Classification

First, the grounds of review tend to overlap, potentially allowing errors to be classified under multiple grounds. Indeed, many judges, including the architects of the tripartite grounds, have warned about approaching the threefold division too clinically.Footnote ⁴⁶ As will be shown, though, the distinctions on which these grounds are based tend to break down.Footnote ⁴⁷ This style of modulation is best illustrated by reference to some of the key dichotomies in judicial review: law versus fact and law versus discretion.

Turning first to the dichotomy between law and fact. Under the grounds of review schema, an alleged error may be classified as one of law or fact, enabling different depth of review.Footnote ⁴⁸ If the error is one of law, then the court can express its own view on whether the decision is correct; if it is a factual error, the deferential Wednesbury standard of unreasonableness would ordinarily apply. The commentary in de Smith reflects this unstable division between law and fact. As in earlier editions, Woolf and Jowell spend some time addressing the boundary between law and fact, which, they repeat, ‘is not always easy to perceive’.Footnote ⁴⁹ While noting that the courts generally leave the assessment of fact to the primary decision-maker, Woolf and Jowell note that factual error ‘can just as easily be absorbed into a traditional legal ground of review’.Footnote ⁵⁰ Other commentators are similarly sceptical, doubting whether there is a sound analytical approach to the distinction and suggesting that pragmatic considerations must be in play. For example, Williams explained that the use of the terms ‘law’ and ‘fact’ in this context are ‘simply flexible concepts that can be used to contain or conceal more pragmatic reasoning’.Footnote ⁵¹ Even more manipulable is the classification of a question as a mixed question of law and fact. This conjugated label has been described as ‘one of the baffling gadgets in the judicial toolbox’.Footnote ⁵² It allows vigilant or restrained review, depending on judicial preference.

Two particular examples illustrate this type of modulation of intensity, in the context of the law–fact dichotomy. The first, illustrated by the Puhlhofer case, demonstrates the difficulty in determining whether a particular administrative finding is based on any (mis-)understanding of law or factual judgement.Footnote ⁵³ In Puhlhofer, the entitlement to government housing assistance depended on being homeless, that is, whether the couple had ‘no accommodation’. The local authority declined assistance because the couple were living temporarily in a bed and breakfast – without cooking or laundry facilities – and were therefore not homeless. But was the alleged error on the part of the local authority an error of law (the local authority misinterpreting the meaning of accommodation) or one of fact (the bed and breakfast was wrongly classified as accommodation)? Ultimately, the House of Lords ruled the critical determination was a factual one; thus intervention was only justified under the irrationality ground if the determination was manifestly unreasonable. But it was also plausible for the determination to be classified as a question of law, which would have entailed more vigilant review.Footnote ⁵⁴ The uncertainty in classification leaves it open to the courts to deploy differing degrees of scrutiny, ostensibly on normative, not descriptive, reasons.

Secondly, factual findings not normally subject to close scrutiny under the irrationality ground may be treated as giving rise to an instance of illegality.Footnote ⁵⁵ Sometimes justified under the ‘jurisdictional or precedent fact’ principle, if a factual finding is a statutory precondition to the exercise of power the courts sometimes (but not always) subject the factual circumstances or criteria to closer review, assessing whether, in their view, it is satisfied.Footnote ⁵⁶ The reasoning goes as follows. Where the legislative framework dictates that the presence of a particular fact is a precondition to the exercise of a statutory power, certain administrative action is only permitted if a particular fact is established. Acting in the absence of the fact being established would be acting without any legal authority or jurisdiction, thereby justifying the greater scrutiny seen under the unlawfulness ground of review. But, as will be shown, the syntactical nature of a precondition is not always determinative and the courts are still sometimes reluctant to apply an exacting eye to certain factual preconditions.

The exemplar case is R (Khawaja) v. Secretary of State for the Home Department, where immigrants successfully challenged moves to deport them.Footnote ⁵⁷ The power to deport was conditional on a factual determination that the immigrants were ‘illegal entrants’. Because the contested determination – that the immigrants were ‘illegal entrants’ – was a ‘precedent or jurisdictional fact’, the House of Lords was prepared to assess itself whether the factual precondition existed or not. This vigilant approach has not been replicated, however, where the factual precondition is slightly indeterminate or involves a degree of evaluation and judgement. While these factual preconditions present the same jurisdictional problems as more objective or determinate facts, the courts have shown greater reticence and reverted to supervising by the more deferential Wednesbury unreasonableness approach.Footnote ⁵⁸ For example, in South Yorkshire Transport, the House of Lords declined to treat the factual determination of whether or not a transport company was operating in ‘a substantial part of the United Kingdom’, when that determination operated as a threshold for investigation by a fair trading commission.Footnote ⁵⁹ Lord Mustill said a ‘clear-cut approach’ cannot be applied in every case of jurisdictional preconditions, especially where the relevant criterion is imprecise.Footnote ⁶⁰ In such cases, the court can intervene ‘if the decision is so aberrant that it cannot be classed as rational’.Footnote ⁶¹

The flexible nature of this style of categorisation is therefore obvious. In some cases, statutory preconditions are treated as touching matters of law; in others, they are treated in the same way as ordinary fact-finding and only subjected to deferential review under the irrationality ground. The elastic nature of the classification problem is seen vividly in the recent case of R (A) v. London Borough of Croydon,Footnote ⁶² where the Supreme Court decided different elements within the same factual precondition should be subject to different degrees of scrutiny. The critical provision required a local authority to provide accommodation for any ‘child in need within their area’. The Supreme Court ruled that if there was a dispute about a local authority's factual finding about whether a person was actually a ‘child’ (based on doubts about official documents), it was for the courts on judicial review to ‘determine where the truth lies on the evidence available’, with ‘no margin of discretion’ applying.Footnote ⁶³ However, this strict standard of scrutiny did not apply to the corresponding element ‘in need’, which would continue to be challengeable only on Wednesbury grounds.Footnote ⁶⁴ The approach ultimately turned on the objective character of the term ‘child’, in contrast to the evaluative nature of the balance of the precondition.

This distinction between law and discretion also presents a subtle reclassification choice to judges. The intensity applied by the courts to legal questions is strict, in contrast to the deferential approach adopted to reviewing the exercise of discretion. Therefore, treating the influence of an external norm such as human rights as speaking to questions of legality – rather than as a factor which must be taken into account in the exercise of discretion – affects the depth of review applied. Both courses remain open to a supervising court under common law review, and the choice between both allows the significant modulation of the depth of review.

The best example of this is the principle of legality, which adopts a strict approach to compliance with human rights norms under the illegality ground.Footnote ⁶⁵ The courts read down legal powers to avoid conflict with (so-called) ‘fundamental’ or ‘constitutional’ rights except where legislation necessarily authorises the rights being limited.Footnote ⁶⁶ The judicial method applied is a strict one, in contrast to the soft-edged evaluation that would otherwise take place under the unreasonableness ground (under either the Wednesbury test or, arguably, any stricter formulation).Footnote ⁶⁷ The application of the principle, under the guise of the legality ground rather than irrationality, produces ‘high-intensity review’.Footnote ⁶⁸ Indeed, the principle of legality is occasionally identified as a sibling to the principle of anxious scrutiny under the unreasonableness ground, because both have been deployed by judges to enable more intensive protection of human rights.Footnote ⁶⁹

Notable cases where this methodology was adopted include R (Pierson) v. Secretary of State for the Home DepartmentFootnote ⁷⁰ and R (Simms) v. Secretary of State for the Home Department.Footnote ⁷¹ In Pierson, a prisoner serving a life sentence had been told he would serve a tariff period of at least fifteen years in prison before being considered for parole but the Home Secretary subsequently raised the tariff period to twenty years. In ruling the Home Secretary's actions unlawful, two Lords referred to a general interpretative principle that general legislative wording should be read subject to human rights. Lord Browne-Wilkinson articulated the interpretative approach as a general principle:Footnote ⁷²

Describing it as a ‘spirit of legality’, Lord Steyn drew an analogy with the long-standing presumption that powers granted to public bodies and officials must be exercised consistently with the common law principle of procedural fairness.Footnote ⁷³ The principle also gained support in Simms, when the House of Lords overturned a blanket ban on prisoners giving interviews to journalists. Lord Hoffmann said: ‘In the absence of express language or necessary implication to the contrary, the courts … presume that even the most general words were intended to be subject to the basic rights of the individual.’Footnote ⁷⁴ Lord Steyn echoed these sentiments, expressing a ‘presumption of general application operating as a constitutional principle’ that if a ‘fundamental or basic right’ was at stake the provisions should where possible be interpreted consistently with those rights.Footnote ⁷⁵

The translation of the interplay between rights norms and other factors from discretion to law, gives rise to potential variability, enabling judges to deploy greater scrutiny if they consider the human rights dimensions or higher level norms justify it. The conditions under which this maximum intensity can be deployed are not definitive. First, conflict with a ‘fundamental human right’ must be identified; however, as these human rights norms are by definition unenumerated, the set of recognised rights is pliable. Secondly, it must be established that it was not the express or implied legislative intent of Parliament to mandate restrictions on those rights when conferring the discretionary power. The threshold for reliance on the principle of legality is therefore versatile, giving the doctrine the character of variable intensity.

Evolution

When the grounds were encapsulated in their tripartite formulation, the door was also left open for other grounds of review to develop.Footnote ⁷⁶ These emergent grounds present different depths of scrutiny – usually more vigilant intensity – to the traditional tripartite grounds. The recognition of potential development of other grounds of review brings other dimensions of variability to the judicial method. This variability is both immediate (where a ground has been recognised within the overall schema) and longer-term (where a ground is able to be explored as prospective ground). Moreover, the emergent grounds tend not to have the universal application of the traditional tripartite grounds; the narrower gateways to reliance on them adds another layer of classification, which further augments the variability associated with them.

Lord Diplock acknowledged further grounds may develop on ‘a case by case basis’, noting particularly the possibility that proportionality might be recognised as a ground of review.Footnote ⁷⁷ The possibility that the suite of grounds might be enlarged was also noted in a number of cases that endorsed the tripartite formulation. While acknowledging the value of the tripartite statement, Lord Scarman in Nottinghamshire was quick to note that it was ‘certainly not exhaustive analysis of the grounds upon which courts will embark on the judicial review’.Footnote ⁷⁸ Similarly, Lord Roskill in Wheeler repeated the three grounds were not exhaustive and ‘further grounds may hereafter require to’;Footnote ⁷⁹ subsequently in Brind he stressed that ‘any such development would be likely to be on a case by case basis’.Footnote ⁸⁰

A number of additional or alternative grounds have been promoted; some grounds have achieved some recognition, although none seem to have yet achieved the same exalted status of the traditional tripartite grounds. The slightly opaque status of grounds of review make it difficult to definitively assess whether and when a basis for judicial intervention is sufficiently recognised so as to be regarded as a ground of review. First, as explained earlier, the focus on individual cases means judges are often agnostic to the overall schema of judicial review and may not herald such developments, beyond identifying an error as a justifiable basis for intervention.

Secondly, the overlapping and multi-dimensional nature of grounds of review mean their genesis is often interwoven with an existing ground of review. Potential grounds of review are sometimes adopted merely as touchstones which may indicate another ground of review has been established. For example, unjustified inconsistent treatment may be the reason why a court holds administrative action to be invalid under the unreasonableness ground.Footnote ⁸¹ Similarly, the failure to honour a legitimate expectation has sometimes been treated as giving rise to a breach of the unfairness or procedural impropriety grounds.Footnote ⁸² Indeed, Lord Diplock's seminal speech systemising the grounds of review noted that previously intervention for irrationality was justified by relying on ‘an inferred though unidentifiable’ error of law.Footnote ⁸³

Thirdly, the recognition of a new ground of review need not await the imprimatur of final appellate courts. The common law may develop within lower courts and become sufficiently imbedded – whether or not final appellate courts have had the opportunity or inclination to comment on the development.Footnote ⁸⁴ As a consequence, some emerging grounds are suspended in a sort of twilight zone. The looseness of stare decisis principle in judicial review and the large doses of discretion available to a reviewing judge means it is still possible for some emergent grounds to be adopted in occasional cases, even though the ground does not then, or subsequently, command widespread support.

A number of substantive grounds of review have been promoted and, in some cases, appear to have assumed new status as potential grounds of review, at least in some circumstances. As Rawlings remarks: ‘Future historians will record that, in the shadow of the ECHR, the pace of development in the grounds for review quickened in the late 1980s and 1990s.’Footnote ⁸⁵ The most prominent emergent grounds include substantive legitimate expectation, mistake of fact, and proportionality. A further set of other grounds have also been promoted, such as inconsistent treatment, substantive fairness, and the innominate ground, but have gained less traction.

Substantive protection of legitimate expectations was recognised in English law in the seminal Coughlan case in 2001 and is often regarded as a separate ground of review.Footnote ⁸⁶ As noted earlier, the sixth edition of de Smith addresses legitimate expectations in a separate chapter; the authors implicitly treat substantive legitimate expectation as having the status of a self-standing ground of review, but are coy in making any formal pronouncement to that effect.Footnote ⁸⁷

Substantive legitimate expectation mandates judicial intervention where an assurance or other action of the administration induces ‘a legitimate expectation of a benefit which is substantive’ and the frustration of such expectation ‘is so unfair that to take a new and different course will amount to an abuse of power’.Footnote ⁸⁸ While expressed as a general ground of review, the ground is only engaged in ‘limited conditions’; ‘[n]o magic formula’, de Smith's sixth edition notes, but a range of contextual factors which seek to balance the ‘relative virtues and defects of certainty and flexibility’.Footnote ⁸⁹

Presented with an assurance or other conduct inducting an expectation, the courts must assess the ground(s) most applicable to the circumstances. The Court of Appeal in Coughlan identified three alternatives.Footnote ⁹⁰ First, the court may determine that it is appropriate only to assess the relevance and weight under the illegality and irrationality grounds, to ensure the assurance has been taken into account and the weight given to it relative to other grounds is not unreasonable. Secondly, under the procedural impropriety ground, a requirement to consult before reneging on the assurance may be imposed. Finally, the court may afford the expectation some substantive protection and assess whether departing from it is unfair as to amount to an abuse of power. The Court conceded that the ‘difficult task will be to decide into which category the decision should be allotted’.Footnote ⁹¹ Little guidance was given about criteria influencing this assessment, apart from noting that, in the particular case, the importance of the promise and fact it was made to only a few people were significant factors, along with the limited (financial only) consequences for the administration if it was required to honour the assurance.Footnote ⁹² This has been taken to require an assessment in other cases of the nature of the express or implied representation or promise, the legitimacy, in the circumstances, of relying on the representation or promise, and the absence of any public interest supporting the administrative change of position.Footnote ⁹³

If the expectation makes it through the gateway and is assessed to be worthy of substantive protection, then the administration's failure to honour that assurance or expectation is subjected to more intensive review than applied that under the traditional grounds of review, especially in relation to the assessment of weight under the irrationality ground of review.Footnote ⁹⁴ The substantive protection path allows the courts ‘to determine whether there is a sufficient overriding interest to justify a departure from what has been previously promised’;Footnote ⁹⁵ in contrast, the default ‘conventional’ or Wednesbury ground focuses solely on ‘rationality and whether the public body has given proper weight to the implications of not fulfilling the promise’.Footnote ⁹⁶ The mandate to assess whether the departure is justified has been equated to high-intensity or correctness review.Footnote ⁹⁷

Since its recognition, the ground continues to be successfully relied on intermittently.Footnote ⁹⁸ In other instances, review based on the ground failed because the qualifying conditions were not made out or departing from the assurance was justified in the public interest.Footnote ⁹⁹

Proportionality is often marked out as a candidate for another ground of review.Footnote ¹⁰⁰ Despite some strong proponents, it has failed to yet crystallise as a ground of universal application. However, it has been endorsed in particular spheres within administrative law; most obviously in the domain of human rights adjudication, but also in relation to the review of excessive penalties and sanctions and directly effective European Community law. The nature of the deployment of proportionality within the schema of judicial review remains unclear, though. On one account, proportionality is a ground of review of limited application; an alternative account treats proportionality merely as an interpretative test or calculus to determine compliance with human rights instruments under the illegality ground of review. As with other emergent grounds, proportionality has the potential to mandate greater depth of review than the traditional irrationality ground allows, although this is not always the case necessarily, given the tractable nature of proportionality review.

The possibility that proportionality might be adopted as an additional ground was foreshadowed in CCSU. However, such a development was subsequently forestalled by the House of Lords in Brind, with judges then expressing concern that adoption of proportionality would amount to an inappropriate move towards merits review.Footnote ¹⁰¹ The question has since been left open. The Court of Appeal in R (Association of British Civilian Internees) v. Secretary of State for Defence suggested proportionality should be substituted for Wednesbury unreasonableness, but accepted that any such change could only be made by a court at the highest level.Footnote ¹⁰² While the possibility has been acknowledged, the House of Lords and Supreme Court have left the question open.Footnote ¹⁰³ Debate continues about whether proportionality ought to supplant irrationality as the leading substantive ground of review. Craig continues to be one of the leading advocates for the embrace of proportionality as a universal ground.Footnote ¹⁰⁴ Others have joined in his promotion of proportionality but on a more limited basis. For example, Taggart argued in favour of an enhanced role for proportionality for all cases addressing human rights (including rights beyond those recorded in statutory bills of rights), but also argued for the retention of reasonableness review for the remaining cases addressing ‘public wrongs’.Footnote ¹⁰⁵

While not assuming status as a universal ground of review, the language of disproportionality has, however, still been relied on in (non-human rights) cases to quash disproportionate or excessive sanctions or penalties.Footnote ¹⁰⁶ Lord Denning's remarks in Hook stating the courts can intervene if ‘punishment is altogether excessive and out of proportion to the occasion’ is often highlighted.Footnote ¹⁰⁷ Following the systemisation of the grounds in CCSU, other judges have adopted similar language suggesting proportionality might be regarded as a ground of review in a particular set of circumstances.Footnote ¹⁰⁸ However, the magnitude of disproportionality required for intervention is generally analogous to the threshold for intervention under the irrationality ground. The degree of coincidence suggests disproportionality may operate merely as an indicator of unreasonableness under the rubric of the established irrationality ground, rather than an independent ground in its own right.Footnote ¹⁰⁹ In any event, the expression of a different basis of intervention may still allow judges to avoid the demands of the Wednesbury irrationality test and rely on disproportionality to modulate the supervisory intensity (albeit any difference may be marginal).

Further, a structured form of proportionality has been established as the dominant method of review under bills of right, such as the Human Rights Act 1998. Indeed, the adoption of the proportionality doctrine has been described as ‘one of the most profound changes to judicial reasoning brought about by the [Human Rights Act]’.Footnote ¹¹⁰ Proportionality is not referred to in the Human Rights Act or European Convention; however, the English courts have, consistently with the approach in other jurisdictions, interpreted limitations clauses as involving a proportionality assessment.Footnote ¹¹¹ The European Court of Human Rights ruled that the phrase ‘necessary in a democratic society’ in the European Convention requires the court to assess the proportionality of the rights-infringing measure.Footnote ¹¹² A similarly styled test has been adopted under the Human Rights Act 1998, assessing whether the government objective is ‘sufficiently important’ to justify limiting rights, whether measures adopted are ‘rationally connected’ to that objective, whether the impairment of rights are ‘no more than is necessary’ to achieve that objective, and whether a ‘fair balance’ was struck.Footnote ¹¹³

For present purposes, our interest in the deployment of this particular form of proportionality lies in its relationship with the traditional grounds of review. As outlined earlier, reliance on proportionality in the human rights domain can be explained in different ways. The first account is evolutionary. That is, the courts have modified the substantive grounds of review when human rights instruments such as the Human Rights Act 1998 are directly impugned. In other words, rather than subjecting decisions to scrutiny under the irrationality ground of review, the courts have endorsed, to a limited extent, proportionality as a ground to review the substance of a decision. Under this account, administrative law is effectively bifurcated and cases where human rights instruments are engaged are subjected to different principles of review. The second account addresses proportionality in the colours of (il)legality. The judicial enquiry is on the lawfulness of the actions of the administration, given the legislative decree in section 6(1) of the Human Rights Act 1998 that it is ‘unlawful for a public authority to act in a way which is incompatible with a Convention right’. Thus, proportionality is the interpretative method for assessing whether a statutory obligation has been violated, not a freshly endorsed substantive ground of review. In other words, whether or not, as a matter of law, interference with rights is incompatible is settled through the proportionality calculus (at least for those cases where limitation clauses apply).

This point has received limited attention. Much of the discussion about the place of proportionality identifies a contrast with the Wednesbury unreasonableness ground.Footnote ¹¹⁴ Indeed, Lord Steyn's endorsement of proportionality in Daly compared proportionality to the ‘traditional grounds of review’.Footnote ¹¹⁵ Framing proportionality in this way – ‘through the lens of administrative law’, as Hickman put it – points to proportionality operating as a new ground of review.Footnote ¹¹⁶ Others characterise proportionality as an aspect of the illegality ground of review. For example, Leigh suggests ‘a new form of over-arching illegality – in the sense that Lord Diplock used that term in [CCSU]’ – is created.Footnote ¹¹⁷ In particular, the obligation to act compatibly with Convention rights enables legality review, noting that ‘there is nothing which suggests that its sole effect is to modify the Wednesbury ground of review, as seems universally to be assumed’.Footnote ¹¹⁸ Craig similarly acknowledges that proportionality under the Human Rights Act operates as a legal ‘test’ but, at the same time, also tends to speak of proportionality as a new ground of review.Footnote ¹¹⁹ ‘Section 6(1) creates a new statutory head of illegality for breach of a Convention right’, Craig says. ‘It is a free-standing ground of challenge.’Footnote ¹²⁰ On the one hand, this is consistent with Leigh's illegality formulation. On the other hand, it also tends to mark a breach of Convention rights (along with the proportionality test implicitly involved) as an independent ground of review; this is reinforced by Craig's treatment elsewhere of proportionality, where he analyses its status in tandem with the existing Wednesbury ground.Footnote ¹²¹

In any event, this structured form of proportionality potentially mandates depth of review that is more intense compared to substantive review under the Wednesbury standard (regardless of whether deployed under the guise of illegality or as an alternative substantive ground). It has generally been understood and applied by judges as involving close scrutiny of the governmental action.Footnote ¹²² This point is often overstated because proportionality does not, in itself, dictate the depth of scrutiny and is better understood as a relational concept which relies on other factors to settle the depth of scrutiny that is applied. Indeed, it is increasingly acknowledged that proportionality inquiry necessarily operates in tandem with notions of deference or intensity.Footnote ¹²³ The modulation implicit in proportionality therefore enables greater variability, over and above the evolutionary dimension. Further, the components of the proportionality review – particularly the touchstones of ‘sufficiently’, ‘rationally’, ‘necessary’, and ‘fair balance’ – provide ample room for judicial discretion.Footnote ¹²⁴

Mistake of fact has also received some recognition as a ground of general application, beyond the limited circumstances in which fact finding could traditionally be challenged.Footnote ¹²⁵ In E v. Secretary of State for the Home Department, the Court of Appeal recognised that ‘a mistake of fact giving rise to unfairness is a separate head of challenge’.Footnote ¹²⁶ The factual mistake must be ‘established’ (that is, shown ‘by objective and uncontentious evidence’), must have played a ‘material’, albeit not decisive, part in the reasoning, and the claimant must not have been responsible for the error.Footnote ¹²⁷ The linkage with fairness casts some doubt on its status as a free-standing ground of review; it is unclear whether the reference stands merely as a conclusion or whether the mistake is merely a touchstone within the broader rubric of unfairness.Footnote ¹²⁸

Finally, a range of other possible substantive grounds have been proposed as counterpoints to the Wednesbury unreasonableness ground, but their impact on judicial doctrines has been quite mixed. These include potential grounds like inequality or inconsistency,Footnote ¹²⁹ substantive fairness,Footnote ¹³⁰ and the so-called ‘innominate’ ground.Footnote ¹³¹ These grounds either have assumed a subsidiary role informing the assessment under other grounds or appear to have been overtaken by other developments (such as substantive protection of legitimate expectations and variegated forms of unreasonableness). While their impact on judicial review doctrines nowadays is therefore limited, dalliances with these alternative grounds speak to the judicial penchant for promoting alternative doctrinal grounds to circumvent Wednesbury's deferential degree of scrutiny.

Reformulation

In addition to the development of novel or emerging grounds, there has been some attempt to recast the grounds of review in a way which would affect the depth of review. The most obvious example is the attempt to simplify the irrationality or unreasonableness ground. That is, rather than seeking to variegate unreasonableness into differing degrees, a number of judges and scholars promoted the idea that the deferential Wednesbury unreasonableness ground ought to be given unified and simplified expression.Footnote ¹³² These attempts can be viewed in different ways. On the one hand, they potentially represent a ground of unreasonableness which has a greater vigilance than is apparent under its original Wednesbury formulation. To this extent, these efforts can be explained as the revision of an existing ground of review, hence their mention at this point. On the other hand, they potentially signal a new – more contextual – style of review. Their emphasis on context and circumstance, teamed with the mandate of significant judicial discretion, tends towards a departure from the categorical approach which underlies the grounds of review schema. These developments are therefore discussed under contextual review later.Footnote ¹³³

Circumscription

Access to the traditional grounds of review may be circumscribed through the application of the principle of non-justiciability. In its strongest formulation, review may not be permitted at all – that is, none of the grounds of review are treated as being applicable.Footnote ¹³⁴ In its softer formulation, the suite of grounds of review may be circumscribed or modified to take account of the context of particular cases. Harris has described these two different techniques as exhibiting ‘primary’ and ‘secondary’ non-justiciability respectively.Footnote ¹³⁵ In doing so, greater weight is placed on judicial restraint and more deferential supervision results.

The principle of primary non-justiciability is well recognised. The courts may decline to review a matter, as a preliminary or jurisdictional matter, on the basis that it is non-justiciable. It is perhaps most famously seen in the House of Lords in CCSU where their Lordships ruled, on the one hand, that the royal prerogative was not automatically non-justiciable but, on the other hand, that the national security issues raised by the application for review were not suitable for judicial determination.Footnote ¹³⁶ The case is also notable for Lord Roskill's categorical list of certain types of decision that would be non-justiciable.Footnote ¹³⁷ The categories of cases that are non-justiciable under English law have been whittled down over time.Footnote ¹³⁸

The principle of secondary non-justiciability – ‘modified review’, as Fordham puts itFootnote ¹³⁹ – draws a stronger connection with the traditional grounds of review. The traditional grounds of review are modified, either substituting more deferential grounds of review or by disapplying particular grounds. A number of examples illustrate the circumscription or adaptation of the usual suite of grounds of review.

First, the basis for reviewing prosecutorial discretion is narrowly limited in England. Once entirely non-justiciable,Footnote ¹⁴⁰ nowadays decisions to prosecute are reviewable only for dishonesty, bad faith and other exceptional circumstances – not the traditional CCSU grounds. Lord Steyn said in Kebelene, ‘absent dishonesty or mala fides or an exceptional circumstance, the decision of the Director to consent to the prosecution … is not amenable to judicial review’.Footnote ¹⁴¹ Secondly, English courts have adopted a similar circumscribed approach when reviewing commercial decisions.Footnote ¹⁴²

Thirdly, the UK Supreme Court in R (Cart) v. Upper Tribunal recently adopted a form of secondary non-justiciability in relation to the judicial review of the Upper Tribunal's refusal to grant permission to appeal decisions of lower tribunals.Footnote ¹⁴³ Presented with a specially developed tribunals system, with its own provision for appeal and review by appellate bodies comprised of superior court judges, the Supreme Court was reluctant to allow review on ‘the full panoply of grounds’, largely for reasons of comity and pragmatism.Footnote ¹⁴⁴ Instead, the Court ruled that judicial review would only be allowed where the criteria for making a second-tier appeal from the Upper Tribunal to the Court of Appeal is made out (namely, if the matter raises an important point of principle or practice or there is some other compelling reason for the court to hear it).Footnote ¹⁴⁵ This approach was seen to represent a half-way house between the usual grounds that would be available and an even narrower set of grounds proposed by the Court of Appeal below.Footnote ¹⁴⁶

Finally, the English courts have circumscribed the grounds available to review the decisions made within visitatorial jurisdiction. In R (Page) v. Hull University Visitor, for example, a majority of the House of Lords ruled that the long-standing exclusive jurisdiction of visitors on internal affairs of charitable or academic foundations meant that their decisions could not be impugned on the basis of the standard error of law ground.Footnote ¹⁴⁷ Only errors falling outside their jurisdiction (in the narrow sense) were reviewable. However, review for breaches of other grounds such as procedural impropriety remained.

Before leaving the circumscription of the grounds of review, it is important to note the set of cases manifesting a ‘super-Wednesbury’ – that is, more deferential – form of unreasonableness could be conceived as circumscription of the grounds of review akin to the other instances discussed.Footnote ¹⁴⁸ However, I prefer to analyse these developments under the rubric of intensity of review because the theme of the different approach has been more strongly connected to the variegation of the unreasonableness ground.Footnote ¹⁴⁹

Classification

The judicial discretion as to the process of classification is acknowledged in New Zealand. As discussed above, there is a strong recognition of the overlapping character of the grounds and the consequential effect of classifying justiciable matters in each category.

An increasingly common technique in New Zealand is the reliance on the principle of legality which moves matters from the realm of discretion, where deferential grounds apply, into the realm of legality, where strict scrutiny is applied.Footnote ¹⁸⁰ One particular instance of this – the development of the presumption of consistency in relation to the influence of international instruments on domestic administrative law – acutely demonstrates the particular significance of the classification technique in the distinction between law and discretion.Footnote ¹⁸¹ The depth of judicial review differs, depending on whether the challenge is mounted under the illegality ground (under the presumption of consistency, a doctrine which mimics the principle of legality) or the irrationality ground. The presumption of consistency requires any administrative power to be read consistently with international law obligations, except where the statutory matrix is otherwise inconsistent.Footnote ¹⁸² Framed in this way as a matter of law, this approach allows the courts to assess whether or not international law obligations have been correctly applied and effectively circumscribes the discretion available to any public body or official. In contrast, the relevancy approach treats the impact of international instruments as a matter of substance or discretion. As long as the administration has turned its mind to the relevant legal instruments, the weight given to those international law norms can only be impugned under the irrationality ground.Footnote ¹⁸³

Another example of the process of classification determining the depth of review is the approach to statutory preconditions. Following the English approach to jurisdictional facts, the New Zealand courts scrutinise the presence of some – but not all – factual preconditions closely. For example, the courts have mandated close supervision of ‘gate-keeper’ fact-finding that a development proposal has no adverse environmental effects such that public participation can be dispensed with.Footnote ¹⁸⁴ In contrast, a statutory precondition requiring a factual finding that it was desirable to protect shareholders and the public interest before placing a company into statutory administration was not treated as a jurisdictional fact and was only subjected to reasonableness review.Footnote ¹⁸⁵

Reformulation

Efforts to recast the grounds of review in order to modify their depth of scrutiny have been notable in New Zealand, particularly in relation to irrationality and reasonableness review.Footnote ¹⁸⁶

Evolution

Like England, some emergent grounds have crystallised, although their application remains narrow and restricted to particular circumstances.

The recognition of substantive legitimate expectation as a ground of review in New Zealand is somewhat unsettled.Footnote ¹⁸⁷ While the ground has not received the same degree of approval as in England, there is some support (largely within the lower courts) for this ground. The Court of Appeal foreshadowed that intervention may be justified, under the more general rubric of abuse of power or unfairness, where the administration reneges on an assurance or promise.Footnote ¹⁸⁸ A number of decisions in the High Court have also, in principle and in accordance with Coughlan, indicated that an expectation may be afforded substantive protection in some situations.Footnote ¹⁸⁹ Other courts have been more equivocal and, occasionally, hostile;Footnote ¹⁹⁰ further, the courts have sometimes ruled that particular statutory schemes (notably, tax and revenue collection schemes) are incompatible with the substantive protection of expectations.Footnote ¹⁹¹ In any event, most cases fail on the facts at the first stage, failing to demonstrate a clear and unambiguous assurance deserving of protection.

The limited embrace of proportionality is mirrored in New Zealand. There has been reluctance to embrace proportionality as a universal ground of review, but the courts have been prepared to intervene to address (excessive) disproportionality in penalties and sanctions. For example, the Court of Appeal in Institute of Chartered Accountants of New Zealand v. Bevan recognised the courts may intervene to quash penalties which were excessive and disproportionate.Footnote ¹⁹² However, the Court was not prepared to enter ‘the broader question, raised for instance by Lord Diplock as long ago as 1984, whether proportionality is a distinct head of review’, Keith said. ‘Rather, we limit ourselves to the penalty cases such as Hook and take comfort from commentary on proportionality which, while recording the controversy about its separate existence, singles out the penalty area as established.’Footnote ¹⁹³

In the human rights domain, the New Zealand courts have also endorsed proportionality, like their English counterparts, as the test to determine whether government action abridging rights is justified and therefore lawful. Proportionality has been used to assess whether government action amounts to ‘reasonable limits … demonstrably justified in a free and democratic society’ under the general limitation provision in the New Zealand Bill of Rights Act 1990.Footnote ¹⁹⁴ Just as the text of the NZ Bill of Rights Act was drawn from the Canadian experience, so too was the associated proportionality calculus.Footnote ¹⁹⁵

The possibility of a free-standing mistake of fact ground of review was also floated in New Zealand by Lord Cooke in the 1980s in Dagayanasi v. Minister of Immigration, but he failed to secure the support of his then fellow judges for such a development.Footnote ¹⁹⁶ He was prepared to rule that a ministerial decision to deport an overstayer was ‘invalid on the ground of mistake [of fact] as well as on the ground of procedural unfairness’ because it was based on an inaccurate conclusion in a medical report about the overstayer's unwell child to obtain adequate treatment overseas.Footnote ¹⁹⁷ Subsequently, several other cases at appellate level have left the door open for its recognition but have not definitively ruled on its status.Footnote ¹⁹⁸ Despite its tenuous acceptance amongst higher courts, this ground has been successfully relied on in a number of High Court decisions.Footnote ¹⁹⁹ One other instance of intervention for factual error is notable. The Privy Council in Erebus Royal Commission ruled the decision-makers must base their decision ‘upon evidence that has some probative value’, but characterised any failure to do so as a breach of the natural justice ground.Footnote ²⁰⁰ (Again, this reiterates the role of classification and preconception of errors under different grounds in order to attract deeper scrutiny.)

Other substantive grounds, such as unequal treatment,Footnote ²⁰¹ substantive fairness,Footnote ²⁰² and the innominate ground,Footnote ²⁰³ have had limited success, despite some strong efforts to have them recognised.

Circumscription

Circumscription of the grounds of review, under the guise of non-justiciability, is seen in New Zealand jurisprudence too. For example, non-justiciability, in its absolute sense, was deployed in the Court of Appeal's decision in Curtis v. Minister of Defence.Footnote ²⁰⁴ The Court avoided the question of the legality of the disbanding of the air strike force, reasoning that it was a political question which the government of the day should be held accountable for through political – not legal – processes. Partial non-justiciability is also evident in similar areas to those in England. Prosecutorial decisions are only subject to limited review.Footnote ²⁰⁵ Commercial decisions of quasi-public bodies are subjected to a circumscribed set of grounds; a particularly notable instance is Mercury Energy where the Privy Council, hearing an appeal from New Zealand, doubted that decisions of State-owned Enterprises were reviewable in the absence of ‘fraud, corruption or bad faith’.Footnote ²⁰⁶

Generality

The grounds of review schema – like its other categorical sibling, scope of review – is based in a regime of rules. The difference between the two lies in the abstractness or specificity of those rules. A grounds of review approach favours a few generalised triggers for judicial intervention over a myriad of specific rules.

In their most basic and practical sense, grounds of review are the bases on which administrative action can be ruled unlawful in judicial review.Footnote ³⁰⁵ In other words, as Fordham explains them, a ground represents ‘the “flaw” which justifies the Court's interference’.Footnote ³⁰⁶ Lord Phillips, in one of the few judicial definitions, characterised the threefold set of grounds as a ‘received checklist of justiciable errors’.Footnote ³⁰⁷ The definition of grounds of review as the flaw or error justifying judicial intervention is consistent with their common expression in negative terms, as was the case with Lord Diplock's formulation. Thus, the grounds meet Fuller's expectation of generality in the articulation of rules.

When expressed in their positive form, though, grounds represent the principles or norms that regulate administrative decision-making. The inverse of grounds of review may be characterised as ‘principles of good administration’ (Galligan), ‘norms of good public decision-making’ (Cane) or ‘standards of legality’ (Hickman).Footnote ³⁰⁸ Nothing particularly hangs on their negative or positive form, at least as the grounds are presently conceived. The failure to comply with a principle of good administration is treated as enabling the courts to intervene by way of judicial review (although, normatively, this need not be so).

In terms of the mediation of the balance between vigilance and restraint, each of the tripartite grounds of illegality, procedural impropriety and irrationality presents a fixed, but implicit, depth of scrutiny. In general terms, illegality and procedural impropriety enable correctness review, while irrationality poses a deferential standard. The classes of flaws brought together under each ground share a similar depth of scrutiny. The applicable depth of scrutiny is dependent on the characterisation of the impugned norm; in other words, which ground of review is relied on. Hence, the descriptor of (categorical or doctrinal) grounds of review.Footnote ³⁰⁹ As with the scope of review model, classification is still the central judicial function that dictates the depth of review.Footnote ³¹⁰

The traditional tripartite grounds continue, in many respects, to reflect the hallmarks of the ‘classic model’ of judicial review.Footnote ³¹¹ While Lord Diplock's seminal speech in CCSU is heralded as significant in the systemisation of judicial review doctrines, the extent of change should not be overstated. As Harlow and Rawlings observe, ‘Lord Diplock's three principles still conform largely to the classical grounds as they had evolved over the centuries’.Footnote ³¹² Formal distinctions still dominate, although there is less emphasis on ‘rigid legal categories’.Footnote ³¹³ The difference lies in Lord Diplock's attempts at developing generalised organisational principles from the doctrinal morass that existed previously.

The doctrinal shape of the grounds of review is more general than seen under the scope of review approach. Each may be divided into various different and more particular sub-principles. For example, illegality may be treated as capturing doctrines addressing matters such as error of law, improper purpose, relevancy, fettering of discretion and so forth. In this respect, the grounds have an aggregating function. This dimension was picked up by Lord Irvine in Boddington when he commended the way in which the tripartite grounds ‘compendiously grouped’ the various types of challenges that could be mounted.Footnote ³¹⁴ Others have also emphasised the nature of the grounds as categories of more specific doctrines.Footnote ³¹⁵ Allan goes even further and asks rhetorically whether the grounds of review are anything more than ‘labels [that] announce the conclusions of legal analysis’?Footnote ³¹⁶ Consistent with his preference for individualised judicial judgement over the application of general doctrine, he says pejoratively that ‘the settled grounds of review are really only conclusionary labels for judgments made on the facts of each case – judgments invoking controversial moral and political values’.Footnote ³¹⁷ The conclusionary potential of the grounds echo, to some extent, the indirect and rhetorical style seen under the scope of review method.

The final point about the nature of the doctrinal rules is the question of their intended audience.Footnote ³¹⁸ As noted earlier, the grounds of review have been articulated and applied as if they express public law norms and bases of judicial intervention simultaneously. In other words, these concepts are intertwined and conflated as a single (or series) of grounds of review. As Hickman notes, though, it may be helpful to distinguish between the two different functions of these public law principles, even though the different functions have ‘not been clearly distinguished, or even explicitly addressed, in the case law’.Footnote ³¹⁹ Hickman amplifies this distinction when he adopts the language of ‘standards of legality’ and ‘standards of review’ for analytical purposes. He explains the terminology in this way. Standards of legality are the rules or principles that are principally directed at the administration; that is, those standards to be satisfied over-and-above the condition expressed in the empowering instrument. In contrast, standards of review speak to the courts, not the administration, expressing how the courts determine whether standards of legality are breached. Thus, in relation to the illegality ground of review, for example, the standard of legality represents the conditions set by the empowering statute.Footnote ³²⁰ The corresponding standard of review imposed by the court in assessing compliance is correctness. ‘The question of whether the conditions of the statute had been fulfilled [is] a matter for the courts’, Hickman says – in contrast to interfering only if the administration's ‘understanding of the statutory requirements was unreasonable’.Footnote ³²¹ So, too, with procedural impropriety: the common law principle of natural justice operates as a standard of legality, while a correctness standard of review is adopted when assessing compliance.

Irrationality review is more complicated. Hickman describes Wednesbury unreasonableness as (principally) a standard of review. For Hickman, Lord Greene's statement that the courts can quash decisions that are ‘so unreasonable that no reasonable authority could ever have come to it’ describes ‘how the courts should exercise their supervisory jurisdiction over the administration’, not ‘a substantive principle to which government decisions must conform’.Footnote ³²² There is some force in Hickman's analysis that Wednesbury speaks mainly of a standard of review through the injunction to the courts to exercise restraint. However, a standard of review only makes sense if it is related to a standard of legality, that is, a norm which the administration is charged with satisfying. While at times Hickman's claim is that the tradition in English public law is to not impose substantive standards of legality, he tentatively concedes in a footnote that Lord Greene's test may also implicitly recognise a standard of legality:Footnote ³²³

On this analysis, therefore, the standards of legality and review are both reasonableness.Footnote ³²⁴ This is consistent with the positive formulations of grounds of review and principles of good administration which proclaim an obligation on the administration to act reasonably. For present purposes, it is sufficient to acknowledge that under the grounds of review schema, on an explicit level, the common law does not draw such a distinction and, on an implicit level, standards of review and legality are fused within the grounds of review themselves.

Hickman's distinction between standards of legality and standards of review is only employed for descriptive purposes. He expressly disavows any attempt to fashion the distinction in normative terms.Footnote ³²⁵ However, my argument is that schemata that explicitly recognise the distinction may be valuable for the purposes of organising judicial review doctrines and determining the circumstances of judicial intervention.

Even though the grounds of review purport to present prescribed and fixed depths of scrutiny, the grounds of review enable significant variability in judicial supervision. Judicial discretion is recognised in places in the schema but this is generally subordinate to the guidance provided by the rule-structure. That is, the generalised doctrinal structure of tripartite grounds is the centre-piece of the schema; judicial discretion infiltrates, latently, the classification process where the grounds overlap and operate, patently, in cases where the grounds emerge and evolve. Like the scope of review schema, the embrace of rules promotes generality, although the role of judicial discretion places a gloss on this virtue.

First, there is a degree of blurring between the distinctions framing the traditional grounds, explicitly acknowledged by the schema's concession to overlap. In such cases, this enables the depth of scrutiny to be modulated or manipulated, as explained in the doctrinal study. Taylor emphasises the normative dimension to this task in the following metaphor:Footnote ³²⁶

Secondly, we have seen other grounds evolve, along with continuing pressure for others to evolve or be reformulated (albeit they are at different stages of genesis). This enlarges the available suite of grounds, bringing more diversity to the supervisory task. The particular incentive to do so is to access a depth of scrutiny that is more intensive than the default deferential review under the irrationality ground. Once again, questions of classification arise. Access to the recognised emergent grounds is generally conditional: in order to rely on the ground, certain categorical preconditions must first be met. Assessment of whether those preconditions exist, provides a degree of judicial discretion and judgement about whether they are satisfied.

Thirdly, in some respects, the departure from established standards of review found in the traditional grounds has seen the adoption of more open-textured and flexible touchstones for intervention, such as abuse of power, fair balance and so forth. In doing so, the reviewing task is imbued with more discretionary judgement, implicitly enabling greater diversity in depth of scrutiny.Footnote ³²⁷

For the purposes of assessing the generality of the regime, the significant aspect is that the variability and judicial discretion are both latent and patent. The primary judicial task continues to be categorisation, identifying the appropriate ground of review for the circumstances of the particular case. In this part of the judicial reasoning process, the determination of the depth of scrutiny is indirect and the judicial discretion is latent – although there is some recognition of overlap necessitating judicial judgement and choice. In contrast, the evolutionary and circumscribing aspects of the grounds of review schema more openly disclose a degree of variability and judicial discretion.

Thus, the generality of the schema is relatively mixed, but is more favourable than seen in relation to scope of review. The method is grounded in rules. Judicial discretion is apparent and potent. There is some effort to give judicial discretion explicit doctrinal foundation, but aspects still remain latent.

Public Accessibility and Transparency

The generalised and systemised framework of grounds promotes accessibility and transparency. However, some aspects of the schema still disguise the factors determining the mediation of vigilance and restraint.

The high point for accessibility and transparency is the set of generalised grounds of review. They provide a clear framework guiding the judicial task, supporting the virtues of accessibility and transparency. However, the indirect manner by which the depth of scrutiny is set – by categorisation – means this schema still relies on classification as a proxy for the conceptual factors dictating the appropriate depth of review and takes away from this degree of openness. Judicial justification is generally framed in the language or form on which the grounds are cast, not the underlying conceptual drivers of vigilance and restraint.

The evolutionary aspects of the schema present a mixed degree of transparency. The ability to depart from the traditional and default grounds does not guarantee transparency about the conceptual basis for drawing the balance between vigilance and restraint differently. When judges engage recognised alternative grounds such as substantive legitimate expectations or proportionality, the focus tends to be the doctrinal preconditions which regulate the availability of these other grounds. For example, in the case of establishing a substantive legitimate expectation, it must first be established that an assurance was given in the nature of a promise (that is, triggering the prospect that the substantive expectation based on it may be protected); in the case of proportionality, enumerated rights must be implicated or the circumstances must involve the application of punitive sanctions. Thus, like the classification process, the motives for engaging the non-traditional grounds of review are not always apparent and the schema is not fully transparent; judges need only conclude that those preconditions are established.

The fact the schema sets out a traditional and default framework perhaps creates, at least passively, an expectation of reason and justification when departing from the default position. Certainly, when emerging grounds are forged, judges usually provide extensive and developed reasons justifying the new basis for intervention.Footnote ³²⁸ There is, though, no guarantee of the provision, nature or extent of the reasoning supporting evolutionary developments. For example, sometimes judges prefer to engage in more subtle doctrinal evolution to avoid taking dramatic steps or highlighting the new developments. Fordham identifies two main techniques whereby the evolution takes place incrementally through two phases: ‘temporary masking’ (an established principle is stretched to address a new problem but subsequently reinterpreted as a new principle) and ‘temporary divergence’ (a new principle is developed on a narrow basis and the existing orthodoxy is subsequently overruled in favour of the new principle).Footnote ³²⁹

That said, it can be argued that the tripartite grounds also have an explanatory function. The labels – whether conclusory or not – are instrumental in explicating judicial intervention in the circumstances. As Fordham says, the grounds of review are ‘the judges’ way of explaining when a public authority has overstepped the mark and when judicial intervention is warranted’.Footnote ³³⁰ In this sense, the grounds operate as a rhetorical device, marking out in shorthand the basis for intervention. As judicial creations,Footnote ³³¹ the grounds represent an attempt to express in a generalised way appropriate balances between judicial vigilance and restraint.Footnote ³³² The grounds therefore have a legitimising function.Footnote ³³³ The traditional grounds of judicial review represent, according to received wisdom, balances drawn between vigilance and restraint that are legitimate; emergent grounds must surmount the legitimacy threshold before they are accepted as legitimate alternative expressions of judicial oversight.

To illustrate, the underlying distinctions that infuse the grounds of review can be charted along two axes.Footnote ³³⁴ First, a strong distinction between substance and process is evident. Procedural impropriety is separated from illegality and irrationality because it addresses how the decision is made, not the decision itself. Procedural impropriety, in Fordham's words, ‘fits … with the notion of a truly supervisory jurisdiction’ – or, in the present language, ‘is legitimate’ – because it does not interfere with the substance or merits of the decision.Footnote ³³⁵ This distinction can be seen in the oft-cited – but nowadays questionable – mantra from Evans:Footnote ³³⁶

Similarly, while irrationality addresses the substance of a decision, its inherently deferential threshold means judicial oversight is unlikely to dig into the substance of the decision. Both grounds have ‘a built-in merits-avoidance mechanism’, as Fordham puts it: ‘procedural fairness because it is by nature only procedural, irrationality because its formulation is designed to acknowledge a margin of appreciation.’Footnote ³³⁷ Secondly, Fordham identifies a dichotomy between ‘hard-edged’ and ‘soft’ questions.Footnote ³³⁸ For those questions which, at least ostensibly, admit only a single and therefore ‘correct’ answer, it is treated as legitimate for the courts to substitute their view for that of the administration. In the English tradition, questions of law have been treated in this way and the resolution of questions of law remain the sole preserve of the courts.Footnote ³³⁹ Hence, the strictness of the illegality ground of review. In contrast, matters such as fact, judgement or discretion are treated as soft questions – matters on which it is illegitimate for the courts to intervene when exercising their supervisory review function. Questions of law and the strictness of the illegality ground can be contrasted to soft questions and the deferential Wednesbury review under the irrationality ground.

Each ground of review therefore has a central role in addressing questions of legitimacy of the judicial supervision, particularly its interventional legitimacy. Highlighting the trichotomy between review of fact, law, and discretion, Taggart says ‘the different standards of review in each category [of grounds] reflect functional, institutional, and pragmatic considerations, as well as legitimacy concerns’.Footnote ³⁴⁰ Thus, the grounds do, to some limited extent, have a role in rationalising and explicating the basis for intervention. But the explanation is encrypted in a label and therefore lacks transparency.

Prospectivity

This schema, like the others, is generally prospective. There are aspects of the judicial adjudication where there is some retrospective effect, such as when judicial discretion in the classification task affects outcomes or when the evolution of emerging grounds acquires some purchase in particular cases. However, when viewed relative to the other schema, the retrospective effect is not significant. In particular, the evolutionary aspects of the regime are generally not dramatic; emergent grounds are few and their development tends to be foreshadowed before they are actually realised as accepted grounds of review. Adjudicative discretion may generate some retrospective effect but the schema seeks to minimise the latent judicial role by mandating aspects of it and making it transparent.

Clarity

The generalisation and systematisation of the grounds of review aid clarity, avoiding the quagmire of rules that has plagued the scope of review schema. The schema, on its face, aims to present a simple set of principles guiding the circumstances of intervention. It still seeks to promote consistency and predictability, which the criterion of generality seeks to produce. Although not rigidly constructed, the generalised and simplified grounds are designed to be at the forefront of the judicial task and to anchor the questions about whether or not to intervene. The adoption of doctrinal grounds, in lieu of conceptual reasoning, favours clarity over transparency; the conceptual basis for intervention is merely gestured to, in the form of labels which implicitly summarise the legitimacy of intervention in particular circumstances.

Clarity is not universal, however.Footnote ³⁴¹ The admission of overlap, the fact the grounds aggregate a series of more specific sub-grounds, and the role of emergent grounds mean the schema has greater complexity than seen at first blush and greater potential for inconsistency when examined more closely. Moreover, it leaves a distinct role for judicial discretion; however, in contrast to scope of review, this discretion is more apparent on the face of the schema. While there is some departure from the rule-regime ideal of simplicity, clarity and generality, the schema goes some way to exposing the normative judgements involved.

Stability

The schema presents a reasonable degree of stability. For example, the tripartite grounds have endured since their original articulation in the early 1980s. As explained above, the regime provides for some evolution of the grounds. So far, the extent to which emergent grounds have been recognised is modest and limited. Other possible grounds remain inchoate, but are discussed because judicial review jurisprudence has sufficient flexibility to enable them to be explored and promoted in individual cases. The evolution of grounds tends to be measured and generational, rather than frequent and immediate. This means the evolutionary potential does not seriously compromise stability. However, relying on the inherent flexibility of common law review, this longer term evolutionary aspect may allow judges latitude and discretion to seek to deploy novel grounds with increased depths of scrutiny in occasional cases, notwithstanding a ground not yet receiving widespread endorsement. Regardless of whether the novel ground is embraced or disapproved in later cases, in the immediate instance judges may rely on the evolutionary potential to justify its deployment. Again, though, such a practice is not widespread and does not significantly undermine the stability of the schema in the way contemplated by Fuller's virtues.

Non-contradiction and Coherence

The rationale for the systematisation of doctrines into generalised grounds of review was schematic coherence. As explained above, Lord Diplock, in particular, was instrumental in the systemisation project; he proudly proclaimed the systemisation of administration law as the ‘the greatest achievement’ of the English courts in his judicial lifetime.Footnote ³⁴² Lord Donaldson echoed the impact of the tripartite grounds, noting that they were formulated ‘in an attempt to rid the courts of shackles bred of the technicalities surrounding the old prerogative writs’.Footnote ³⁴³ The purpose of the systemisation was to move judicial review doctrine beyond its doctrinal morass and to ensure a simple, unified, and coherent framework was adopted. As seen in the explanation of the nature of the rules above, the grounds of review approach is relatively successful in presenting a coherent structure. Moreover, the schematic coherence of the grounds of review approach is illustrated, in part, by the employment of the grounds by textbook writers. The systemised grounds allow the ‘orderly exposition’ of the bases of intervention.Footnote ³⁴⁴ Indeed, an analogy is frequently found between the grounds and the chapter headings of textbooks.Footnote ³⁴⁵ As Taylor explains, the grounds provide an analytical structure which is useful for instructive purposes; the groupings – or, again, ‘chapters’ – avoid the ‘unwieldy’ exposition of the bases of judicial intervention.Footnote ³⁴⁶ Thus, grounds of review performs well under this criterion.

Non-impossibility and Practicality

The simplified, systemised nature of the generalised grounds also ensures litigation is reasonably practical. The litigation process is supported by the operation of the established tripartite grounds, in most cases. These operate as useful guides for the purpose of framing and arguing cases, although non-traditional grounds present some challenges in the litigation process.

The tripartite grounds usefully frame argument in administrative law courts. ‘The grounds of review are the arguments which a lawyer can put forward as to why a court should hold a public authority's decision to be unlawful.’Footnote ³⁴⁷ The way the grounds of review provide structure in litigation, assisting submissions or judgments to ‘focus on the factual features of the decision or action said to be reviewable’ has been acknowledged.Footnote ³⁴⁸ This is recognised by the rules of civil procedure: claimants are required to identify the grounds on which a judicial review claim is made.Footnote ³⁴⁹ The readily understood depth of scrutiny associated with the grounds flows into the evidential corpus required. Like scope of review, review for legality, procedural fairness and unreasonableness does not necessitate vast amounts of evidence or cross-examination.Footnote ³⁵⁰ The symmetry between the standards of legality and standards of review – that is, their inverse relationship – assists in the deliberation process. Generally, the process of reasoning is simple and uncluttered. Judges’ attention is directed to the circumstances of the particular case, and need not engage in lengthy self-reflection about their own methodology or the applicable standards of review.

The simplicity and straightforward nature of litigation is tested because these emergent grounds may unleash different methodologies and different degrees of scrutiny. More vigilant grounds, like proportionality or legitimate expectation, require greater attention to the justification advanced by the administration and greater examination of the reasoning and supporting evidential basis. The classic model of judicial review – expedited and tightly focused in practical terms – does not fit this more vigilant approach. Meeting the evidential demands of greater scrutiny costs time and preparation.

As an example, the Lab Tests litigation in New Zealand shows the growth of the court record and the enlargement of time associated with arguments of increased vigilance.Footnote ³⁵¹ The incumbent tenderer for diagnostic testing reviewed a decision of a district health board to award the contract to another provider. Amongst other things, it argued that the Court ought to adopt a ‘broad-based probity in public decision-making approach’ to reviewing the decision (an approach that was adopted at first instance but overturned on appeal). As a result, the judicial review hearing took ten days to hear at first instance, and a further seven on appeal. The evidential corpus was large (68 affidavits and nearly 12,000 pages of documents) and written submissions lengthy (over 700 pages).Footnote ³⁵² Concern was expressed about the unorthodox length of the hearing, prompted especially by the more intensive standard of review and associated factual complexity. For example, Arnold J conceded the appeal court's judgment was ‘a lengthy judgment, much longer than is desirable’ but pointed to the need to ‘examine the evidence in some detail’, in the light of the plaintiff's claim for a supervisory approach which mandated ‘almost indeterminate scope for intervention by the courts’; his view was that the ‘factual and other subtleties’ were ‘too great to be dealt with in what is supposed to be “a relatively simple, untechnical and prompt procedure”’.Footnote ³⁵³ This can be compared to the experience and expectation of litigation under a classic model, where proceedings would often be heard in a day or so.Footnote ³⁵⁴

The procedural style of litigation associated with increased scrutiny is markedly different and more involved; we can describe it as the ‘plenary’ style of procedure, in contrast to the more modest ‘constrained’ style of procedure. The existence of two procedural styles raises the question of how the courts can accommodate each within the ordinary framework of review. In other words, practically, how can the courts modulate the quality of procedure in cases in which that is required? On the one hand, it is not desirable or feasible for the plenary style of procedure to be employed in every case. The costs associated with this would be significant. And presenting a full evidential corpus for intensive review in cases where traditional review only applies risks encouraging judges to engage in more vigilant review, without any normative basis. On the other hand, it may be difficult to anticipate those cases where increased scrutiny is justified; nor is it easy to demarcate evidence between low-intensity and high-intensity grounds, or to separate factually dependent grounds from abstract grounds. This compromises predictability and therefore the lessens the clarity of the schema.

That said, the conditional or partial nature of emergent grounds does ameliorate this to some extent. None of the emergent grounds is universal; as mentioned above, preconditions must be first satisfied before they can be reached and they have, to date, been relatively confined in operation. Thus, the courts must first be persuaded that the emergent ground is applicable in the circumstances of the particular case before the high-intensity procedural review is deployed. This limits, to some degree, the need for, and reliance on, high-intensity procedural review. However, these decisions are often not made at or before the substantive hearing of the case, diluting this gate-keeper function. It is possible that more sophisticated and robust examination of the claimed grounds of review at the preliminary permissions stage (where this is available) could improve this gate-keeping function and ensure the evidential corpus is commensurate with the realistically arguable grounds of review. However, there may be some reluctance to make judgements like this based on the sparse evidence available at preliminary hearings.

Congruence and Candour

The grounds of review schema performs relatively well in term of congruence and candour, balancing a need for consistency with flexibility – and generally encouraging congruence and candour on the part of judges. However, the ability to manipulate some of the key classifications takes away from congruence and candour. As explained earlier, doctrinal classification can obfuscate the unstated normative reasons for the classification.

Hortatory Versatility

Finally, the grounds of review schema is well-suited as a hortatory framework. The hortatory function is exemplified by the production of the various bureaucratic manuals, such as the Judge Over Your Shoulder guides produced in England and New Zealand, based on the simplified and systemised grounds of review.Footnote ³⁵⁵ The aim is to seek to improve awareness of the principles of good administration and promote compliance. The grounds of review are often cast in inverse terms to represent what have been described as principles of good administration or standards of legality.Footnote ³⁵⁶ They represent norms which ministers, public bodies and officials must comply with when exercising discretion. In other words, the corollary of grounds of review enabling intervention on the basis of illegality, irrationality and procedural impropriety (as per Lord Diplock) is an obligation on the part of decision-makers to act legally, rationally and with procedural propriety. This helps explain the usefulness of the ground for bureaucrats and their prominence in administrative guidance.

It is commonly observed that the number of administrative decisions which are subject to external review, particularly judicial review, is minuscule compared to the vast number of decisions actually made by the administration.Footnote ³⁵⁷ Thus, the hortatory role is important because it has the potential to reach parts of the administration that are not frequently exposed to external review and supervision. Moreover, these principles of good administration may be utilised by other public functionaries which have a grievance-remedying role, such as ombudsmen, auditors or administrative tribunals. Again, the simple expression of three bases of intervention supports this function.

The increasing complexity of the circumstances of judicial intervention, however, means a simple inversion of the grounds of intervention to articulate principles of good administration is no longer adequate. The growth in emergent grounds, with different depth of scrutiny review in different circumstances, creates tension between the statement of grounds in terms of standards of legality and standards of review. Take, for instance, the legitimate expectation ground. In some circumstances, defeating a legitimate expectation in a way that creates resultant unfairness or an abuse of power justifies judicial intervention. As a standard of review, the intensity may be approximated to simple unreasonableness (or, as some argue, proportionality), if the circumstances justify; if not, the default Wednesbury unreasonableness standard applies.Footnote ³⁵⁸ But how does this translate into a general standard of legality? The contingent and contextual nature of the ground inversion of the standard or ground does not readily translate into a general standard or norm that decision-makers should comply with. Instead, further reflection is required to elaborate a norm. Here, it is one of consistency or legal certainty (either acting consistently with promulgated policies or established practices on the one hand, or not reneging on promises or other assurances on the other) – this is the gist that runs through legitimate expectation. Thus there are some limits to the hortatory role of the grounds, particularly when the emergent grounds are engaged.