Author:	Naomi Sidebotham BA, LLB (Hons), LLM Lecturer in Law, Murdoch University School of Law
Subjects:	Administrative law -- Australia (Other articles) Judicial Review - Australia
Issue:	Volume 8, Number 1 (March 2001)
Category:	Refereed Articles

Contents:

• Introduction
• The interpretation of unreasonableness
• Introduction
• Associated Provincial Picture Houses v Wednesbury
• Early interpretations of unreasonableness
• Unreasonableness reassessed
• Conclusion
• The limits of judicial review
• Proportionality
• Conclusion
• Notes

Introduction

1. As a basis on which to review exercises of administrative discretion unreasonableness has proven to be one of the most problematic areas of administrative law. It raises issues concerning the extent to which both constitutional and practical limitations of judicial power are maintained by the courts when exercising this ground. This is because unreasonableness as a ground of review is difficult to define with any certainty or clarity. One of the most oft cited formulations is that of Lord Greene MR, that a decision is unreasonable if "no reasonable authority could ever come to it."[1] Despite many attempts to refine the concept it remains ill-defined and imprecise. All of the tests are circular; a decision is unreasonable if it is manifestly absurd, if it is "outrageous in its defiance of logic" or if a reasonable person would not have come to it.[2] Consequently, in seeking to maintain the legality of executive decision making on the basis of something as vague as unreasonableness the courts may come dangerously close to reviewing the merits of the decision under review, and so intervene or not based on their opinion of the reasonableness of this decision, rather than its legality.

2. Recently, however this ground of review has come under close scrutiny from both the judiciary and the legislature. The latter has truncated the grounds of review in relation to some decisions under the Migration Act 1958 (Cth). Namely, decisions of the Refugee Review Tribunal which are judicially reviewable to the Federal Court cannot be reviewed on the basis of unreasonableness.[3] In those areas where unreasonableness does remain as an available ground of review both the High Court and the Federal Court have unequivocally held that as a ground of review its scope must be significantly curtailed.[4] In an apparent affirmation of Lord Greene's original intent both courts have insisted that the circumstances in which a decision can be held unreasonable must truly be overwhelming. Unreasonableness is intended to be a safety net only. The courts must preserve the limits of judicial power. They must not use unreasonableness as a guise for hearing an appeal and so review the merits of a case and substitute their own decision for that under review. In exercising this ground of review courts must exert appropriate restraint.

3. The question which demands attention as a result of these developments is thus what role, if any, does or should unreasonableness play as a ground of judicial review? It is argued here that unreasonableness is a very difficult standard against which to measure the legality of a decision but that it nevertheless provides a legitimate and useful basis for review. This article examines both the limitations that must be respected by the courts when concluding that a decision is or is not unreasonable and judicial interpretation and application of this ground of review. In particular it analyses the recent case law which advocates a restriction on the application of unreasonableness. It is argued here that these restrictions may be too severe, as unreasonableness is an important means by which the courts can protect the rights of the individual. It is concluded that there is a vital role for unreasonableness as a ground of review and that while it must be exercised with caution and restraint it must not be restricted so as to become essentially otiose.

   The interpretation of unreasonableness

   Introduction

4. In this section the development of unreasonableness as a ground of review is examined. There is a brief discussion of its emergence in the Wednesbury case and subsequent development by Australian courts followed by an analysis of recent High Court and Federal Court decisions which have sought to limit this ground of review. The focus of discussion is on establishing that unreasonableness is a useful ground of review but that the lack of clarity in its definition can impact significantly on its efficacy. It is also argued that while it is essential for courts to respect the limitations of this ground of review it should not be limited to such an extent that it becomes ineffective.

   Associated Provincial Picture Houses v Wednesbury

5. It was Lord Greene MR who firmly introduced the concept of unreasonableness as a ground of review in Associated Provincial Picture Houses v Wednesbury.[5] It was in this case that a regulation preventing the admission of any child under the age of 15 into the local picture house was castigated as unreasonable. In arriving at his conclusion Lord Greene presented unreasonableness as having two meanings. The first was merely as a general rubric containing the other grounds of review, such as relevancy and impropriety of purpose.[6] However, it is evident from his proposition that "if a decision on a competent matter is so unreasonable that no reasonable authority could ever come to it then the court can interfere," that he envisaged a second meaning of unreasonableness, that is, as an independent substantive ground of review.[7] He emphasised that to prove unreasonableness in this sense would require something overwhelming. It was intended only as a safety net to catch those decisions that were manifestly absurd but may escape review on any of the more specific grounds of attack.[8] Hence as a substantive ground of review unreasonableness would have only very limited application. From here unreasonableness has continued to develop as a legitimate basis upon which to challenge an exercise of administrative discretion. In the fifty years that have passed since its emergence there has been much debate, both academic and judicial, over the definition and appropriate scope of unreasonableness as a ground of review

   Early interpretations of unreasonableness

6. As indicated above one of the most intractable problems of unreasonableness is that of defining it with any certainty. The formula appears simple. It is certainly easy to articulate. Determining the exact meaning of this articulation, however, is not so simple. Lord Greene's formulation given above, while a convenient starting point, does not provide any real guidance on how it should be applied other than only in very rare circumstances. He defines unreasonableness simply as something that is not reasonable. Australian courts, while emphasising that they must display restraint when exercising this ground of review, have generally adopted Lord Greene's definition. This is also the definition that is incorporated into the Administrative Decisions (Judicial Review) Act 1977 (Cth).[9] Clearly, this conception of unreasonableness is inherently indeterminate. Whether or not a particular decision is or is not unreasonable is often nothing more than a question of opinion and degree. The difficulty inherent in the application of such a criteria must in consequence raise questions over its effectiveness. Can a ground of review that cannot be defined with any certainty be considered a viable and useful basis upon which to review an exercise of administrative discretion? Analysis of a series of cases in which Australian courts made it clear that they were happy to adopt an expansive reading of this ground of review raises these very concerns. While this does not of necessity mean that as a ground of review unreasonableness should be abandoned, it does clearly demonstrate that examination and consideration as to its scope is required.

7. The effectiveness of unreasonableness as a ground of review was blatantly called into question in Chan v Minister for Immigration and Ethnic Affairs.[10] In this case the refusal of the Minister to grant an entry permit into Australia to Mr Chan was clearly unreasonable. This indeed was the conclusion of the High Court. It was not, however, the conclusion of the Federal Court. The reasonable minded judges adjudicating over the reasonableness of a ministerial determination were unable to arrive at a common consensus as to what was beyond the limits of reason. In this case the Minister had the power, pursuant to the Migration Act 1958 (Cth), to grant an entry permit if the applicant "has the status of a refugee." This is defined as arising if there is a "well founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion."[11] Here the Minister determined that the evidence to the effect that the appellant, a Chinese national, had been persecuted by six years imprisonment and exile as a result of the anti-revolutionary opinion held by his family was not sufficient to satisfy this requirement. Although he "may have been discriminated against to a limited degree due to the apparent perception the local authorities had of his family," this could not be considered persecution. Nor could it be considered that any fear held by Mr Chan was well founded, since he had indicated to immigration officials that if he were to be deported his preferred destination was China not the available alternatives of Macau or Hong Kong. Following an extensive analysis of what was meant by a "well founded fear of persecution" the High Court unanimously held that it was unreasonable for the Minister to hold that Mr Chan should not be granted refugee status. Their Honours held that discrimination of the kind in question did amount to persecution, and that as there existed a "real chance" of it continuing should Mr Chan be returned to China "it would be unreasonable to hold that [his] fear of further exile or detention was not well founded."[12] Unreasonableness was applied as a safety net to a decision that was manifestly absurd. On its face this case would seem to present a good example of unreasonableness being applied in exactly the kind of situation that Lord Greene MR envisaged. Indeed the conclusion is probably a good one.

8. Yet the Federal Court did not determine that the Minister's conclusion was unreasonable. The High Court and Federal Court heard the same arguments and the same evidence and yet reached opposing conclusions. The High Court was of the opinion that it was absurd, given the evidence, to conclude that there was not a well founded fear of persecution. In contrast, the Federal Court determined that the decision not to grant Mr Chan refugee status was a decision that was open to a reasonable mind and that as such it was not for the Court to interfere in this decision. Clearly, the efficacy and utility of unreasonableness as a ground of review must be questioned.

9. In a series of other cases including Prasad v Minister for Immigration and Ethnic Affairs,[13] Luu v Renevier[14] and Minister for Aboriginal Affairs v Peko-Wallsend[15] it is apparent that an expansive reading of unreasonableness is preferred. In Prasad Wilcox J held that a determination was unreasonable because there was readily available evidence that contradicted the finding of a ministerial delegate. It was the failure of the delegate to conduct further inquiries and have regard to this evidence that meant that his decision was unreasonable. In Wilcox J's opinion the failure to make inquiries where further information is both important and easily accessible renders a decision so unreasonable that no reasonable person could have come to it. Wilcox J's interpretation of unreasonableness would certainly seem wider than that envisaged by Lord Greene. Likewise, in Luu the full Federal Court found that the failure of the delegate to consult all of the available evidence was an unreasonable exercise of discretion. He had failed to consult material that was, to his knowledge, readily available to him and likely to be of critical importance in making the determination.

10. Yet in neither of these cases can it be said that the delegate's decision evidenced something "overwhelming" or "manifestly absurd." A decision was simply made on the basis of evidence furnished by the applicants. At some level it can be argued that the Federal Court simply reassessed the evidence before the respective decision makers and formulated its own conclusions on its own interpretations of the evidence. In its opinion further inquiries should have been conducted. It is difficult to conclude that the delegates' decisions were so unreasonable that no reasonable person could have come to them. Unreasonableness was simply expanded to include this duty to make inquires and consult all available evidence before arriving at a final decision. Likewise in Peko-Wallsend it was expanded by Mason J to encompass the situation where a decision maker has accorded inappropriate weight to the issue before him or her. He stated that if a decision maker has given too much weight to an irrelevant factor or too little weight to a relevant one then the decision is "manifestly unreasonable."[16] Clearly it would seem in these cases that there was a considerable move away from the stringent ground of review envisaged by Lord Greene MR.

11. The wider the interpretation unreasonableness receives the more dangerous it becomes as a ground of review as courts will be given greater opportunity to intervene in the merits of a decision. It will also become less effective as it is applied with less consistency. Indeed, even in cases such as Prasad and Luu the courts have warned that they must not overstep the limits of their powers. They must not use unreasonableness as a guise for reviewing the merits of a decision. It is such concerns that have led to recent attempts to restrict and more clearly define this ground of review.

    Unreasonableness reassessed

12. In a number of recent cases both the High Court and the Federal Court have advocated restraint and the test for unreasonableness has been re-stated in the restrictive and extreme language of Lord Greene MR. The first of these cases, Minister for Immigration and Multicultural Affairs v Eshetu, concerned an application by an Ethiopian national for a protection visa under the Migration Act 1958 (Cth).[17] The central issue in the case was whether or not the respondent could be accorded refugee status, that is whether he had a "well founded fear of being persecuted." The Minister rejected his application and a review of that decision by the Refugee Review Tribunal was not successful. The respondent sought review of the Tribunal's decision in the High Court arguing that it had reached a decision that was so unreasonable that no reasonable authority could have come to it.

13. The respondent argued that he had been involved in anti government activity in Ethiopia, as a consequence of which he and a number of others were arrested, starved, beaten and threatened with death if they continued with their political activity. He also submitted that since he left Ethiopia those who were arrested with him had disappeared or been killed and that consequently he was too scared to return. The Tribunal found that the respondent did not have a well-founded fear of persecution. It had been unable to find any independent corroboration of his account of the arrests, deaths and disappearances. Also information from a number of government sources, such as the Department of Foreign Affairs and Trade, indicated that although some of those who continued to oppose the government would be at serious danger most would not. The Tribunal did not believe that the respondent would be someone who would be in danger, as he had not been a prominent member of a political organisation or an organisation that advocated violence.

14. The High Court rejected the argument that the Tribunal had acted unreasonably. In reaching this conclusion the Court did not focus on the facts but rather on the reasoning adopted by the Tribunal and, most importantly, on the scope and definition of unreasonableness as a ground of review. In particular, the Court prescribed very limited boundaries for unreasonableness insisting that it is to be used only in the most extreme of circumstances. Gleeson CJ and McHugh J in particular emphasised that when using unreasonableness a court must exercise caution as it must be vigilant to examine only the legality of an impugned decision. Unreasonableness must not be used as a guise under which courts seek to exercise review of the merits of a decision. They state that:

    Someone who disagrees strongly with someone else's process of reasoning on an issue of fact may express such disagreement by describing the reasoning as "illogical" or "unreasonable" or even "so unreasonable that no reasonable person could adopt it." If these are merely emphatic ways of saying that the reasoning is wrong, then they may have no particular legal consequence.[18]

    In order to satisfy the unreasonableness test there must indeed be something more than mere divergence of opinion. There must be something overwhelming.[19] To prove something overwhelming the evidence needs to definitively indicate only one possible conclusion and not be supportive of alternative conclusions. This was not the case here. There was some probative material and some logical support for the Tribunal's conclusions. It is not for the courts to progress beyond this and express disagreement and then castigate the decision as unreasonable.

15. This conception of unreasonableness as laid down in Eshetu has been applied by the Federal Court in two recent decisions. The first of these, Minister for Immigration and Multicultural Affairs v Betkhoshabeh involved an Iranian national who was granted refugee status in Australia.[20] While in Australia, however, he developed a psychiatric illness which manifested itself as paranoid delusions directed towards his interpreter. He was convicted of unlawfully being on her premises and threatening to kill her and sentenced to three and a half years imprisonment. This prompted the Minister to determine that he should be deported.[21] This decision was upheld by the Administrative Appeals Tribunal. The Tribunal found that the respondent may well suffer discrimination if returned to Iran, but that his "life or freedom would not be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion."[22] The primary judge in the Federal Court found the Tribunal's decision to be unreasonable. This was reversed by the Full Court. It determined that the Tribunal had not acted unreasonably in upholding the Minister's decision to deport the respondent.

16. In reaching its conclusion the Court once again emphasised the difference between reviewing questions of law and interfering with questions of fact. The role of a court is to maintain legality. It must not transgress into the merits of a decision. The Court applied Eshetu saying that unreasonableness cannot be used simply because there is disagreement with someone else's reasoning. Again, the Court looked at whether or not the evidence was supportive of only one conclusion. In the present case it was not. There was evidence that deportees from other countries were not suffering discrimination upon return to Iran. Consequently, the court determined it was "not perverse for the Tribunal to reach the conclusion it did."[23] It was for the Tribunal not the Court to formulate a conclusion on the evidence. The conclusion it reached may have been debatable but it was certainly not inconceivable. This case clearly marks a turnaround from the approach of earlier cases, such as Prasad.

17. This stringent approach received further reinforcement in Cong Tam Dong v Minister for Immigration and Multicultural Affairs.[24] The appellant in this case had come to Australia in 1984. He was later convicted of a "serious offence," namely a home invasion. Once his term of imprisonment was completed, a deportation order was issued against the appellant in 1997. This decision was approved by the Administrative Appeals Tribunal. The appellant sought review of this decision asserting that the Tribunal had acted unreasonably in arriving at this conclusion.

18. The appellant argued before the Tribunal that his participation in the offence which led to the deportation order was due to the pressure exerted by his co-offenders as he was concerned for his safety if he did not agree to participate. He also argued that as his family was in Australia and as he had spent much of his life in Australia he should be permitted to remain here. The Tribunal dismissed these arguments. It was concerned by the fact that the appellant had a criminal record and that there was a considerable risk of his re-offending. It also said that although he had been in Australia for over half his life and although his family was a decent family, he should be deported. The appellant argued that this was unreasonable as the Tribunal had merely substituted its own opinion for that of the expert evidence of psychologists and parole officers and that there was no factual basis to support the Tribunal's conclusion. The full Federal Court rejected the appellant's argument holding that the Administrative Appeals Tribunal had not acted unreasonably. The reasoning is very much in the vein of the earlier two cases. It stated that the Tribunal was entitled to reach its own conclusions and assessment as to whether expert evidence is to be relied on. In this case the Tribunal disagreed with the factual basis upon which such evidence was based. This was something within the Tribunal's competence to do.[25]

19. Clearly these recent determinations suggest a more circumscribed role for unreasonableness than that envisaged by the earlier cases discussed above. The ground has been narrowed so that it is to be used only where there is unreasonableness in the very strict sense of the word. That is, courts can intervene on the basis of unreasonableness only where the evidence can lead to only one possible (as opposed to reasonable or preferable) conclusion and this is not the one arrived at by the decision maker. Certainly, this approach is preferable to the widening of the ground favoured by earlier decisions. Unreasonableness cannot be used as a guise to allow the courts to intervene in decisions they simply disagree with. As discussed below, there are significant and well founded limitations on the courts' power of review and these limitation must be accorded sufficient respect. These recent decision have re-assessed the importance of these limitations and highlighted the need for their respect. This should allow for unreasonableness to operate more effectively as a ground of review. This is because it has become more clearly defined. Consequently, greater consistency in its application should be fostered.

20. These restrictions should not however become so severe as to render this ground of review otiose. Unreasonableness is an important and useful ground of review. It is necessary to capture decisions such as that in Chan which would appear to be manifestly absurd. Certainly as made clear in Eshetu something overwhelming, not just a mere divergence of opinion, is required. The problem remains however of how to determine when something is or is not overwhelming. In Eshetu, for example, the Court concluded that there was some material supporting the Tribunal's conclusion and that the Tribunal reached a decision which on the evidence it was entitled to make. Likewise in Betkhoshabeh and Comg Tam Dong the Federal Court held that it could not hold the decisions in question unreasonable as the decision makers arrived at conclusions that were open to them on the evidence. This may indeed have been the appropriate conclusion in these cases. There must however be some scope for the courts to intervene in cases in which the decision maker arrives at a conclusion open to it on the evidence but which is nevertheless unreasonable.[26]

    Conclusion

21. From these cases and the above discussion it is clear that unreasonableness has been severely curtailed as a ground of review. It is only to be applied in the most strict of circumstances. Certainly this deals with some of the problems associated with too wide review and also addresses some of the issues of the efficiency and utility associated with this ground of review. The question that demands attention, following these decisions, however, is what role, if any, should unreasonableness play as a ground of review.

    The limits of judicial review

22. The answer to the question "what are the limits of judicial review for unreasonableness" is easy to articulate, at least in theory. The fundamental guiding constraint is that the courts must examine only the legality of a decision. They must not entertain an appeal on its merits. Under no circumstance must the court substitute its own decision for that under review. This was emphasised by Lord Greene in the Wednesbury case itself. He asserted that although the courts can intervene if a decision is so unreasonable that no reasonable authority could have come to it once the decision making body is acting within the four corners of its jurisdiction it is not for the courts to intervene simply because it disagrees with the decision in question.[27] Essentially, the distinction stipulates that in reviewing exercises of administrative discretion the courts must confine themselves to ensuring that the decision maker has acted in accordance with the terms of his or her power and in compliance with other overriding principles of legality such as relevancy, propriety of purpose and reasonableness. They must not review the appropriateness of a decision nor must they in any way become involved with the policy considerations underlying it in the sense of simply criticising the substance of the policy. This constitutes the merits of the decision and it is the body in whom Parliament has vested the discretion that is alone to be the final arbiter on the merits of a decision.

23. There is good reason for maintaining the distinction between the legality and the merits of a decision and for limiting the courts' task to reviewing the former. There are considerable practical and constitutional limitations on judicial power that demand respect when the courts are exercising judicial review. Practical limitations derive from the institution of the judiciary itself. In particular, the adversarial process is not designed to deal with administrative disputes which, far from being two party disputes, are polycentric. Yet within the confines of this system they are reduced to two party disputes and so wider administrative and policy issues are ignored. Nor is the judiciary well equipped in terms of expertise to deal with the kinds of policy based issues that arise in administrative disputes. Hence the danger is that if the courts do become too involved in these disputes and overstep the legality - merits distinction they will be making assessments of policy issues about which they have not been sufficiently informed. This in turn threatens to undermine the efficiency and integrity of the administrative process itself. Increased judicial scrutiny of the general policy considerations that underlie particular decisions will engender uncertainty amongst administrators who may focus on avoiding a specific challenge to an exercise of discretion rather than, for example, pursue initiatives or make long range plans.[28]

24. Perhaps more significant is the apparent disregard for the constitutional limits of judicial review that stems from judicial involvement in the merits of a decision. Although it is acknowledged that the courts are constitutionally empowered to test the legality of executive action, the danger is that they themselves will exercise the power vested by Parliament in the original decision maker and substitute their own decision for that under review. Judicial review would thereby become a process substantially indistinguishable from that of the executive. A "surrogate political process" exercised by the courts in defiance of the doctrines of the separation of powers and parliamentary sovereignty would eventuate.[29] This would be a dangerous encroachment upon the democratic process as the powers formally reposed in the elected representatives of the people would come to be exercised by bodies which are neither democratically elected nor politically accountable. Constitutional propriety and ultimately the authority and legitimacy of judicial review itself would be at the risk of being undermined.

25. In practice however this theory is not so easily applied or adhered to. Any ground of review potentially involves the courts in an evaluative process which brings them close to an examination of the merits of a decision. Given that discretionary powers are often conferred in language that is "elliptical, ambiguous and inherently open textured" a determination as to relevancy or propriety of purpose will entail a value judgement as to the limits of the discretion and the purposes for which it should be employed.[30] These potential dangers are unquestionably more pronounced with respect to review for unreasonableness. The very nature of this ground of review invites an examination of the merits of the decision for although the empowering statute will require that the decision maker act reasonably, it will not define what is and is not reasonable. Hence, determining whether or not a decision maker has acted reasonably is not simply a question of whether he or she has done something which is authorised by the statute. Rather, unreasonableness asks the court to measure a decision against an ideal standard of reasonableness. This standard is a vague, ill-defined standard that allows examination of the substance, not just the process, of the decision.

26. This proposition is adequately borne out by an analysis of many of those cases in which the courts have held the decision under review to be unreasonable. An examination of Prasad or Luu, for example, reveals judicial intervention in decisions that can hardly be held to be so unreasonable that no reasonable person could have made them. In Prasad, Wilcox J did not reach his conclusion by examining whether the decision was so unreasonable that no reasonable person could have made it. He simply re-assessed the evidence before the Minister's delegate and formulated his own conclusion on his own interpretation of that evidence. Regardless of the merit or otherwise of Wilcox J's ultimate decision it can be said that the issue was one involving individual interpretation and judgement and in this instance the legislature had vested the relevant responsibility for making that judgement in the Minister not the courts. Arguably, Wilcox J intervened to review the merits of the decision in apparent contravention of the limits of judicial review.

27. These limitations are now being accorded far greater respect by the courts. It is apparent from cases such as Eshetu and Betkhoshabeh that both the High Court and the Federal Court have limited the application of unreasonableness to extreme circumstances only. Such extreme circumstances will arise when the evidence is capable of supporting only one possible conclusion. If the decision maker does not arrive at this one possible conclusion his or her decision will be castigated as unreasonable. Mere disagreement with the opinion of the decision maker will not be sufficient to render a decision unreasonable. Clearly, this envisages an extremely attenuated conception of unreasonableness, one which will not only have a very limited sphere of application but which is dependent on the court determining what is and is not the only possible interpretation of the evidence. It would seem that, at least to some extent, these decisions have replaced the standard of reasonableness with that of correctness. That is, has the decision maker correctly interpreted the evidence. If not the decision is unreasonable. In cases where there is more than one possible interpretation of the evidence, that is no correct interpretation, the decision will not be held to be unreasonable. It is for the designated decision maker to determine which is the appropriate conclusion in these cases and it is not for the courts to intervene on the basis of unreasonableness.[31]

28. From here the questions that must be addressed are whether this ground of review should be maintained at all, given that it has been so narrowly interpreted as to become virtually otiose, and, if it is to be retained, what is the appropriate scope of judicial review for unreasonableness? The answer to the former question must be yes. If the courts are to curtail to the optimum degree the misuse of discretionary power it is essential that unreasonableness be maintained as a ground of review. The need for its retention is clearly demonstrated by cases such as Chan v Minister for Immigration and Ethnic Affairs. In this case unreasonableness was the only basis upon which the court could have castigated the delegate's decision that Mr Chan had no reason to fear for himself and that any fear was not well founded. The facts clearly indicated that Mr Chan did have a well founded fear of persecution. To hold otherwise on the basis of the material before the decision maker was grossly unreasonable.[32]

29. More problematic is defining with certainty the appropriate ambit of this ground of review. The argument in favour of adopting the restrictive approach advocated in Eshetu is that a wider interpretation and application of unreasonableness may eventuate in the breach of the limits of judicial power discussed above. To avoid this usurpation of the administrative process unreasonableness must be available as a ground of review only in extremely limited circumstances. One of the problems with this formulation of unreasonableness, however, is that it is intended to be so restrictive that as a ground of review it would become essentially otiose. The danger is that there would consequently be established "a zone of immunity ... around many oppressive or improper decisions that are in reality vulnerable to judicial review."[33] There must be a ground of review that can capture decisions such as that in Chan that would otherwise escape scrutiny.

30. Furthermore, the formulation of unreasonableness posited in Eshetu and the following decisions does not in itself resolve many of the problems associated with review for unreasonableness. The need for something overwhelming or for the evidence to support only one possible conclusion has the same defects as other formulations of this ground of review. Conceptions of correctness are no more determinate than those of reasonableness. Both require the courts to engage in an evaluative process. Both require an examination of the substance of the decision itself in order to determine whether the evidence has been interpreted either correctly or reasonably. The test is open to precisely the kind of manipulation that it seeks to eliminate. The line between mere divergence of opinion and a correct interpretation of the evidence is no more determinate than that between disagreement with the decision under review and concluding that it is unreasonable. The danger of illegitimate judicial incursion into the merits of the decision remains.

31. Nevertheless, this ground of review must be maintained. It is essential if decisions as manifestly absurd as that in Chan are to be subject to review. It must however be exercised with restraint. Any open intervention in the merits of a decision cannot be countenanced. The limitations on judicial review exist for good reason and so must be preserved. However, it is not necessary that it be as limited as a ground of review as suggested in recent case law. To allow unreasonableness to be applied only in cases where the evidence supports only one conclusion is to confine the ground of review to non-existence. While concerns that unreasonableness involves the courts in too much of a qualitative exercise are legitimate it should not be forgotten that much of the role of the courts in administrative law in assessing relevance, propriety and even standing is an evaluative process. If on occasion this results in an unavoidable judicial incursion into areas not appropriate for judicial consideration this is the cost of charging the courts with the task of protecting individual rights. This ground of review must be exercised sparingly and the courts must exert self restraint. But there must also be some scope for this ground of review to be utilised.

    Proportionality

32. Attempts have been made to move away from the language of unreasonableness in an effort to provide greater clarity and consistency in reviewing administrative discretion. Proportionality, for example, has been suggested as a ground of reviewing administrative action. This principle of proportionality "requires that the means employed by the decision maker must be no more than is reasonably necessary to achieve his legitimate aims."[34] Hence, the decision maker must have regard to the legitimate objectives which he or she seeks to attain, and pursue those objectives by means which are not "oppressively excessive." It is a principle that is objective driven. By definition it requires the courts to engage in a balancing process to determine if the objectives of the discretion have been pursued by means that are proportionate to that end.

33. It was Lord Diplock who first alluded to the introduction of proportionality into English law in the CCSU case where he referred to the grounds of review as irrationality, illegality and procedural impropriety and then went on to foresee the adoption into English law of the principle of proportionality.[35] This statement sparked debate as to whether proportionality should operate as part of the unreasonableness ground of review, whether it is in itself a completely separate ground of review or indeed whether it should have any role to play at all in reviewing discretion at all. Arguments in favour of it operating as a totally separate ground of review replacing that of unreasonableness, assert that whereas unreasonableness is a vague standard, incapable of definite application,

    proportionality advocates a relatively specific principle - one that is at any rate far more specific than 'unreasonableness' or 'irrationality' - [hence] it focuses more clearly than those vaguer standards on the precise conduct it seeks to prevent. By concentrating on the specific it is more effective in excluding general considerations based on policy rather than principle.[36]

    In short, proportionality is presented as a principle capable of objective and certain application, one which does not encourage judicial usurpation of the administrative function.[37]

34. In Australia proportionality has been adopted as a principle in some areas of law, notably criminal law and constitutional law.[38] Yet, despite the growing body of literature in its favour, it has not been accepted as a separate ground of review in administrative law. Although it has been recognised as relevant in determining the validity of regulations it has no place in assessing the legality of an exercise of administrative discretion. It has received some judicial consideration, notably by Deane J in Australian Broadcasting Tribunal v Bond.[39] However, proportionality seems to be envisaged as simply an indicia for determining if there has been unreasonableness in the Wednesbury sense rather than as a separate ground of review.[40] Proportionality should not be introduced into Australian administrative law as anything more than a guide to determining whether a decision maker has acted unreasonably.

35. It is difficult to accept that the use of this principle rather than that of unreasonableness would render judicial review any less indeterminate or any less dependent on judicial value judgements. "[I]t does not resolve in and of itself the actual standard of review ... there is still the further crucial question of how far the authority can deviate from what the court believes to be the correct proportionate balance between the respective interests."[41] A value judgement is still required to be made by the court as to what it perceives to be a proportionate balance which is reasonable and rational in the circumstances before it. At most any test of proportion should be considered merely as an indicator of unreasonableness.

    Conclusion

36. The answer to the question is there still a role for unreasonableness is definitely yes. Cases such as Chan indicate that there is a need for this ground of review. Clearly, unreasonableness is an inherently subjective and indeterminate ground of review and the question of the extent to which the courts should be able to exercise review on this ground is very complex. Too much intervention ignores the basis and legitimacy of the limitations on judicial power. Excessive restraint fails to accord sufficient respect to the courts' role of protecting the individual and defining the limits of governmental power. Despite this difficulty of definition unreasonableness must be maintained as a ground of review. It does provide a useful and legitimate basis for judicial review.

37. This ground of review must, however, be exercised with caution. Its scope must not be so restricted that it becomes essentially otiose. Yet, it must also be tempered by an appreciation of the constitutional and practical limitations on judicial review. The courts must be aware that as the policy content of an impugned decision increases it will be become increasingly inappropriate for judicial intervention. This need, on the one hand, for the courts to be able to review unreasonable exercises of discretion and, on the other, the apparently incompatible necessity for the limits of judicial review to be respected prescribes that the ambit of review on the ground of unreasonableness cannot be precisely and conclusively defined. Rather, what is required is the striking of a balance between too much and too little intervention. Where the balance will lie will depend largely and inevitably upon the facts of a particular case.