), Art. 8 (effective remedy before national tribunals for breach of fundamental rights), and Art. 21 (political participation). In this context, it is noted that the Fiji government made a statement to the 53rd Assembly dated 7 May 1998, in which it stated that, ‘The Government of Fiji honours strictly its commitments and obligations under the Universal Declaration of Human Rights and, as such, the Fiji Government does not have any legislation which are of coercive nature against any member of the United Nations.’
 The UN Charter, Article 21 of the UDHR and Article 25 of the ICCPR (bolstered by state practice referred to above, the writings of jurists and the decisions of the Human Rights Committee (see its General Comment No.25) confirm that the right to vote and to be elected, and to freely determine political status and choose governments, has crystallized into a norm of international law: see Thomas Franck, ‘The Emerging Right to Democratic Governance’, cited above. It is not contended that the ‘right to democracy’ has entered customary international law in an absolute form, but rather as a normative presumption or rule that individuals (like Chandrika Prasad) or ‘peoples’ who have enjoyed democratic governance may not have it removed by force. State practice in support of the emergence of this rule may be seen in the unanimous vote of the General Assembly for the restoration of the democratic constitution of Haiti in September 1991, and most recently in the UN intervention in East Timor to protect the fledgling right to democratic self-determination.
 There are various ways by which the Court may take cognisance of this development in international law, but it is bound to do so. The test for ‘efficacy’ expounded by Haynes P. in Mitchell is now bolstered and extended by the emergence of the norm. It is contended that no military junta or ‘interim’ government that overthrows a democratic government and lawful constitution should be regarded in law as ‘efficacious’. Protestations by ‘caretakers’ that they will introduce another democratic constitution in the fullness of time cannot be taken at face value, and should not be credited by this Court.
 The Appellants skeleton makes various claims about the IMG policy for the future (para. 5(xvii), 69-73, 92), as does the caretaker Prime Minister Qarese (affidavit). The court should not act upon these claims: ‘emergency rule tends to become a way of life’ (see Das, p.1: ‘only Cincinnatus returned to his farm’). A similar pledge to that made by Qarase was made by General Zia ul-Haq in 1977, and the court in Bhutto unwisely acted upon it, thinking that he would restore democracy. He was still in office, and Pakistan was still under martial law, when he died over a decade later.
 The court must assume that whatever decision it takes will be respected by the police and military powers of the state. It has the opportunity to reaffirm Fiji’s commitment to constitutionality through the continuing vibrancy and vitality of those legal principles which underlie constitutionalism across the common law world. The facts proven by the Respondents in this case do not discharge the heavy burden of showing that there has been a fully successful and broadly recognised revolution, or (as if one prefers, an irreversible alteration of the Grundnorm). For the Court to find otherwise would be to lend premature and inappropriate support to an illegitimate regime. For those reasons outlined in Mitchell and Jilani, this course is not one which should be contemplated.
 This case concerns fundamental human rights, such as racial equality and the concept of non-discrimination, which are the concern of the Respondent, Chandrika Prasad, who has suffered considerable hardship as a result of recent events in Fiji and has had his basic rights violated. In such a case, it was appropriate that Gates J exercise his discretion to make the following orders:
1 The attempted coup of May 19th was unsuccessful.
2 The declaration of the State of Emergency by the President Ratu Sir Kamisese Mara in the circumstances then facing the nation, though not strictly proclaimed within the terms of the Constitution, is hereby granted validity ab initio under the doctrine of necessity.
3 The revocation of the 1997 Constitution was not made within the doctrine of necessity and such revocation was unconstitutional and of no effect. The 1997 Constitution is the supreme and extant law of Fiji today.
4 The Parliament of Fiji, consisting of the President, the Senate, and the House of Representatives, is still in being. Its incumbents on and prior to 19 May 2000 still hold office, that is Ratu Sir Kamisese Mara, who had stepped aside, and who remains President as originally appointed by the Bose Levu Vakaturaga (Great Council of Chiefs); the Senators are still Members of the Senate; the elected Members of Parliament are still Members of the House of Representatives. The status quo is restored. Parliament should be summoned by the President at his discretion but as soon as practicable.
5 Meanwhile owing to uncertainty over the status of the Government, it will remain for the President to appoint as soon as possible as Prime Minister, the member of the House of Representatives who in the President’s opinion can form a government that has the confidence of the House of Representatives pursuant to Sections 47 and 98 of the Constitution, and that government shall be the government of Fiji.
 Such orders were necessary in order to restore the rule of law to Fiji and to return Fiji to the democratic system of government created by the 1997 Constitution. The decision of Gates J to grant the above orders was thus underpinned by considerations of the highest importance.
 The orders made by Gates J cannot be said by this Court to be futile. The leadership of the Military has indicated that it will abide by the decision of this Court. The Parliament of Fiji may have been prorogued, but it has not been dissolved.
 Nor are the orders of Gates J futile if Ratu Sir Kamisese Mara has validly resigned from the office of President. In that case, the situation is governed by Section 88(4) of the Constitution:
If the office of President be comes vacant, a new President and Vice-President must be appointed in accordance with Part 2, but the incumbent Vice-President has the authority under this section to perform the functions of President, for a period of no longer than 3 months, pending the filling of the vacancy.
 If the 3 month period has elapsed and neither the President nor the Vice‑President is available to perform the functions of the President (including those functions specified in the orders of Gates J), such functions may be performed by the Speaker of the House of Representatives in accordance with section 88(3) of the Constitution:
If neither the President nor the Vice-President is available to perform a function of the President, it may be performed by the Speaker of the House of Representatives.
 Hence, it is consistent with the orders granted by Gates J that the Parliament be summoned. The Parliament, once summoned, would have the task of forming a government. In accordance with section 90 of the 1997 Constitution, the President and Vice‑President would then be appointed by the Bose Levu Vakaturaga after consultation by the Bose Levu Vakaturaga with the Prime Minister.
 The decision in Jilani v Government of the Punjab 1972 PLD SC 139 illustrates the dangers of a court not making appropriate orders in a case such as the present. In overruling State v Dosso, in which the Supreme Court of Pakistan recognised a new regime after only 21 days, Yaqub Ali J found at 219 that as a result of the case:
As a commentator has remarked, a perfectly good country was made into a laughing stock. A country which came into being with a written Constitution providing for a parliamentary form of Government with distribution of State power between the Executive, Legislature, and the Judiciary, was soon converted into an autocracy and eventually degenerated into military dictatorship.
 Similarly, Fieldsend AJA of the High Court of Southern Rhodesia in Madzimbamuto v Lardner-Burke 1968 2 SA 284 at 424 stated:
the powers and duties of the separate organs of government are prescribed, and for any organ to act in contravention of those provisions is illegal. This puts a particular responsibility upon the judiciary, which is raised in effect to a position above that of the other elements of government, and which is the only sure guardian of the terms of the constitution and of the trust reposed in the people by the grant of the constitution.
 This Court should dismiss the appeal with costs.
DATED: 13 FEBRUARY 2001 GEOFFREY ROBERTSON QC PROFESSOR GEORGE WILLIAMS ANU PATEL NEEL SHIVAM