Groves, M --- "Treaties and Legitimate Expectations - The Rise and Fall of Teoh in Australia" [2010] UMonashLRS 8

Last Updated: 14 October 2011

Treaties and Legitimate Expectations – The Rise and Fall of Teoh in Australia

Matthew Groves[∗]

This is the pre-peer reviewed version of the following article which has been published in final form at: http://www.ingentaconnect.com/content/hart/judrev/2010/00000015/00000004/art00003

Introduction

1. In an earlier edition of this journal, Sanjay Patel argued that unincorporated treaties could and should provide the foundation for legitimate expectations by individuals seeking judicial review of administrative action.[1] This suggestion drew support from the decision of the High Court of Australia in Minister for Immigration and Ethnic Affairs v Teoh [1995] HCA 20 (1995) 183 CLR 273. That case broke new ground by holding that an unincorporated treaty could found a legitimate expectation that administrative officials would normally act in accordance with the terms of the treaty. Teoh’s case appeared to signal a greater recourse to both legitimate expectations and treaties in administrative decision making but was soon overtaken in Australian by wider constitutional issues in Minister for Immigration and Ethnic Affairs; Ex parte Lam [2003] HCA 6 (2003) 214 CLR 1. In that case the High Court of Australia strongly doubted both Teoh and the doctrine of legitimate expectations that underpinned it, though it seemed mainly concerned to ward off any recourse to the substantive variant of the legitimate expectation that was accepted in R v North and East Devon Health Authority; Ex parte Coughlan [2001] 1 QB 213. Teoh was thus left in limbo in its country of origin, yet it continues to attract interest in other jurisdictions.
   
2. This article examines the Australian response to Teoh and explains why the case waned in Australia. It will be argued that the government and judicial responses to Teoh are contradictory and have failed to confront the central questions posed by Teoh. The article also explains why any substantive version of legitimate expectations remains beyond the reach of Australian courts. Legitimate expectations of both the procedural and substantive kind have become casualties of a much wider constitutional battle. The courts appear to have won the day and secured the central elements of their supervisory role but they have ceded territory in areas such as legitimate expectations.
   

Teoh and its aftermath

3. Legitimate expectations appeared to become redundant in the immediate years before Teoh with the adoption by the High Court of an expansive principle that the rules of natural justice would apply to the exercise of any public power that might “destroy, defeat or prejudice a person's rights, interests or legitimate expectations.”[2] At the same time, the High Court adopted an increasingly strict approach to legislative attempts to limit of exclude the rules of natural justice.[3] The result was that principles governing the implication of any duty to observe natural justice and its possible exclusion were both settled by the adoption of presumptions weighted in favour of individuals challenging government decisions.[4] The legitimate expectation provided a useful device to strengthen otherwise weak claims to specific procedural rights or natural justice more generally but that function appeared to be overtaken by the adoption of a more holistic approach which simply assumed that fairness was presumed to apply and instead asked what it might require in the circumstances.[5]
   
4. Teoh was an unlikely vehicle to revisit the legitimate expectation. Mr Teoh freely admitted he had never heard of the Convention on the Rights of the Child and also that the decision maker had neither adopted a policy nor made a representation based upon the Convention. A majority of the High Court instead accepted that the mere ratification of the Convention was sufficient to create a legitimate expectation that public officials would normally act in accordance with the Convention. This expectation was held to apply when officials proposed to act contrary to the Convention, to require that people affected be given notice of this intention and a chance to argue against it.[6] Mason CJ and Deane J reasoned that such an expectation could arise because:
   

“...ratification of a convention is a positive statement by the executive government of this country to the world and to the Australian people that the executive government and its agencies will act in accordance with the Convention. That positive statement is an adequate foundation for a legitimate expectation, absent statutory of executive indications to the contrary, that administrative decision-makers will act in conformity with the Convention.”[7]

On this view, ratification of an international instrument was a sufficiently clear and serious statement to both the international community and the Australian public to support a legitimate expectation. It might differ in character to the behaviour that had traditionally supported a legitimate expectation, such as the public promulgation by a minister of an official policy or the making of a promise or representation by an agency, but it was deemed to have the same basic character and should therefore be given the same basic effect.

5. The sole dissent in Teoh came from McHugh J, who questioned both the effect given to unincorporated treaties and the coherence of any legitimate expectation founded upon them. His Honour held that the recognition of a legitimate expectation would breach rather than simply circumvent the traditional rule that treaties gained no domestic force until they were given legislative effect.[8] McHugh J also reasoned that any recognition of expectations based upon unincorporated treaties would essentially enable the executive to bypass the parliamentary process and confer a de-facto force upon treaties without the use of legislation.[9] His Honour also fired what would prove to be two prophetic salvos at the reliance the majority judges placed on natural justice. McHugh doubted that any legitimate expectation could provide a significant procedural benefit because the wide application of natural justice in Australia required officials to provide notice of any key issue upon which a decision might turn to the attention of a person affected and provide an opportunity to respond. In that case, his Honour reasoned, the legitimate expectation gave way to the more practical question of “what does fairness require in all the circumstances of the case?”[10] McHugh J also highlighted the circular nature of an expectation in the circumstances that the majority judges accepted it could be evaded. Officials could be required to provide notice that they do not intend to follow a treaty, which they are not strictly obliged to follow in any case, to a person who knows nothing of the treaty that will in any event not be considered.[11] That absurd possibility greatly diminishes any claim that Teoh might foster greater respect for rights or good administration. The discussion of Lam’s case below will explain that these criticisms of the internal logic of the legitimate expectation recognised in Teoh eventually gained traction.
   
6. Teoh is often misunderstood or misused, so it is important to note two key limits in its reasoning. First, the majority judges rejected any suggestion that the novel legitimate expectation they had recognised made the Convention a part of domestic law or somehow directly enforceable. They pointedly affirmed the orthodox principle that international instruments are not directly enforceable in domestic law unless and until they are given legislative effect.[12] The High Court made clear that ratification of a treaty did not confer it with a direct or enforceable quality. Any procedural expectation that administrative officials would normally act in accordance with the terms of a treaty was exactly that: a procedural presumption that could be rebutted generally by the executive or in particular instances by decision makers. In theory the requirements of Teoh can be satisfied if decision makers provide notice that they do not intend to take notice of a principle contained in an unincorporated treaty and provide an opportunity to a person affected to argue why this course should not be taken. The regular suggestion by courts of other jurisdictions that Teoh enables a treaty to “be directly enforceable in domestic law”[13] or gives rise to “enforceable” expectations[14] clearly therefore overstate and oversimplify a central element of the case.
   
7. The second important feature of Teoh was its uncertain scope. Mason CJ and Deane J stressed that ratification of a treaty could give rise to a legitimate expectation “particularly when the instruments evidences internationally accepted standards to be applied by courts and administrative authorities in dealing with basic human rights affecting the family and children.”[15] In the immediate aftermath of Teoh some commentators argued that this passage confirmed the limited scope of Teoh and that the step taken by the majority judges was an incremental rather than radical step.[16] Upon closer inspection, the passage quoted is not so limited. It simply states that a legitimate expectation is more likely to be triggered by ratification of a particular form of human rights treaty, but does not disclaim the possibility that expectations may arise from other human rights instruments or those outside the area of human rights. That possible uncertainty provides a major flaw in Teoh because there is little to indicate when, why or how expectations are more or less likely to arise.
   
8. Any judicial suggestion that legitimate expectations might arise only, or are more likely to arise from, human rights instruments is also problematic. Lord Steyn favoured this view in Re McKerr [2004] UKHL 12 [2004] 1 WLR 807 at [51] when he noted the “growing support for the view that human rights treaties enjoy a special status.” Sales and Clement have argued that the scholarly authority relied upon by Lord Steyn provided no significant reason of authority or principle for a sweeping distinction between human rights and other treaties.[17] The absence of judicial authority alone may not preclude the adoption of such a principle, particularly if judges are able to construct the first footholds from scholarly sources. The distinct and more difficult question which arises from the suggestion by Lord Steyn and Teoh more generally is upon what basis judges can judges claim the expertise and authority to categorise international instruments? Any attempt to do so by reliance on Teoh would require the courts to sift through the many ratifications of treaties by the executive to determine which ones provided a sufficiently “positive statement” to require judicial action.[18] Lord Steyn’s suggestion would require the equally difficult process of devising a judicial taxonomy to determine which unincorporated treaties may have a domestic effect. Neither option would be attractive.
   

The aftermath of Teoh

9. Teoh was that rare administrative law case which attracted widespread public attention. The mostly polemic commentary ranged from one extreme of complaints about judicial trashing of the separation powers to the other of bold predictions on the rebirth of integrity in government. The real consequences of the case were far more mundane but had three key features.
   

(a) The joint Ministerial statements

10. The first was the immediate and sweeping joint statements issued by the federal Attorney-General and Minister for Foreign Affairs. These statements sought to exploit the suggestion of Mason CJ and Deane J the ratification of a treaty could generate a legitimate expectation “absent statutory or executive indications to the contrary...”[19] The joint statements did so in bold terms. They declared that the ratification of a treaty had not ever been and was not now intended to be capable of generating any form of legitimate expectation on the part of administrative officials in the federal government.[20]
    
11. The joint statements attracted considerable attention from scholars and the media but were rarely raised in the courts, which was surprising in light of the explosion of migration litigation in the years after Teoh. The joint statements were never considered by the High Court, so their precise status remains unsettled but it seems clear they did not inflict death by press release upon Teoh. A small number of decisions from the Federal Court of Australia suggested that an executive statement must be timely and precise in order to stymie any legitimate expectation arising from a treaty. On this view, any statement must normally be made at or close to ratification.[21] Very broad or sweeping statements, such as the attempt by the joint statements to address every international instrument, may also not be sufficient.[22] In Lam v Minister for Immigration and Multicultural Affairs [2006] FCAFC 184 (2006) 157 FCR 215 at 226 a Full Court of the Federal Court suggested that an effective executive statement would “need to direct attention to the manner in which decision-makers are required to undertake the task of making particular decisions.” This reasoning suggests that the executive statements must reflect with some precision the environment in which an expectation they seek to frustrate will be administered.[23] It also highlights the curious result that ratification of a treaty may be deemed to create an expectation in general terms which could only be removed by much more precisely focussed behaviour by the executive.
    
    12. The issue of a joint statement by two government ministers highlights a neglected aspect of Teoh that recently divided the House of Lords in another legitimate expectation case. In R v (BAPIO Action Ltd) v SSHD [2008] UKHL 47 [2008] 1 AC 1003 Lord Scott suggested that a legitimate expectation could normally only be constructed by use of a decision-maker’s own assurances or conduct.[24] In other words, the expectations possibly created by one part of government could not be raised against another part. Lord Rodger flatly rejected “not only as a matter of constitutional theory, but as a matter of substance” the idea that the “powers, duties and responsibilities” of government could be segmented when government was ultimately “a single entity.”[25] The latter approach is surely correct and also consistent with other cases that have acknowledged the complexity of modern government. If governments can rely on that complexity to invite the courts to take a more holistic approach to the valid exercise of official authority in cases such as Carltona Ltd v Commissioners of Works [1943] 2 All ER 560 they can hardly complain when the courts allow citizens to view government in the same holistic way. Teoh was notable for treating this assumption as orthodox in the highest levels of government.
        

(b) Reform to the parliamentary scrutiny of treaty making

13. After Teoh Australia gave serious consideration to denouncing the Convention on the Rights of the Child so that it could re-enter the Convention with a reservation to overcome Teoh. This possibility was dismissed, albeit with some reluctance, on the ground that denunciation would greatly damage Australia’s international reputation. [26] The Parliament instead reformed the treaty making process itself, mainly by establishing a Joint Parliamentary Committee (drawn from both Houses of Parliament) to advise upon Australia’s entry into all international agreements. Many commentators have questioned whether the advisory role of the Committee enables it to exert any real influence over the treaty process, particularly when its advice is tendered to the executive which continues to enjoy the same untrammelled discretion to enter a treaty on the terms on whatever terms it wishes. The advisory powers of the Joint Committee also make it powerless to address the continued habit of governments to ratify treaties and then largely ignore them.[27]
    
14. Although the Joint Committee was plainly intended to improve oversight of the treaty process it might also inadvertently foster one part of Teoh, which was the notion that ratification of a treaty was a solemn statement to the wider public and not simply a platitudinous official exercise. That judicial fiction is harder to deny now that the treaty process is overseen by politicians of all persuasions and both chambers of the parliament. On this view, the Joint Committee may not be toothless. It may instead bite the hand that created it.
    
    • (c) The displacement of expectations by detailed guidelines
15. In the aftermath of Teoh the Minister for Immigration promulgated increasingly detailed guidelines to govern the continued stream of cases similar to Teoh, namely the deportation of a non-citizen who had been convicted of substantial criminal offences but who also had significant family ties to Australian citizens.[28] It was eventually accepted that these increasingly detailed documents contained “such an elaborate regime” they extinguished any scope for a legitimate expectation based upon the Convention on the Rights of the Child.[29] The courts seemed comfortable with that outcome because the ministerial directions contained detailed reference to the need to take account of the impact that any decision to deport a person might have on Australian children. The substance of Teoh was therefore largely integrated into the lexicon of migration decision-making by its incorporation into the ministerial policy. It was at this point that the legitimate expectation became relatively easy to displace. The ironic consequence is that the only sure way to effectively displace Teoh was by a process that amounted to an indirect acceptance of the decision.
    
16. The possibility that Teoh could be expunged by ministerial directions was curious on several counts. Although the High Court anticipated that “executive indications” might overcome any legitimate expectation, it was widely assumed that any indication would need sufficiently formal trappings to provide a clear indication of an intention to counteract the consequences of ratification of a treaty. It hardly seemed likely that some of the endless stream of mundane procedural guidelines issued by immigration authorities might succeed where a solemn joint statement by the Attorney-General and Minister for Foreign Affairs might not. More puzzling was the light touch the courts often gave to these ministerial directions. They quickly accepted that the Minister need not follow his own direction and, even if he did, he could misconstrue it without falling into legally reviewable error.[30]
    
17. Although Teoh was clearly a significant decision it largely faded in Australian law. Once the substance of the case had been adopted in migration guidelines, bureaucrats were largely able to adopt a formulaic approach to the case. They would normally make reference to the case as part of their reasoning and pay careful attention to the impact of any migration decision upon children but would then invariably decide deportation of the person in question was the better decision.[31] The High Court did not reconsider Teoh for eight years. During that period the ground completely shifted underneath Teoh. The next section explains that shift and how constitutional considerations served to both empower and constrain the High Court of Australia.
    

The rebirth of constitutional judicial review of administrative action in Australia

18. The seeds of Teoh’s demise were planted in another legitimate expectation several years earlier. In Quin v Attorney-General (NSW) [1990] HCA 21 (1990) 170 CLR 1 a magistrate (Quin) contested a change of policy that frustrated his appointment to a new court that replaced the one upon which he had sat. Quin essentially claimed a legitimate expectation that his application would be treated according to the announced policy of the Attorney which was later changed to his disadvantage. The High Court rejected the claim on many grounds, largely due to its reluctance to accept that a legitimate expectation could be used to enhance a claim to judicial appointment. The enduring feature of the case was the detailed explanation that Brennan J gave of the purpose and scope of judicial review.[32] Brennan J drew upon the famous American case of Marbury v Madison [1803] USSC 16; 5 US 137 (1803) to explain that it was the “province and duty” of the judicial branch of government to declare and apply the law. Brennan J also reasoned that this duty simultaneously defined and constrained judicial power. He explained:
    

“The duty and jurisdiction of the court to review administrative action do not go beyond the declaration and enforcing of the law which determinates the limits and governs the exercise of the repository’s power. If, in doing so, the court avoids administrative injustice or error, so be it; but the court has no jurisdiction to simply cure administrative injustice or error. The merits of administrative action, to the extent that they can be distinguished from legality, are for the repository of the relevant power alone and, subject to political control, for the repository alone.”[33]

35. The most notable feature of this passage and the wider reasoning of Brennan J was to transpose the doctrine of Marbury v Madison from judicial review of legislative action to administrative action. In doing, Brennan J confirmed several important features of Australian judicial review. One was its operation within a constitutional separation of powers that divided and allocated functions to different arms of government. The result imposed necessary limitations upon each of government, including the judicial branch. It also entrenched a sharp distinction between judicial and merits review. Brennan J described the latter as a “forbidden field” to Australian judges for constitutional reasons.[34] On this view, the scope of judicial review was not to be determined by reference to the “protection of individual interests but in terms of the extent of power and the legality of its exercise.”[35] Another striking feature of Brennan J’s reasoning was it is permeated with constitutional principles but the key pages do not contain a single reference to the Australian Constitution. In the absence of direct reference to the federal Constitution it seemed clear that Brennan J anticipated no significant difference between judicial review at the State or federal level.
    
36. The next important step occurred in Craig v South Australia [1995] HCA 58 (1995) 184 CLR 163 when the High Court emphatically endorsed jurisdictional error concept as the device by which the courts could determine the law and police the constitutional boundaries imposed upon tribunals and administrative officials. The distinction between jurisdictional and non-jurisdictional errors that was swept away in England by Anisminic Ltd v Foreign Compensation Commission [1968] UKHL 6; [1969] 2 AC 147 was pointedly affirmed in Craig. The central elements of Brennan J’s reasoning in Quin and the Craig case were then melded in Corporation of the City of Enfield v Development Assessment Commission [2000] HCA 5 (2000) 199 CLR 135. In that case the High Court rejected the doctrine devised by the Supreme Court of America in Chevron USA Inc v Natural Resources Council Inc [1984] USSC 140; 467 US 837 (1984). The High Court reasoned that the deference Chevron granted to administrators to determine competing interpretations of the law was antithetical to the role Australia’s constitutional arrangements granted to executive agencies. The High Court also drew upon the central propositions of Quin’s case to affirm that equivalent restrictions applied to the courts. Just as administrative tribunals and officials could not tread into the judicial terrain of declaring and law, the courts could not undertake the executive function of deciding the merits of administrative decisions.[36]
    
37. The story so far might appear one of judicial restraint and limitation but the benefit of this arrangements was starkly illustrated in Plaintiff S157/2002 v Commonwealth [HCA] 2 [2003] HCA 2; (2003) 211 CLR 476. In that case the High Court faced a privative clause that appeared to oust every form of review of every possible decision that might be made under migration legislation to determine asylum claims. The clause had been drafted with careful reference to a much earlier decision of the High Court which sanctioned the use of privative clauses, so long as the clauses met very particular requirements.[37] The clause in issue in Plaintiff S157/2002 took advantage of the possibility in the widest possible terms and seemed to entirely deprive asylum claimants of any recourse to the courts. The High Court held that privative clauses were to be interpreted against the “minimum entrenched provision of judicial review” introduced by s75(v) of the Constitution. It followed, the Court reasoned, that:
    

“The centrality and protective purpose of this jurisdiction of this Court...places significant barriers in the way of legislative attempts to impair judicial review of administrative action. Such jurisdiction exists to maintain the federal compact by ensuring that propounded laws are constitutionally valid and ministerial and other official action lawful and within jurisdiction...”[38]

The High Court confirmed that the concept of jurisdictional error was central to determining the proper scope of administrative officials. The courts would determine and correct any error of a jurisdictional character as part of their constitutionally mandated role. The High Court also confirmed that the constitutional role it was granted effectively precluded the enactment of any privative clause that purported to limit or exclude this judicial function. In view of the extremely expansive and open ended conception of jurisdictional error adopted by the High Court, the guarantee of judicial review to police jurisdictional error amounted to nothing less than the constitutional entrenchment of judicial review more generally.[39]

22. This emphatic assertion of the entrenched status of judicial review effectively closed a long and intensely fought dispute between the courts and the federal parliament, during which Australia saw the enactment of the strictest privative clauses in the common law world. These clauses were, almost without exception, directed to migration decisions and most commonly asylum claims. Plaintiff S157/2002 marked a constitutional settlement of sorts because the privative clause was not held unconstitutional but was instead interpreted to apply only to exclude review of decisions not infected with jurisdictional error. In short, the facade of the clause remained but its substance was obliterated. In the recent case of Kirk v Industrial Relations Commission of NSW [2010] HCA 1 (2010) 239 CLR 531 the High Court applied similar reasoning to the State Courts. The Court held that the integrated judicial system established by s73 of the Commonwealth Constitution and which invested the High Court with an appellate jurisdiction from the State Supreme Courts presumed that the State courts would continue to both exist and exercise their core function of supervisory review at the State level. It was also held that the jurisdiction of the State Supreme Courts would, like their federal counterparts, be delineated by reference to their function of detecting and preventing jurisdictional error at the State level.
    

The return of the constitutional chickens to the roost in Lam

23. Just a few months after the dust had settled in the constitutional battle culminating in Plaintiff S157/2002, the legitimate expectation returned to the High Court in Minister for Immigration and Ethnic Affairs; Ex parte Lam [2003] HCA 6 (2003) 214 CLR 1. Lam was an unlikely vehicle to revisit the issue because the applicant claimed a denial of natural justice in a very limited basis. Lam was eligible for deportation by reason of his criminal offences and interviewed at length while in prison about his circumstances. He provided detailed information, material from relatives and friends and a letter from the person caring for his children. Government officials sought further information, including contact details for the carer of the children so that the officials could assess Lam’s relationship with the children and the effect his deportation might have upon them. The officials did not contact the carer. They instead collated the information and provided a detailed brief of advice to the Minister who decided to cancel’s Lam’s visa. Lam claimed a denial of procedural fairness and argued that the later letter he received created a legitimate expectation that the Minister would not decide the case until his subordinates had spoken to the carer of Lam’s children. This claim seemed stronger than one than the one in Teoh because it was based upon a direct representation and it did not depend on any treaty. For this reason, Lam did not rely directly upon either Teoh or the newly established substantive expectation of Coughlan [2001] 1 QB 213 but the High Court took the occasion to strongly disapprove of both.
    

(a) Treaties as the foundation for legitimate expectations

41. For present purposes the key findings into Lam can be divided into three parts. The first concerned Teoh and legitimate expectations more generally. Four of the five members of the High Court expressed strong reservations about Teoh but the case was not overruled.[40] McHugh and Gummow JJ doubted whether the ratification of a treaty could provide a “positive statement” of the sort accepted in Teoh. Their Honours also questioned whether courts should interpret or accord meaning to the ratification of treaties. McHugh and Gummow JJ reasoned that the courts also should not “add to or vary” the powers granted to administrative officials “by taking a particular view of the conduct of the Executive of external affairs.” The judicial arm of government, they cautioned, should “declare and enforce the limits of the power conferred by statute upon administrative decision-makers, but not by reference to the conduct of external affairs, to supplement the criteria for the exercise of that power.”[41] This conception of the judicial role clearly draws from the one adopted by Brennan in Quin’s case, in that it envisages a relatively divided and orderly allocation of functions. Whether the implication of a legitimate expectation from the ratification of a treaty necessarily transgresses those boundaries is, however, contestable.
    
42. Callinan J also doubted whether ratification of a treaty could provide a secure foundation for a legitimate expectation, mainly because the process of judicial implication might overlook important issues about the treaty process, such as whether parliament might enact the treaty in whole or in part or whether parliament believed any obligations arising from the treaty might be satisfied by existing domestic law.[42] The underlying concern of Callinan J, which was the potential that judicial recognition of expectation based upon the executive ratification of a treaty might underline the role of parliament, echoed similar concerns of McHugh J’s dissent in Teoh.
    
26. These aspects of the reasoning of McHugh, Gummow and Callinan JJ cast clear doubt over the possibility that ratification of a treaty can and should provide the foundation for a legitimate expectation. Importantly, however, McHugh and Gummow JJ placed considerable reliance on the principles that flow from the structure of the Australian Constitution in this part of their reasoning. Whether that might provide a coherent point of limitation is another matter because the concerns of Callinan J, which question the wisdom of courts ascribing a particular meaning to the ratification of treaties, clearly cannot be confined to Australia.
    

• (b) Doubting the continued role of the legitimate expectation

46. McHugh and Gummow JJ also expressed grave doubt about the continued value of legitimate expectations. Their Honours thought the concept was at best of “limited utility” in light of wider evolution of natural justice. McHugh and Gummow JJ suggested that the legitimate expectation required further consideration if it was to remain a useful legal concept. [43] Hayne and Callinan JJ both also doubted the continued value of legitimate expectations, though to differing extents. Hayne J did not express a concluded view but suggested the concept should not be used “without careful articulation of the content of the principle which is said to be engaged” in any case.[44] Callinan J was openly sceptical about the “invention” of the legitimate expectation and reasoned that the concept was “an unfortunate one, and apt to mislead.”[45] His Honour concluded that Teoh probably represented the “high watermark” of the doctrine, which should in future at least be restricted to cases where people had turned their mind “consciously to the matter...[and] reasonably have believe and expected that certain procedures would be follows.”[46]
    
28. These various remarks are at least cautious and often sceptical of the legitimate expectation. They clearly doubt a central theme of Teoh, which was that government actions such as the ratification of a treaty may provide a signal to the world or the domestic population sufficient to support a legitimate expectation in a particular person. While they did not eradicate the legitimate expectation they clearly confine its use in many situations, including those like Teoh. The logical question that follows is whether that limited room leaves any room at all for the legitimate expectation. Gleeson CJ offered a conception of natural justice that suggested not.
    

• (c) Is natural justice a better fit?

48. Gleeson CJ was the only member of the court in Lam to consider the requirements of natural justice in detail. He characterised the principle of Teoh, so far as it related to fairness more generally, to hold that unfairness will arise when decision-makers informs a person that a particular procedure would be followed and without warning this does not occur. Gleeson CJ reasoned that any such principle was “far too broad”. If it was based on a legitimate expectation, he cautioned, “it required the concept of legitimate expectation to carry more weight than it will bear. If such a proposition were accepted, it would elevate judicial review...to a level of high and arid technicality.”[47] Gleeson CJ instead favoured a more practical conception of fairness. He explained that “Whether one talks in terms of procedural fairness or natural justice, the concern of the law is to avoid practical injustice.”[48] On this view, the important issue was not the nature or source of an expectation but the effect it had. The problem for Lam was that he could not point to any further information that he or the carers of his children could have given to the already detailed information he supplied. The different problem for Teoh, one could speculate, was that he could not explain how the failure to observe a treaty provision he did not know about caused any unfairness to him.
    
30. An important consequence of the reasoning of Gleeson CJ was to shift the onus to prove unfairness squarely upon the person claiming it. This highlights a neglected feature of Teoh, namely a personal and subjective claim to fairness that should ultimately underpin any legitimate expectation was turned on its head. Teoh did not know about the treaty and almost certainly would not have offered further information, yet the failure to draw attention to a matter of no personal knowledge or apparent practical value was held to cause unfairness. The approach of Gleeson CJ requires those who claim a legitimate expectation, or unfairness more generally, to explain more clearly not simply the source of that claim but precisely what unfairness has arisen. Although Gleeson CJ did not address the legitimate expectation or Teoh, there is little doubt that his Honour’s avowedly practical search for unfairness would certainly have led Teoh to fail.
    

• (d) The constitutional heresy of substantive unfairness
  

51. Just as the Teoh and its particular legitimate expectation were not relied upon in Lam, the doctrine of substantive unfairness recognised in R v North and East Devon Health Authority; Ex parte Coughlan [2001] 1 QB 213 was also not raised directly by the parties. Nonetheless the High Court signalled the serious constitutional obstacles that lay before any Australian adoption of Coughlan. The English notion of substantive unfairness is difficult to describe in an outside view but the main concern arises in the third category recognised by the Court of Appeal. This was a promise or practice where the benefit is substantive rather than procedural and the court “will in a proper case decide whether to frustrate the expectation is so unfair that to take a new and different course will amount to an abuse of power.”[49] Neither this passage nor the many later cases that have sought to explain it appear to offer a clear method to determine when or how official behaviour becomes “so unfair” that it requires enforcement.[50] A separate but related concern is that the balancing exercise, to weigh individual and wider public interests, which the Court of Appeal suggested would guide this exercise, appears to edge close or even into merits review.[51]
    
    32. Four members of the High Court hinted that Coughlan was clearly beyond the constitutional reach of Australian courts. Gleeson CJ reasoned that the constitutional jurisdiction granted to the courts “does not exist for the purpose of enabling the judicial branch of government to impose upon the executive branch its ideas of good administration.”[52] On this view, the normative concepts of good administration and the like that have been offered in cases following Coughlan to explain the ultimate foundation of substantive unfairness are ones which the judicial branch of Australian government cannot enter. McHugh and Gummow JJ, with whom Callinan J agreed on this point, similarly reasoned that Coughlan and later cases appeared

“...concerned with the judicial supervision of administrative decision-making by the application of certain minimum standards now identified by the English common law. These standards fix upon the quality of the decision-making and thus the merits of the outcome...this represents an attempted assimilation into the English common law of doctrines derived from European civilian systems.”[53]

Their Honours continued:

“In Australia, the existence of a basic law which is a written federal constitution, with separation of the judicial power, necessarily presents a frame of reference which differs from both the English and other European systems...An aspect of the rule of law under the Constitution is that the role or function of Ch III courts does not extend to the performance of the legislative function of translating policy into statutory form or the executive function of administration.”[54]

33. These passages draw together several different threads. The first is that English and Australian public law have clearly separated and will remain so for the foreseeable future. The High Court of Australia regularly attributes this to the “tectonic shifts” caused by the increasing influence of European principles on English law.[55] Sir Robin Cooke similarly suggested that the common law is becoming less English.[56] The better view is that the United Kingdom and other Commonwealth countries are now guided by different principles and the combined effect of those principles is to draw both bodies of law in different directions. In the case of Australia, the dominant influence in public law is clearly the Constitution. Any principle of judicial review must conform to constitutional doctrine or face inevitable rejection. The Constitution entrenches not only a minimum protected level of judicial review of administrative action but has also given rise to a form of judicial power with clear limits. The principles that flow from the conception of judicial review expounded by Brennan J in Quin’s case simultaneously define and confine judicial review. They demarcate the boundaries within which the judicial function is necessarily exclusive and secure but they also mark out other boundaries beyond which the courts cannot step. The most obvious is the traditional divide between judicial and merits review but, according to Lam, this almost certainly includes those grounds of judicial review such as substantive unfairness that draw the courts uncomfortably close to merits review. Proportionality would also almost certainly fall within the same prohibited category. It is clear that the minimum entrenched level of judicial recognised in Plaintiff S157/2002 is accompanied by important limitations. The price of secured judicial power was the effective surrender of some of the more recently developed grounds of review.
    
34. The apparent winding back of Teoh by Lam is less explicable by Australia’s constitutional arrangements. It seemed more likely that the court doubted both the coherence of any claim to a legitimate expectation based upon the ratification of a treaty and also the practical value that any such expectation might occupy in light of the expanded Australian approach to natural justice. The emphasis of Gleeson CJ upon the practical focus of fairness, simply highlights the inherent contradiction in Teoh, namely the legitimacy of an expectation based upon an act (whether the ratification or a treaty or from some other source) of which its claimant is unaware. How can people legitimately expect anything based upon something they do not know of? The answer is simple. The law allows the invocation of all manner of common law principles and presumptions that the people in whose favour they operate are almost always blissfully unaware. The difficulty arises when this possibility is characterised as an expectation. The notion is artificial and inevitably tenuous.
    

Is there a possible solution to the use of unincorporated treaties?

35. Taggart long ago pointed to a possible solution when he acknowledged the many internal contradictions of Teoh and suggested that the case might provide a “wobbly stepping stone” to a more coherent principle in which treaty issues could become relevant considerations in some administrative decision making.[57] To regard treaty issues as relevant considerations would overcome any artificial connection between treaties and legitimate expectations, or natural justice more generally. McHugh and Gummow JJ were plainly unsympathetic to this possibility in Lam,[58] mainly because their Honours thought the possibility was also implicitly rejected in Teoh. It is also worth noting that Taggart did not explain when or how treaty considerations might become relevant. Perhaps the solution might be to regard treaty issues relevant considerations when they deal with fundamental rights or freedoms. This possibility would admittedly require further consideration of what treaties do and do not concern fundamental rights or freedoms and also the argument of Sales and Clement that no such taxonomy in treaties does or can exist. But the nomenclature of rights and freedoms did not trouble acceptance of the principle of legality, so one can question whether it can provide a necessary stumbling block to any recourse to treaties. This possibility would enable the courts to consider the central question raised by Teoh, which was arguably ignored in the barrage of technical objections taken to the case in Lam. Should public officials who exercise public powers be required, in some limited circumstances yet to be finally settled, to consider whether to take account of fundamental principles that are contained in treaties to which their country is a signatory? People might disagree on the answer but few would disagree that the question should be asked more directly than it was in Teoh.

---

[∗] Monash University, Australia.

[1] Sanjay Patel, ‘Founding Legitimate Expectations on Unincorporated Treaties’ [2010] Judicial Review 74.

[2] Annetts v McCann [1990] HCA 57 (1990) 170 CLR 596 at 598; Ainsworth v Criminal Justice Commission [1992] 10 [1992] HCA 10; (1992) 175 CLR 564 at 577.

[3] Annetts v McCann [1990] HCA 57 (1990) 170 CLR 596 at 598-9; Re Minister for Immigration and Multicultural Affairs; Ex parte Miah [2001] HCA 22 (2001) 206 CLR 57, at 73-5, 84-5, 95-8, 111-5.

[4] Legislative attempts to exclude the duty to observe natural justice abound in Australian migration law, in the form of schemes that seek to create an exhaustive framework intended to exclude recourse to the common law. The High Court invariably finds schemes are not sufficiently clear or exhaustive. See, eg, Saeed v Minister for Immigration and Citizenship [2010] HCA 23 (2010) 267 ALR 204.

[5] A point laboured by McHugh J in Teoh [1995] HCA 20 (1995) 183 CLR 273 at 311-2.

[6] Teoh [1995] HCA 20 (1995) 183 CLR 273 at 291-2 per Mason CJ and Deane J.

[7] [1995] HCA 20 (1995) 183 CLR 273 at 291.

[8] [1995] HCA 20 (1995) 183 CLR 273 at 315-9.

[9] [1995] HCA 20 (1995) 183 CLR 273 at 316.

[10] [1995] HCA 20 (1995) 183 CLR 273 at 312.

[11] [1995] HCA 20 (1995) 183 CLR 273 at 314.

[12] [1995] HCA 20 (1995) 183 CLR 273 at 287-8, 291 per Mason CJ and Deane J, (Gaudron J agreeing on this point), 302 per Toohey J.

[13] In the matter of an Application by MacFiolla Cathain [2009] NIQB 66 at [36] per Treacey J.

[14] European Roma Rights Centre v Immigration Officer at Prague Airport [2003] EWCA Civ 666 [2003] 4 All ER 247 at [101] per Laws LJ. The description is particularly inapt for Teoh, given the explanation below that Australian law does not allow substantive expectations of any form.

[15] Teoh [1995] HCA 20 (1995) 183 CLR 273 at 291. Gaudron J agreed generally with Mason CJ and Deane J but placed greater weight on the separate citizenship rights she accorded to Mr Teoh’s children: 304. While this reasoning identifies a quite separate base for the expectation, her Honour’s strong reliance on the vulnerable status of the children and the impact of the deportation of their father upon the wider family reinforce is broadly consistent the focus of Mason CJ and Deane J upon specific categories of treaties. Later cases have acknowledged that this possibility has not taken root. See, eg, Lam v Minister for Immigration and Multicultural Affairs [2006] FCAFC 184 (2006) 157 FCR 215 at [35].

[16] Kristen Walker and Penny Mathew, “Minister for Immigration v Ah Hin Teoh” [1995] MelbULawRw 16; (2005) 20 MULR 236, 246-7, 249.

[17] Philip Sales and Joanne Clement, “International Law in Domestic Courts” (2008) 124 LQR 388 at 398-9.

[18] McHugh J expressed a similar concern on the difficulties that bureaucrats would face in deciding, first, which treaties to consider and, secondly, whether and how particular treaties might give rise to legitimate expectations: Teoh [1995] HCA 20 (1995) 183 CLR 273 at 316-7. His Honour also noted that the task was made more difficult by the more than 900 treaties to which Australia was a party at that time. The same problems could easily vex judges.

[19] Teoh [1995] HCA 20 (1995) 183 CLR 273 at 291.

[20] Joint Statement by the Minister for Foreign Affairs and the Attorney-General, “International Treaties and the High Court Decision in Teoh” (10 May 1995) and ‘The Effect of Treaties in Administrative Decision-Making’ (25 February 1997). The first statement was issued by a Labour government and the second by a conservative Liberal government. The preparedness of both sides of politics to disclaim Teoh was a rare example of political consensus in Australia.

[21] Department of Immigration and Ethnic Affairs v Ram (1996) 69 FCR 431 at 437-8. Some exceptions to this general temporal requirement were canvassed in Lam v Minister for Immigration and Multicultural Affairs [2006] FCAFC 184 (2006) 157 FCR 215 at 227.

[22] Tien v Minister for Immigration and Multicultural Affairs [1998] FCA 1552 (1998) 89 FCR 80 at 103.

[23] See also Tien v Minister for Immigration and Multicultural Affairs [1998] FCA 1552 (1998) 89 FCR 80 at 103; Browne v Minister for Immigration and Multicultural Affairs (1998) 27 AAR 353 at 369.

[24] [2008] UKHL 47 [2008] 1 AC 1003 at 1016, citing R v Secretary of State for the Home Department; Ex parte Mapere [2000] EWHC 633; [2001] Imm AR 89 at [34], [36].

[25] [2008] UKHL 47 [2008] 1 AC 1003 at 1018.

[26] Parliament of Australia, Joint Standing Committee on Treaties, ‘United Nations Convention on the Rights of the Child’ (17th Report, 1998) at 65-6 (available at http://www.aph.gov.au/house/committee/jsct/reports/report17/rept17ex.pdf)

[27] See Ben Saul, ‘Administrative Law and Human Rights’ in Matthew Groves and HP Lee (eds), Australian Administrative Law: Fundamentals, Principles and Doctrines (Cambridge UP, 2007).

[28] Migration Act 1958 (Cth) s499 grants the Minister a broad power to issue directions to migration officials. These directions are subject to parliamentary oversight.

[29] See, eg, Baldini v Minister for Immigration and Multicultural Affairs [2000] FCA 173 at [30].

[30] Hopkins v Minister for Immigration and Citizenship [2007] FCA 1101 at [40].

[31] The formulaic nature of the reasons of migration officials increased shortly after Teoh due to a separate decision of the High Court essentially accepted the use of standard form paragraphs by migration officials: Minister for Immigration and Ethnic Affairs v Wu [1996] HCA 6 (1996) 185 CLR 259.

[32] This description is especially appropriate because the judgment plainly echoes an earlier speech of Brennan J published as ‘The Purpose and Scope of Judicial Review’ (1986) 24 Australian Bar Review 93.

33. [33]Quin [1990] HCA 21 (1990) 170 CLR 1 at 35-6.
34. [34]Quin v Attorney-General (NSW) [1990] HCA 21 (1990) 170 CLR 1 at 38.
35. [35]Quin v Attorney-General (NSW) [1990] HCA 21 (1990) 170 CLR 1 at 36.
36. [36]Corporation of the City of Enfield v Development Assessment Commission [2000] HCA 5 (2000) 199 CLR 135 at 152-4 [43]- [44] per Gleeson CJ, Gummow, Kirby and Hayne JJ, citing Attorney-General (NSW) v Quin [1990] 21 HCA (1990) 170 CLR 1 at 35-6 per Brennan J.
37. [37]R v Hickman; Ex parte Fox and Clinton [1945] HCA 53 (1945) 70 CLR 598.
38. [38]Plaintiff S157/2002 v Commonwealth [HCA] 2 [2003] HCA 2; (2003) 211 CLR 476 at 513-4 [104] per Gaudron, McHugh, Gummow, Kirby and Hayne JJ.
39. [39]The High Court has repeatedly confirmed that the categories of error that may be jurisdictional is not closed. See, eg, Kirk v Industrial Relations Commission of NSW [2010] HCA 1 (2010) 239 CLR 531 at 573-4 [72] per French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ.
40. [40]Teoh was likely saved because Lam was only heard by a bench of five of the seven members of the High Court. Any decision to overrule a relatively recent case of the High Court would clearly have required all of its members to preside.
41. [41]Lam [2003] HCA 6 (2003) 214 CLR 1 at 33-4 [102].
42. [42]Lam [2003] HCA 6 (2003) 214 CLR 1 at 49 [152].
43. [43]Lam [2003] HCA 6 (2003) 214 CLR 1 at 16 [61]- [64].
44. [44]Lam [2003] HCA 6 (2003) 214 CLR 1 at 37 [119].
45. [45]Lam [2003] HCA 6 (2003) 214 CLR 1 at 45-6 [140]- [141].
46. [46]Lam [2003] HCA 6 (2003) 214 CLR 1 at 47 [145].
47. [47]Lam [2003] HCA 6 (2003) 214 CLR 1 at 9 [25].
48. [48]Lam [2003] HCA 6 (2003) 214 CLR 1 at 14 [38]. Hayne J similarly suggested that fairness should focus simply on “the fairness of the procedures adopted...If the procedure was fair, reference to expectations, legitimate or not, is unhelpful, even distracting”: ibid at 35-6 [111].
49. [49]Coughlan [2001] 1 QB 213 at 242.
50. [50]That difficult was amplified by the Court of Appeal’s assertion that it was clearly for the court to pronounce “whether the consequent frustration of the individual’s expectation is so unfair as to be a misuse...of power”: Coughlan [2001] 1 QB 213 at 251. This approach appears to offer little more than the court’s own opinion as a touchstone.
51. [51]A pointed tacitly conceded when the Court of Appeal acknowledged it would take account of the fairness of any outcome: Coughlan [2001] 1 QB 213 at 246.

[52] Lam [2003] HCA 6 (2003) 213 CLR 1 at 11-12 [32].

[53] Lam [2003] HCA 6 (2003) 213 CLR 1 at 23-4 [73].

[54] Lam [2003] HCA 6 (2003) 213 CLR 1 at 24-5 [76].

[55] A phrase recently used by Gummow ACJ and Keifel J in Minister for Immigration and Citizenship v SZMDS [2010] HCA 16 at [21].

[56] A point made by Sir Robin Cooke in “The Road Ahead for the Common Law” (2004) 53 ICLQ 273 at 274.

[57] Michael Taggart, ‘Legitimate Expectation and Treaties in the High Court of Australia’ (1996) 112 LQR 50 at 54.

[58] Lam [2003] HCA 6 (2003) 213 CLR 1 at 24-5 [76].