ńňđ. 14
(âńĺăî 19)



(4) The catalogue continued, with a list of things which are unauthorised (albeit, ‘less
obviously’)36 even though the decision makers are acting within their ‘general area
of jurisdiction’. Acting on the mistaken assumption or opinion as to the existence
of a certain event, occurrence or fact (commonly called a jurisdictional fact) or
other requirement, when the Act makes the validity of the decision maker’s acts
contingent on the actual or objective existence of those things, rather than on the
decision maker’s subjective opinion.

(5) Disregarding a relevant consideration which the Act required to be considered or
paying regard to an irrelevant consideration which the Act required not to be con-
sidered, in circumstances where the Act’s requirements constitute preconditions
to the validity of the decision maker’s act or decision. These mistakes constitute
one form of error of law. For the same reasons as apply to item 6 below, an infe-
rior court’s ‘relevancy’ errors will ordinarily be non-jurisdictional errors of law,
whereas the same error by a tribunal is more likely to be jurisdictional.
(6) Misconstruing the decision maker’s Act (another form of error of law) in such a
way as to misconceive the nature of the function being performed or the extent of
the decision maker’s powers.

Craig acknowledged that this item can involve very fine line-drawing. It added
that, generally speaking, errors of law committed by inferior courts will not
amount to such misconceptions of the nature of their functions, because it is
an ordinary part of an inferior court’s functions to give authoritative interpreta-
tions of the law. In this respect, Craig proposed a presumptive difference between
inferior courts and tribunals, because the latter’s functions do not ordinarily
include the authoritative determination of questions of law. Being merely an
interpretive presumption, not all cases have adopted Craig’s approach of char-
acterising all tribunal errors of law as jurisdictional. In particular, industrial
bodies and those dealing with ‘small claims’ seem still to be accorded some lee-
way for making non-jurisdictional errors of law even though they are ‘tribunals’
rather than ‘inferior courts’.37 Craig’s reasons for according the concept of a non-
jurisdictional error of law less space in tribunals than in inferior courts apply with
even greater strength to bureaucrats, although they did not rate a mention in the
Craig was built on an earlier catalogue given by Lord Reid in Anisminic Ltd
v Foreign Compensation Commission.38 Lord Reid’s list was longer by two items.
However there is no reason to doubt that Craig would have included Anisminic’s
further items had it thought of them.39 These were:

1. Bad faith.
2. Breach of natural justice.

Interestingly, Australia’s inclusion of natural justice in the catalogue was not
finally confirmed until Re Refugee Review Tribunal; Ex parte Aala,40 decided some
five years after Craig. The government had tried to persuade the Aala court that
when the Constitution came into force, prohibition was issued only for ‘want or
excess of jurisdiction’, not for breach of natural justice. The argument seemed
to tolerate the contraction of ‘want or excess of jurisdiction’ to ‘jurisdictional
error’,41 but resisted its incorporation of the natural justice grounds of review
which should, it maintained, be regarded as entirely distinct. The court gave
a number of reasons for rejecting the government’s argument. There were in
fact some pre-1900 precedents for treating a breach of natural justice as taking
its author beyond ‘jurisdiction’, at least partially undermining the government’s

argument from history. But even if one were to concede the history, the court
said that the real concern of the constitutional writs was to check want or excess
of federal power. It said further that it was consistent with this concern that
breaches of natural justice be treated as ‘jurisdictional errors’ for the purposes
of the constitutional writs, just as they are so treated nowadays at common law.
Gaudron and McHugh JJ subjected the government’s argument to their own
historical examination, not (as Kirby J alleged) because they thought that any
earlier word usage might be controlling, but because they thought that there
was a consistency of theme to all of the cases, old and new. The theme was
the restraint of breaches of statutory power. Kirby J was dismissive of the joint
judgment’s ‘tedious and largely unilluminating’ historical examination which was
borrowed, he said, from the Minister’s submission. The joint judgment returned
fire, criticising Kirby J’s preference for essentialist ‘intuition or divination’ over
legal scholarship.42 Hayne J had no desire to enlist in that particular battle, and
was content to acknowledge the difficulties in distinguishing between errors of
law that were jurisdictional and those which were not.43 His Honour’s approach
was to say that there was no harm in assimilating the natural justice grounds into
‘jurisdictional error if that had not already occurred’.44
Aala’s overall message could not have been clearer. Regardless of whether one
regards ‘jurisdictional error’ as a handy conclusion (joint judgment) or unneces-
sary historical distraction (Kirby J), the judgments are all agreed that it has assim-
ilated all the grounds for attacking the exercise of a statutory power as a nullity.
There are only two differences of any note between Craig’s catalogue and
Anisminic’s. First, Craig left more room for unreviewable legal errors if com-
mitted by inferior courts. The English judges had also lingered briefly with a
‘presumption’ that some legal issues were within the final ‘jurisdiction’ of inferior
courts but not other decision-making bodies,45 but they soon expanded juris-
dictional error to encompass almost all errors of law.46 Of course, the virtual
abolition of non-jurisdictional errors of law would not necessarily entail a con-
clusion that all errors are henceforth jurisdictional, because there would still be
room for non-jurisdictional errors of fact or discretion. However, at least one
English commentary now contains considerable slippage between announcing
the near-irrelevance of the distinction between the two sorts of errors of law, and
the near-irrelevance of the more general distinction between jurisdictional and
non-jurisdictional errors.47
Other commentaries are more careful,48 but it was perhaps inevitable that one
would begin to see cases characterising all reviewable errors as ‘errors of law’.
The English Court of Appeal now sees errors of fact as errors of law if they were
decisive and unfair.49 The court acknowledged that its conversion of fact to law
might appear ‘paradoxical’,50 but it is surely worse than that. Its net effect would
be to replace one conclusory term (jurisdictional error) with a term (error of law)
which until then had not been conclusory. That is an effect which Professor Craig
would regret.51 Australian law is far from clear about the distinction between

errors of law and fact,52 but it has shown no signs of treating it as a catch-all label
to summarise all of the available grounds of review, whether they be for errors
of fact or law.53
The second difference between the Anisminic and Craig catalogues is evi-
denced by Lord Reid’s preference of restricting ‘jurisdiction’ to its ‘original sense’
of a ‘tribunal being entitled to enter on the inquiry in question’. Lord Reid was
happy to acknowledge that the reviewable errors committed after that point
would result in nullity, but thought that describing them in terms of jurisdiction
would be to use that term ‘in a very wide sense’.54 There is an explicit reference
in the Australian High Court to Lord Reid’s distinction between what he called
the original and wider usages of ‘jurisdiction’,55 and there is no shortage of other
cases giving what might be called a ‘chronological tinge’ to ‘jurisdiction’.
Anisminic’s catalogue had concluded with the observation that it was not
exhaustive. Similarly, the High Court acknowledges that its Craig catalogue is not
exhaustive.56 Indeed, how could it have said anything else? The straight-forward
examples of jurisdictional error are far out-numbered by the sorts of errors which
Lord Reid thought had stretched the original understanding of the term, and
which Craig admitted were less obvious57 and sometimes required the drawing
of lines which ‘may be particularly difficult to discern’.58 Such admissions go back
a long way in Australia. Many High Court judgments have distinguished between
‘actual’ and ‘constructive’ failures to exercise jurisdiction.59 That can be traced to
a judgment of Jordan CJ in Ex parte Hebburn Ltd; Re Kearsley Shire Council,60 in
a passage to which the High Court has referred many times, and which Gaudron
J once described as ‘[t]he classic statement as to what constitutes constructive
failure to exercise jurisdiction’.61 It is worth quoting a portion of the old classic,
because its indeterminacy is revealing:62

I quite agree that the mere fact that a tribunal has made a mistake of law, even as to the
proper construction of a statute, does not necessarily constitute a constructive failure
to exercise jurisdiction. But there are mistakes and mistakes; and if a mistake of law
as to the proper construction of a statute investing a tribunal with jurisdiction leads
it to misunderstand the nature of the jurisdiction which it is to exercise, and to apply
‘a wrong and inadmissible test’; or to ‘misconceive its duty’, or ‘not to apply itself to
the question which the law prescribes’; or ‘to misunderstand the nature of the opinion
which it is to form’, in giving a decision in exercise of its jurisdiction or authority, a
decision so given will be regarded as given in a purported and not a real exercise of
jurisdiction, leaving the jurisdiction in law constructively unexercised, and the tribunal
liable to the issue of a prerogative writ of mandamus to hear and determine the matter
according to law.

These ‘constructive’ failures to exercise ‘jurisdiction’ have some common
threads. They are not exhaustive. Further, they are all committed during a pro-
cess which was properly commenced, but which went off the rails because of an
error going to ‘jurisdiction’, if one uses that term in a way described by Lord Reid
as wide and by the High Court as less obvious.

Death of the pure theory of jurisdiction

Lord Reid was right. ‘Jurisdiction’ used to have a narrower meaning, although
one might well ask why that should still matter.
Mr D M Gordon wrote a series of articles advocating what he called a ‘pure
theory of jurisdiction’.63 He had wanted this to apply to all cases of judicial
review of administrative action of an adjudicative nature, no matter what the
context, and no matter how egregious the decision maker’s error might have
been. Gordon’s starting point was a definition, that ‘jurisdiction’ meant only the
authority to decide. His next step was to deny that there could be any exceptions
to or extensions of his definition.
Gordon’s easiest examples came from decisions involving challenges to deci-
sions of justices of the peace. He acknowledged that their jurisdiction could be
limited in terms of subject matter (for example, they had no jurisdiction in equity),
or by the monetary amount at stake, or by the relief claimed or claimable (for
example, they could not determine title to land or grant injunctions). These were
matters, he said, which could be determined at the outset. Once a matter properly
got under way, however, the justices could not ‘lose’ or ‘exceed’ their jurisdiction
by an error of reasoning or procedure. He could cite an impressive list of author-
ities in support of his argument, perhaps the most famous being Lord Sumner’s
judgment in R v Nat Bell Liquors Ltd:64

A justice who convicts without evidence is doing something that he ought not to do,
but he is doing it as a judge, and if his jurisdiction to entertain the charge is not open to
impeachment, his subsequent error, however grave, is a wrong exercise of a jurisdiction
which he has, and not a usurpation of a jurisdiction which he has not.

The idea that decision makers cannot lose their jurisdiction once they have
properly passed their starting points is completely at odds with the now-classic
acceptance of constructive failures to exercise jurisdiction. Anisminic65 discussed
many cases, but the only case it felt compelled to overrule was Davies v Price,66
which had depended on an acceptance of Lord Sumner’s logic to reject an argu-
ment that applying the wrong legal test could sometimes amount to a jurisdic-
tional error.
Mr Gordon’s articles contained some valuable historical insights into the evo-
lution of the prerogative writs, and they are still cited in that context.67 They
also acknowledged an impressive line of precedents which were contrary to the
approach he advocated, making his argument more than just an appeal to a nar-
row, definitional logic. It was in effect an appeal not just to restrict, but to reverse
the growth of the grounds of judicial review. That appeal is at least seventy years
too late.
Even at the time Mr Gordon started advocating his position, however, his
definition left too much either unexplained or explained unconvincingly. Most
obviously, it failed either to explain or even to have a label for judicial review for
fraud, improper purpose, or breach of natural justice. It also failed to allow for

jurisdictional fact review,68 although others got around that problem by sub-
dividing a matter that might otherwise be properly before a decision maker. Case
after case talked of issues that were collateral, or preliminary, or even ‘essential
preliminaries’, and some of those predated Lord Sumner by at least a generation.69
Despite the terminology, however, there was no pre-ordained way of recognising
what issues were core or preliminary. Even the idea of an essential preliminary
lost its tether to any sense of chronology70 when it found itself being deployed
to ‘explain’ jurisdictional fact review (jurisdictional facts being essential prelimi-
naries). It is still used in the High Court71 and elsewhere, but it seems to indicate
nothing more than a severable or separate issue which is ‘jurisdictional in the
sense that the decision maker must get it right on pain of being overturned on
judicial review for jurisdictional error’.
To be fair, Mr Gordon’s arguments distinguished between the truly judicial
decision makers (inferior courts) and tribunals and others. Indeed, he was criti-
cal of overarching accounts of the judicial review grounds which failed to draw
such contextual distinctions, and he always opposed resort to the terminological
devices of preliminary issues and jurisdictional facts.72 However that left him
opposing too many decisions for want of proper fit with his logical definition,
and too large a scope of judicial review not just opposed but wholly unexplained.
For example, he failed to adjust for the merger of what counts as ‘jurisdictional
error’ and what counts as ‘ultra vires’. The two doctrines started life in different
fields, the first in relation to judicial officers and (eventually) those with a duty
to act judicially, the latter in relation to statutory bodies (classically, corporations
in the days when most companies owed their existence to individualised Acts).
However, each was driven by similar rationales of containing bodies within the
limits of their powers, and searching for those limits in the context of the par-
ticular Act.73 Separating the two doctrines became impossible with the spread
of the administrative state, which saw the administrative branch of government
invested with powers properly characterised as legislative, executive and judicial.
Anisminic ended any attempt at papering over their merger. The terms are now
used interchangeably.74
Context is indeed important to any judicial review case, and the prime con-
textual variant is surely the relevant statute. If the statute is better interpreted
as stipulating certain preconditions to a valid decision, why should it matter that
some of them fail to fit an a priori definition of jurisdiction, even if that definition
was once an accurate description in times well-past?
History aside, however, there is one aspect of Gordon’s critique that still has
resonance. Gordon repeatedly criticised the judges for failing to come up with
a rational theory to explain why they would choose to characterise some issues
as jurisdictional and others as non-jurisdictional. He declared that ‘in no branch
of English law [than that dealing with jurisdiction] is there more confusion and
conflict’.75 Any serious examination of the cases, he said, ‘must convince the
open-minded inquirer that there is virtually no proposition so preposterous that
some show of authority to support it cannot be found’.76 The cases presented ‘a

hodgepodge of contradictory and inconsistent rulings, and . . . an aggregation
of sophistical and even absurd reasons to justify rulings’.77
One cannot sustain an argument based ultimately on the proposition that
there is (or was) only one meaning of a term as slippery as ‘jurisdiction’.78 Nor
should one hope for a single, over-arching theory of statutory interpretation.
That said, it is often as difficult now as it was when Gordon was writing to
predict which statutory requirements will be treated as mandatory (using the old
language), and which will be characterised as directory. The usual consequence
of ‘jurisdictional error’ is court action treating the challenged decision as a nullity.
The High Court tells us that nullity is ultimately a consequence directed by the
particular Act in question, which is only another way of telling us that whether
a requirement is mandatory (and therefore jurisdictional) is ultimately a matter
of statutory construction in each particular case. Small wonder, then, that the
same High Court decision which tells us these things also admits that the result of
this particular constructional exercise often ‘reflects a contestable judgment’.79
It cannot be otherwise.
Pure theories have no place in statutory construction, but is statutory inter-
pretation entirely unpredictable? Is it utterly variable between cases, depending
on the predilections of each judge, who is free to insert whatever implied con-
ditions they like into the particular statute and as free to characterise statutory
conditions as mandatory or directory as they like? This chapter opened with one
of Kirby J’s more colourful criticisms of jurisdictional error. His Honour said in
another case that the terminology is ‘meaningless’80 unless it is either informed81
(or preferably replaced)82 by higher-level general principles. That approach is
surely correct. Forget hang-ups with labels, and try and work out what lies behind

Jurisdictional errors: Can values be rules?

It is not too much of an exaggeration to characterise Australia’s law of judicial
review of administrative action as predominantly a bottom-up affair.83 We do
have our grand principles, of course, but they are so grand as to belong more
properly to the field of constitutional law. The rule of law, the principle of legality,
the separation of powers, even the recently advanced ‘integrity principle’ – these
offer very little guidance to anyone wanting to know what the courts might
commonly regard as the minimum standards of good public administration.
Indeed, most Australian judges deny a direct role in articulating or setting
minimum standards of administrative decision-making.84 Kirby J is a dissenter.
His Honour acknowledges that judicial review can usually make do with its spe-
cific grounds of review, perhaps informed by the judge’s perception of ‘serious
administrative injustice’. However, in the ‘exceptional’ case where there is serious
administrative injustice which does not fit within any of the established review
grounds, his Honour believes that the court can and should act.85

An interesting trend in recent reforms to public service legislation has been
the statutory formulation of a list of ‘values’ such as political neutrality, honesty
and fairness,86 but there is a complete absence of judicial review cases treating
the violation of a prescribed value as a ground of review.87 There seems to be
a tacit consensus that these issues are better left to the Ombudsmen, or to the
public sector managers. There is a good deal of sense in this at what one might
call the micro-management level, because one should not expect judges to know
much about theories of good and effective public sector management, let alone
to keep abreast of developments in management theory and practice. Even at the
more general level, however, we find Australia’s senior judiciary reacting strongly
against recent English suggestions of working from the ‘top down’.
English courts pushed the ground of Wednesbury unreasonableness to its lim-
its, increasingly impatient with the idea that it applied only to decisions which
were utterly bizarre. They finally kicked over the traces when the European Court
of Human Rights told them that even if it was applied in a stretched fashion
(‘anxious scrutiny’ was the English term), Wednesbury unreasonableness was too
lenient with the administration when it came to the protection of human rights.88
So the English courts went further, and embraced ‘proportionality’ review. At
roughly the same time, they were also reinterpreting some of their older cases,
particularly those which involved Wednesbury unreasonableness or ‘legitimate
expectations’. They decided that some of the more restrictive cases could no
longer stand, and that a good many of the remainder would be better understood
as cases evidencing judicial review for substantive unfairness per se. Having taken
the leap into substantive unfairness, it was but a short step to their recognition
of judicial review for bureaucratic unfairness, or abuse of power,89 and then only
a slightly bigger step further into judicial review for a tribunal decision that was
‘unfair’ because ‘fresh evidence’ had subsequently revealed error in its factual
This is all very different from the typical Australian judgment, which con-
centrates on lower-order issues, the specific ‘grounds’ of judicial review. ‘Abuse
of power’ might well be a correct summation of the values underlying the spe-
cific grounds, but it lacks ‘an immediate normative operation’91 in this country.
In other words, ‘abuse of power’ or substantive ‘unfairness’ or the protection of
‘legitimate expectations’ might justify the operation of more specific legal rules
or principles, but they are not grounds of review in their own right.
It is submitted that the courts cannot sensibly operate solely on either a top-
down or a bottom-up approach. They need to do both, and probably cannot avoid
it even if this is not always acknowledged. Grand value statements are usually too
indeterminate to do more than provide general guidance, but by the same token,
one cannot operate the more precise grounds of review in a vacuum, divorced
from any sense of their proper fit with each relevant administrative or regulatory
A recent High Court decision provides a good example of the tensions involved
in determining the relevance or status of notions of ‘good administration’. The

challenge in NAIS v Minister for Immigration and Multicultural and Indigenous
Affairs92 was to a Refugee Review Tribunal’s decision which had been six-and-
a-half years in the making, with the main oral hearing held roughly five years
before the decision itself. The challengers quite reasonably suggested that the
long delays would have seriously diminished the Tribunal’s capacity to remem-
ber the witnesses’ demeanour or body language when they had testified, and yet
demeanour may well have played a real part in the Tribunal’s assessment of their
credibility. For the majority, that meant that the procedure had been so unfair as
to amount to a breach of natural justice. There was a real risk that the delay had
deprived the Tribunal of its capacity to make demeanour-based assessments of
credibility. This was a breach of natural justice. Gummow J protested in dissent
that this amounted to the imposition of judicial process standards upon an admin-
istrative process, and his Honour implied that this in turn amounted to review
for maladministration per se. Kirby J’s response was frank. His Honour acknowl-
edged the parallels between this case and the principles relevant to appeals from
tardy judges,93 and said that the distinction in this context between review on
the merits and judicial review failed to give him ‘much assistance’:94

Where judicial review is sought on the grounds of breach of the requirements of pro-
cedural fairness, it is precisely the merits of the way the decision-making power was
carried out that is at issue. If that power is exercised in a manner that is unfair, within
the authorities on procedural fairness, the decision may be invalidated by jurisdictional
error for that reason.

Going beyond jurisdictional error

It has been seen that in the cases of certiorari, injunctions and declaratory relief,
the general law allows judicial review for illegality not necessarily constituting
jurisdictional error. So, too, do some of the grounds of review in the ADJR Act.
Most of its grounds are a direct steal from the common law, but at least some of
these may not be tethered to a requirement of jurisdictional error or invalidity.
There are explicit hints in that regard from the High Court as regards ADJR’s
‘procedural error’ ground,95 and it is an assumption underlying much of the
reasoning in Jadwan Pty Ltd v Secretary, Department of Health and Aged Care,96
where the ADJR ground was taking irrelevant considerations into account. As for
ADJR’s ‘error of law’ and ‘no evidence’ grounds, these self-evidently push further
than the common law’s analogues, and in so doing, push beyond jurisdictional
The net result is that many contexts no longer require jurisdictional error, and
in the case of certiorari, injunction and declaratory relief, some contexts never did
have such a requirement. One consequence must be that nullity (jurisdictional
error’s usual consequence) is less of an organising concept than in former times.
Another consequence will surely be that as more and more contexts begin to
de-couple from ‘nullity’ and ‘jurisdictional error’, the courts will have to become

more context-sensitive. They will increasingly need to give more consideration to
granting remedies which do not automatically treat the successfully challenged
decision as having been a nullity from its inception.98

Is jurisdictional error conclusory? Yes, but that in itself is not a bad thing, provided
we focus on what counts as jurisdictional error, and why. Should jurisdictional
error be abolished? No, if that means copying England’s switch to an equally
conclusory label of ‘error of law’, and no, if it means ditching the concept of
nullity altogether. It is clear, however, that jurisdictional error is in retreat in
several contexts. From this author’s perspective, that is a cause for neither tears
nor celebration, but simply a reminder of the need to get behind the labels,
conclusions and grounds to try and ascertain their rationales and assess their
Privative clauses and the limits
of the law
Mary Crock and Edward Santow

Judicial review and privative clauses

The idea that courts or other legal bodies should play a role in overseeing admin-
istrative action is central to modern notions of democratic governance. However,
it seems to be in the common law countries – United Kingdom, United States,
Canada and the nations of the British Commonwealth – that the most complex
oversight regimes have been created. This may be because of the sometimes nebu-
lous distinction drawn in those countries between administrative review (review
of the merits of administrative action) and judicial review (review of the legality
of such action). The distinction is manifest on the one hand in the creation of both
specialist and multi-jurisdictional tribunals or agencies charged with the review
and/or re-making of administrative decisions. On the other hand are courts of
law vested with constitutional or statutory authority to check that administrative
decisions (including decisions made by those tribunals or agencies) have been
made in accordance with the law.
It is a system built on a sequencing of functions between administrators, tri-
bunals and courts, which are arranged in a natural hierarchy. This can militate
against ‘efficiency’ in both administration and governance, in that a system involv-
ing decision-making by representatives of two branches of government creates
a necessarily complex matrix of avenues for review. Thus in a sense, the Anglo-
Australian system of judicial review is less efficient than, say, the French droit
administratif. However, as Sir Gerard Brennan has observed, judicial review is
an ‘intended . . . fetter on the Executive’s pursuit of its policies’ and so a certain
level of inefficiency is ‘inevitable, if not intended’.1


Put simply, in Australia, as in other common law countries, judicial review
has developed as a forceful weapon in the hands of dissenters. Recourse to the
courts can lead to judicial pronouncements on matters of process or on the inter-
pretation of legal principle that invoke antagonistic responses from government.
At a practical level, the sheer volume of court challenges lodged can result in
lengthy delays in the final disposition of matters. Sometimes where the outcome
of an administrative process represents a potential detriment to the applicant,
delay can represent an advantage to the applicant. In cases where the respon-
dent wishes to avoid the imposition of a penalty or the making of an order, the
delays inherent in judicial review proceedings can be an equally potent tactical
In this chapter, we examine various ways in which government has responded
to these challenges by attempting to constrain review of administrative action
by the courts. The most contentious vehicle of restriction is known in Aus-
tralia as the ‘privative clause’.2 This presents most often as a statutory device
by which Parliament purports expressly to restrict the manner or extent of judi-
cial review in respect of an identified class or classes of administrative decision.
The intended effect is to reduce (or even to remove altogether) the tiers of judicial
review available to challenge certain administrative (and sometimes, judicial3 )
Privative clauses in their various incarnations are at the cutting edge of admin-
istrative law inasmuch as they highlight significant points of conflict in the admin-
istrative process. They also represent a point of reference within the broader
context of relations between politics and the legal process. Privative clauses are
subject ultimately to the constraints imposed by the federal and state Constitu-
tions, to the extent that these pronounce on the respective powers and duties of
the three arms of government. Accordingly, while parliaments enacting privative
clauses may indicate a strong intention that the judiciary refrain from review-
ing certain classes of decision, the courts have on occasion resisted the call for
restraint in deference to higher duties. It is the thrust and parry between parlia-
ment and the judiciary that makes the subject of privative clauses both confusing
for students of administrative law and endlessly fascinating for practitioners and
One of the reasons why privative clauses engender great controversy is that
they often stand at the boundary – and even seek to define the boundary –
of matters that are within the province of the courts and matters that are not.
Indeed, rulings by the courts on the effectiveness of attempts to oust judicial
review are often tied closely to the doctrines that have developed around the
central question of what is and is not justiciable by the courts. If privative clauses
raise issues about when courts can be excluded from review, they also raise ques-
tions about what are the limits of the law, or what will constitute illegality in
any given situation.5 In this context, it is not surprising that the discourse on
privative clauses is full of what Julius Stone would call ‘categories of illusory
reference’.6 When presented with such clauses, the courts have often asserted

their right to review, but then declined to offer relief to applicants because no
legal error is apparent.7 On other occasions they have acknowledged the pres-
ence of legal errors, but declined to intervene on the basis that the privative
clause is an effective barrier to review. This is where we see judicial review at its
most political; where cases cannot be understood outside of their context and a
complex interplay of factors.
It is beyond the scope of this chapter to compare in detail Australia’s expe-
rience in the use of privative clauses with that of other common law countries.
The United Kingdom and the United States make two comparators of particular
interest. The United Kingdom is a country without a written constitution but with
a rich tradition of respect for the judiciary as the final arbiter of what passes for
the rule of law. Its historical jurisprudence on privative clauses forms the start-
ing point for judicial thinking on the subject in Australia as in most common law
countries. The United States, on the other hand, does have a written constitu-
tion but this document does not entrench expressly the right to judicial review
of administrative action. Such rights have had to be implied using elements of
America’s constitutionally entrenched Bill of Rights – in particular the right to
due process and the ‘habeas’ clause, which entrenches the right to freedom from
arbitrary imprisonment. If Australia’s constitutional matrix is different again, its
jurisprudence finds interesting parallels in both of these foreign jurisdictions.
In the United Kingdom and the United States, Parliament and Congress respec-
tively have chafed against the supervisory role of the courts, enacting various
iterations of privative clauses. Although the debate is far from closed in either of
these countries, it would seem fair to say that in each instance the courts have
mounted spirited defences against the constraints placed on them.8 How the
Australian courts have reacted to similar challenges is the focus of this chapter.
In this chapter, we examine the significance of judicial oversight and the dif-
ferent ways in which governments have tried to curtail curial review: privative
clauses present in many different guises. This leads inevitably to a discussion
of the weapons in the judicial armoury – most particularly to the concept of
‘jurisdictional error’. As kryptonite is to Superman, so is the ‘jurisdictional error’
fatal to the effectiveness of most privative clauses. The frustration of this area
of administrative law is that every attempt to enunciate firm principle must be
heavily qualified. Attempting to unravel all the strands of doctrine is no easy
task. Our central argument is that attempts to constrain judicial review cannot
be understood fully at the level of high legal principle. While viewed first and
foremost as aids to statutory interpretation – and an expression of the will of
Parliament – privative clauses can only ever be understood when fully context-
ualised. They are introduced to deal with specific problems or issues and have
been interpreted by the courts in ways that reflect the nuances of context. Two
case studies are used in detail to explore this concept. We examine the courts’
treatment of privative clauses in the fields of immigration and industrial rela-
tions. The chapter concludes with a broader reflection on the future of privative
clauses (see pp. 363–7).

An overview of the issue

In the common law system, judicial review is a vital check on administrative
action, designed to guard against the natural fallibility of administrative deci-
sion makers.9 This is so whether the role of the courts is viewed as one based
on upholding principles laid down by Parliament (in statutory law)10 or by
the courts themselves in the ever evolving common law process.11 Privative
clauses are controversial because they have the potential to reduce the means
by which a person wishing to challenge an administrative decision may seek judi-
cial review of that decision. Although devised most commonly to limit oversight
of administrative review bodies, such clauses can operate also to prevent appeals
from judicial bodies in the form of both inferior courts and superior courts of
Administrative decision-making has expanded such that, by the twenty-first
century, there has developed a vast range of activities subject to ‘government
scrutiny, permit or control’.13 In areas such as migration law and industrial rela-
tions (areas in which legislatures have traditionally sought to enact privative
clauses), administrative decision makers are empowered to make determinations
affecting a person’s fundamental rights. If the decision-making process is flawed,
a privative clause can operate to prevent an affected person from obtaining legal
redress. This, in turn, can have a devastating impact on the individual affected.
For instance, a person wrongly denied refugee status, who is then sent back to
their country of origin, may conceivably face torture or death. As we explore later
(see pp. 360–3), it is not surprising that the courts have resisted constraints on
their ability to review these types of decisions.
Having said this, it is a mistake to assume that privative clauses will have no
impact at all on the manner in which a court will approach the task of judicial
review. In this context, it is well to return to the question of why legislatures enact
privative clauses at all.14
A privative clause is, perhaps above all else, a mechanism by which the Exec-
utive attempts to retain control over a particular decision-making process. This
might be deemed desirable where, for instance, an administrative body is given
the power to make decisions over the allocation of a finite resource. Parliament
may be concerned that judicial intervention in this process could ultimately lead
to the administrative body being required to make allocations which go beyond
the availability of the resource in question. By the same token, the decisions in
question may be ones of ‘high’ policy, even presented in the form of subordi-
nate legislation: judicial review in such instances being unwelcome as a judicial
usurpation of executive power.
Secondly, where the ability to make a certain decision requires considerable
expertise in a particular field of knowledge, the legislature may wish to ensure
that the relevant decision makers are sufficiently qualified in that field. Parlia-
ment may believe that it would be too difficult to reach a rational, fair and just
decision unless the decision maker possesses the relevant expert knowledge.

Almost by definition, courts of general jurisdiction are not expert in every field
of knowledge. For this reason, Parliament may take the view that a court is only
equipped to ensure that the minimum procedural standards are met by the expert
decision makers, but that any question touching on the merits of the decision is
beyond the court’s competence. The Administrative Review Council (ARC) refers
in this context to the inappropriate use of judicial review in decision-making that
requires the balancing of ‘polycentric factors’.15
Finally, legislatures enact privative clauses because they perceive limitations
with judicial review itself. Certainly, judicial review is no panacea. It does not
provide the means for rectifying every type of administrative error. Judicial
review can only be used to remedy errors relating to the process of decision-
making, rather than the merits of the decision itself. In most instances, success
in judicial review proceedings will result in no more than the referral of a mat-
ter back to the original decision maker so that the decision-making process can
start afresh. Moreover, when courts review the legality of an administrative pro-
cess they tend to enforce minimum standards. Put bluntly, judicial review cannot
transform a statutory decision-making process that is elementally inequitable or
discriminatory into one founded on equality and respect for individual dignity.
On the other hand, judicial review is susceptible to abuse in instances where the
delays involved in seeking curial review can constitute an advantage for one or
other party.16
All of these matters (and more) must be weighed by the courts when presented
with legislation that appears on its face to be saying ‘Judges: Keep out’.

Identifying privative clauses
There are many ways in which legislatures attempt to shield administrative pro-
cesses from curial review. Perhaps the most obvious mechanism is through the
codification of decision-making in a way that reduces or removes altogether the
discretion in decision makers. For example, a prime motivation for the creation
of detailed regulations in the areas of taxation, social security and immigration
is that statutory rules reduce the scope for arguing about how decisions can be
made (as a matter of law). The problem, of course, is that lawyers are born and
bred to argue about the interpretation of legal rules that appear on their face to
be straight forward. Regulations that provide ‘shopping lists’ of matters required
to be taken into account have attracted applications for judicial review where the
courts have attacked even minute departures from the text of the legislation.17
Accordingly, legislators on occasion have resorted to the interposition of ‘sub-
jective’ powers in decision makers. The legislation in this instance is characterised
by phrases such as ‘in the decision maker’s opinion’ and ‘where the decision maker
is satisfied that’. This statutory language is designed to restrict curial review in
that the essence of legal decision-making is no longer a shopping list of provi-
sions, but whether there is any legal basis upon which the decision maker could
have reached the decision in question.18

Another way in which the role of the courts can be minimised is by creating
structures that redefine the domus of administrative decision-making – for exam-
ple, by vesting powers in ‘private’ bodies. In these instances, the statutory scheme
will permit aspects of a decision-making process to be performed by individuals
or corporations engaged as private contractors to carry out specific activities.
Controversially, such contractors have been found to be operating outside of the
accountability restraints of public administrative law.19
The traditional privative clause, however, is a more direct statutory device that
purports to limit the courts’ ability to review nominated classes of administrative
decisions. The devices go by many names, including ‘ouster clauses’, ‘preclusive
clauses’ and ‘finality clauses’.20 There are also various sub-species of these general
terms denoting the various linguistic formulae that legislatures have developed to
indicate their intention to limit or preclude judicial review. These include clauses
purporting to prevent the operation of the writs of ‘prohibition or certiorari’;
those stating that a class of administrative decision ‘shall not be questioned’;
those stating that certain administrative decisions provide ‘conclusive evidence’
on a particular question; and those stating that an administrative decision should
apply ‘as if it were enacted’.21 We explore the different approaches that have
been taken to different types of preclusory clauses in the following part. For
present purposes, however, it suffices to note that in the case of the most ‘extreme’
prescriptive legislation, the courts have generally treated the various provisions
Statutes imposing time limits on appeals or restricting the jurisdiction of courts
are not generally included within the scope of the term ‘privative clause’. Nev-
ertheless, it is acknowledged that, where a provision permits judicial review
only within a very short period of time following an administrative decision,
the practical effect of the provision is very similar to that of a privative clause.23
Interestingly, in some very contentious cases, the courts have treated time limit
provisions as if they were privative clauses.24

The privative clause and kryptonite: Jurisdictional error
Leaving to one side the devices used to exclude or limit judicial review, the other
constants in the discourse on privative clauses are the factors that lead courts
to rule such clauses to be ineffective. The touch-stone at this point is the notion
of ‘jurisdictional error’. These are errors of such gravity that the decisions made
or proposed to be made are regarded at law to be no decisions at all – nullities.
That is, where a jurisdictional error occurs in the making of an administrative
decision, a person with standing to challenge the ruling will be able to ‘pierce’ any
privative clause purporting to prevent judicial review of that class of decision.
This means that the person will have the legal right to seek judicial review of the
decision in the relevant court.
The problem, of course, is in defining a jurisdictional error. The variables are
many. Much will depend on who is making a decision and on the nature of the

error made. As a rule of thumb, administrative decision makers in the form of
bureaucrats or tribunal members will be given less lee way than judicial officers.
Within the court hierarchy, it will be easier to show a jurisdictional error in the
hands of an inferior court than in a superior court of record. This is due to the
fact that superior courts are empowered to make binding rulings about their own
As to the type of error made, distinctions are often drawn between errors
made at point of entry into a process and those made after an inquiry has been
validly commenced. Few privative clauses will hold against a decision maker who,
from the outset, has no legal authority to enter upon the inquiry in question.25
In most instances, however, the issue is not so clear cut. The decision maker
may have begun ‘within jurisdiction’. The question rather is whether he or she
does anything that takes the process into jurisdictional error. As explored in the
following section, privative clauses can prevent judicial review of some kinds
of legal errors. For example, a long line of jurisprudence suggests that where a
decision maker has jurisdiction to enter upon an inquiry, a privative clause will
prevent a reviewing court from intervening to correct subsequent errors unless
the errors go to jurisdiction and are apparent ‘on the face of the record’. Whether
an error amounts to a ‘jurisdictional error’ will again involve an examination of
who is making the alleged error and of their status relative to the reviewing court.
In the domain of federal administrative law, the critical point of reference is
Chapter III of the Constitution, which vests the ‘judicial power’ of the Common-
wealth exclusively in courts of law. This means that under the federal Constitu-
tion, administrative bodies cannot be vested with judicial power. Section 75(v)
of the Constitution provides further that the High Court has original jurisdiction
to hear cases relating to ‘all matters in which . . . a writ of Mandamus or prohibi-
tion or an injunction is sought against an officer of the Commonwealth’. As this
jurisdiction is conferred upon the High Court by the Constitution, the legislative
and executive branches of government cannot prevent the Court from hearing
cases in which these remedies will lie. Section 75(v) specifies that judicial review
must be sought in one of two ways. The first is by operation of the prerogative
writs (sometimes referred to as ‘constitutional writs’), while the second is via the
equitable remedies of declaration and injunction.
In Plaintiff S157/2002 v Commonwealth (Plaintiff S157),26 the High Court
ruled that if a decision made by an officer of the Commonwealth27 is infected with
jurisdictional error, the decision would not fall within the ambit of a privative
clause drafted in apparently express terms. In other words, the privative clause
notwithstanding, the decision will remain amenable to judicial review. Kerr and
Williams note that the case recognises, in Australia, a ‘minimum standard of
judicial review of executive action that cannot be abrogated by legislation’ which
is supported by the text of the Constitution, its structure and ‘the concepts and
principles that can be derived from the rule of law’.28
Both the prerogative writs and the remedy of declaration are discretionary.
Where the High Court is the only avenue left open to litigants, however, this

discretion is limited. The Court will have no option but to grant the appropriate
remedy where there has been a transgression, and where the applicants have
complied with the necessary procedures. It is in this context that privative clauses
and distinctions in types of legal error become relevant.
It remains an open question – dealt with in more detail later in this chapter –
whether the Commonwealth Parliament, or one of the state or territory parlia-
ments, could formulate a privative clause that would render even jurisdictional
errors impervious to judicial review. As we explore below, a critical difference
between the federal and state constitutions is that state constitutions contain no
equivalent to Chapter III of the Federal Constitution, meaning that the distinc-
tion between law and policy or questions of ‘merit’ is not as bright in the province
of state law. At state level, judicial powers can be vested in bodies other than
courts. There is also no equivalent of s75(v). By the same token, the effectiveness
of a privative clause will inevitably depend – at least to some extent – on the
rationale for its inclusion and on whether a court is persuaded of its legitimacy
in the circumstances of a particular case. While the desire to achieve certainty is
without doubt a motivator for the legislator in virtually every instance, there are
few privative clauses that are certain in their application.

Towards a general interpretative approach
to privative clauses

The courts in Australia and elsewhere tend to interpret privative clauses narrowly.
This means that privative clauses have not traditionally been invalidated, but their
impact has been diluted and in some cases almost completely nullified. The result
is that access to judicial review has been constrained, but not as much as might
be expected given the wording of the clauses in question.
Perhaps the first point to make is that privative clauses are designed as aids to
statutory interpretation. As privative clauses are themselves creatures of statute,
the traditional approach has been to examine their literal or ordinary meaning.29
Section 15AA of the Acts Interpretation Act 1901 (Cth) requires that provisions
also be construed purposively.30 This approach mandates an analysis of legislative
intent. Despite concerns about the difficulty of gauging such intent simply by
examining the record of parliamentary debates in Hansard,31 this remains an
orthodox mode of statutory interpretation.
The problem with privative clauses is that judges have often adopted construc-
tions that appear not to accord with ordinary meaning. For example, one might
well think that where an Act provides that a particular class of administrative
decision is ‘final and conclusive’ and ‘must not be challenged or reviewed by a
court’ and is not subject to the various constitutional writs, a court would need
to strain to construe the Act so as to permit judicial review of the relevant class of
decision.32 As Wade and Forsyth put it: ‘Statutory restrictions on judicial reme-
dies are given the narrowest possible construction, sometimes even against the
plain meaning of the words.’33

Perhaps the oldest of the approaches apparent in the jurisprudence on pri-
vative clauses is that which has seen the judiciary focussing on the language
used by the legislature so as to draw distinctions between different types of legal
errors. Privative clauses were said to bar judicial intervention to correct ‘non-
jurisdictional errors apparent on the face of the record’. In the 1760 case of R v
Moreley34 the Court of King’s Bench ruled that in the absence of strict words of
statutory intendment, the court will always be empowered to issue a writ of cer-
tiorari to quash a decision that displays an error of law. The statute in that case

No other court whatsoever shall intermeddle with any cause or causes of appeal upon
this act: but they shall be finally determined in the quarter sessions only . . . [and]
No record, warrant or mittimus to be made by virtue of this act, or any proceedings
thereupon, shall be reversed, avoided or any way impeached, by reason of any default
in form.

The Court agreed with the petitioner’s submission that the clauses merely pre-
vented the reviewing court from re-examining the facts of the matters in question.
The idea that the courts’ power to quash decisions (by certiorari) will only be
taken away by the ‘most clear and explicit’ words in a statute went on to become
accepted doctrine. For their part, successive judges pushed the boundaries of
language. So, for example, in R v Medical Appeals Tribunal; Ex parte Gilmore35
the English Court of Appeal, led by Lord Denning MR, ruled that statements to
the effect that a tribunal’s decisions were ‘final’ meant no more than final on the
facts of a case. That case involved a workman blinded in an industrial accident
who was awarded compensation only for the loss of sight in one eye – the other
having been sightless before the accident! The Court had no hesitation in finding
that the tribunal’s ruling involved a fundamental error of law. Romer LJ noted
that ‘it is not in the public interest that inferior tribunals of any kind should
be the ultimate arbiters on questions of law. Parliament, of course, can make
them so; but . . . a legislative intention to do so is not sufficiently expressed by the
mere provision that the decision of such-and-such a tribunal shall be “final”’.36
In many cases involving prescriptive legislation, the courts have distinguished
between issues of fact and issues of law;37 between errors going to jurisdiction
and non-jurisdictional errors;38 and between a right to appeal (on matters of
fact and law) and a right to judicial review.39 Lord Denning was sanguine in his
assessment of the jurisprudence, commenting that judges have tended to adopt
a doctrinal approach that will deliver the desired outcome in the case before
So, in Hockey v Yelland41 the High Court of Australia held that a privative
clause stating that the determination of a medical tribunal should be ‘final and
conclusive’ did not prevent judicial review of the legality of the decision made. In
the result, however, the Court found that the tribunal in question had committed
no legal error and that the applicant’s complaint went to matters of fact or perhaps
to errors of law that were not apparent on the face of the record. The Court
also made it difficult for the applicant to challenge the ruling on the basis of

non-jurisdictional error by adopting a very narrow interpretation of the phrase
‘on the face of the record’.42
In essence, there is an apparent disjuncture between what might be described
as the ‘literal’ meaning of a privative clause, and what we would term its ‘constitu-
tional’ meaning. We use the word ‘apparent’ advisedly: the dichotomy is arguably
a false one. As Professor Allan observes, in one ‘crucial sense’, statutes have no
‘literal’ meaning:

The application of a statutory provision is always a matter of legal and constitutional
argument: it entails an interpretation of the statute’s general purpose or policy sensitive
to those enduring legal values that are part of the language or communication between
legislature and judiciary, or between legislator and citizen.43

Having said this, the distinction between the literal and constitutional import of
a privative clause is a useful shorthand to denote how a privative clause might be
construed without regard to those ‘enduring legal values’ (the ‘literal’ meaning),
and how a privative clause should be interpreted when due regard is paid to all
relevant legal and constitutional considerations (the ‘constitutional’ meaning).44
The one constant has been that privative clauses complicate any judicial review
process. There has been argument over whether they decrease the scope of judi-
cial review or whether they increase the discretion given to an administrative
decision maker. Logically, if a statute operates to preclude an administrative
decision from being reviewed, this must increase (potentially ad infinitum)
the decision maker’s discretion or decision-making powers. However, the truth
of this statement is often denied.45 For example, under the Hickman doctrine,46
it has been argued that privative clauses are permissible provided their operation
could be ‘characterised not as restricting judicial review but rather as defining
the true reach of the decision maker’s power’.47 However, this approach carries
with it more than the faint odour of sophistry,48 as the High Court recognised in
Plaintiff S157.49
In the process of enacting the privative clause in s474 of the Migration Act 1958
(Cth) (see pp. 362–3 for further details), the Australian Parliament made plain
its intentions, even seeking to direct how the provision should be interpreted.
In the Second Reading Speech, the then-Immigration Minister, Philip Ruddock,
explained that the government intended the privative clause to apply to the High
Court as well as the inferior courts.50 Further, the Explanatory Memorandum
purported to express the legislature’s intention that the privative clause should
be interpreted consistently with a particular view of the ‘Hickman doctrine’ so as
to permit judicial review, but on three grounds only:

Such a clause has been interpreted by the High Court, in a line of authority stemming
from the judgment of Dixon J in R v Hickman; ex parte Fox and Clinton (1945) 70 CLR
598, to mean that a court can still review matters but the available grounds are confined
to exceeding constitutional limits, narrow jurisdictional error or mala fides.51

In oral argument in Plaintiff S157, the Solicitor-General asserted that the Court
was bound to follow this interpretative guide.52 He argued further that even if the

government’s interpretation of the Hickman doctrine were found to be incorrect
as a matter of law, this would not matter ‘because [the Minister] has said very
clearly . . . exactly what is intended’.53
The High Court rejected this argument, the majority observing that a ‘Minis-
ter’s understanding of the decision in Hickman cannot give s474 an effect that is
inconsistent with the terms of the Act as a whole’.54
Claiming that the narrow construction of a privative clause vindicates leg-
islative intent causes a number of problems for legislators. First, it can lead to
confusion. For example, during debates prior to the introduction of the privative
clause in the Migration Act, Senator Andrew Bartlett expressed his concern that
the Commonwealth Parliament was enacting a provision, the effect of which it
could not know:

If we are going to pass a bad law, at least we have to know how bad the damn thing
is before we pass it. The government do not even know what they did. They probably
wanted bad law; they might not have wanted very bad law. We probably got extraordi-
narily appalling law – and we do not even know if we have got it yet. We will not even
know for another year or two, until we finally get a High Court case about what the hell
this privative clause means.55

Secondly, it can cause frustration on the part of legislators. It can lead to a
game of cat-and-mouse wherein the courts, in effect, goad Parliament to for-
mulate new forms of privative clause which will be impervious to the courts’
meddling. The result is an escalation in hostilities and parliament will be encour-
aged to enact increasingly draconian privative clauses until the courts find them
constitutionally invalid.
This analysis leads to the conclusion that the vindication of legislative intent
is an unsatisfactory explanation for narrowly construing privative clauses. It is
also arguably unnecessary. Sir John Laws argues that, in reality, decisions that
construe privative clauses narrowly

. . . owe neither their existence nor their acceptance to the will of the legislature. They
have nothing to do with the intention of Parliament, save as a fig-leaf to cover their true
origins. We do not need the fig-leaf any more.56

Forsyth responds that ‘[t]he fig-leaf, like the swimming-costume on a crowded
beach’, is necessary to ‘preserve the decencies’.57 The question, therefore, is
whether it is indecent for the courts to construe privative clauses narrowly with-
out paying lip-service to the old justification based on legislative intent. It will be
our argument that a more satisfactory approach is to acknowledge that statutory
interpretation must always be conducted within the framework of constitutional
law and the broader context in which administrative decisions are made. As
explored earlier, there may well be circumstances in which it is reasonable and
appropriate to limit access to curial review. By the same token, it is not possible
to make legitimate distinctions between types of legal error (jurisdictional or
non-jurisdictional) without acknowledging the nature and significance of the
administrative decision sought to be reviewed. Another interesting question is

whether there are legitimate distinctions to be found between the powers of the
federal courts to review decisions (given the terms of the Federal Constitution)
and those of superior state courts of record.

Privative clauses and the importance of context

The operation of a privative clause, like any statutory provision, is dependent on
how it is interpreted. The foregoing discussion reveals that different privative
clauses imply differing degrees of restriction on judicial review. One feature that
emerges from the case law is that such restrictions appear to depend only partly
on the wording of particular clauses. Context is highly relevant in the courts’
interpretation of privative clause. The starting point must be the constitutional
requirements of the jurisdiction in which a privative clause is enacted. The case
law makes clear that a privative clause enacted by a state is subject to fewer lim-
itations on its operation than one enacted by the Commonwealth Parliament.58
As noted earlier, this is largely due to the operation of s75 of the Federal Con-
stitution, which constitutionally enshrines the High Court’s role in conducting
judicial review of administrative action.
It is also clear that distinctions are drawn between administrative and judicial
bodies and between inferior and superior courts of record. Australia has never
adopted the broadly interventionist approach favoured by the English courts
(where the tendency has been to treat all errors of law as going to jurisdiction).59
However, the High Court has generally made it plain that tribunals and admin-
istrative bodies cannot be protected from judicial oversight where they fall into
legal error.60
Other contextual factors are also relevant. It seems that Australian courts
have been influenced – consciously or sub-consciously – by the type of decision
to which the privative clause applies. So, for example, more deference appears
to be shown in the case of tribunals and fact finding bodies vested with special
expertise in a particular area or matter. In such contexts, as noted earlier, it is not
uncommon for the courts to give with one hand while taking away with the other.
In some cases they have reaffirmed their power to intervene, but then drawn back
from finding any legal error.61
In this section we explore two areas of law in which legislatures have attempted
repeatedly to restrict access to judicial review. The first is that of industrial rela-
tions. This is interesting for the depth of history in attempts to shut out the courts;
and for the complex interplay between state and federal powers. The second case
study is immigration, interesting because of the battle royal that has raged in that
area between parliament and the courts.62

Privative clauses in industrial relations
There is a long history of privative clauses being used in the industrial relations
field in Australia. These go back to the early years of the twentieth century when

the Federal Government and several of the state governments established labour
courts to settle industrial disputes by conciliation and, where conciliation failed,
by final and binding compulsory interest arbitration.63 The purpose of the pri-
vative clauses was to limit judicial review of the decisions of specialist labour
courts. The view of Parliament was that it was appropriate for arbitrated set-
tlements, which bestowed terms and conditions of employment on the relevant
industry and occupation, to be undisturbed from judicial review unless a court
had manifestly exceeded its jurisdiction.
One noteworthy feature of the early labour cases is that the decision-making
authorities at first instance were courts – one fact that may explain the early
deference shown to rulings in this area. In more recent times, industrial relations
arbitration has become the preserve – in the main – of industrial relations tri-
bunals. Although the courts’ treatment of preclusive provisions in this area has
varied over the years, the move does appear to have engendered a less deferential
approach to commission rulings. The commissions – with the exception of the
New South Wales Commission in court session64 – tend now to be regarded as
no more than specialist tribunals with narrow fields of expertise.
During the first half of the twentieth century, the state Supreme Courts and
the High Court paid deference to the state labour courts. As recently as 1960, the
High Court confirmed that a privative clause inserted in the Industrial Arbitration
Act 1912–1952 (WA)65 had the effect of limiting the ability of superior courts to
review the rulings made by the state Court of Arbitration. The case involved
a mining dispute in which a union obtained an ex parte order against a mining
company prohibiting the company from dismissing its employees or engaging in a
‘lock out’ at the coal mine in question. The company had threatened its employees
that unless a drop in productivity could be reversed, it would be forced to close the
mine. Instead of fighting the matter in the Arbitration Court, the company took
its grievance straight to the Supreme Court of Western Australia (and thence to
the Full Court of the Supreme Court). The Court obliged the company by issuing
writs of certiorari and prohibition on the basis that the Arbitration Court had
misconstrued the terms of the relevant legislation.
Leading the High Court in overturning the Supreme Court’s ruling, Dixon CJ
was at pains to emphasise the difference between privative clauses operating in
the domain of state law. He said:

It might be thought that the exclusion of certiorari expressed by this provision would
afford an insuperable objection to the use of that remedy to quash an order of the
Arbitration Court; but reliance is placed upon the restrictive construction placed upon
similar provisions in decided cases which say that they do not operate to remove the
remedy where the subject matter is outside the scope of the authority of the inferior
In this Court the fact that s75 (v) of the Constitution invests jurisdiction in matters in
which a writ of prohibition is sought against an officer of the Commonwealth has neces-
sarily affected the interpretation of similar clauses in Commonwealth legislation. Such
a provision cannot deny the remedy where it properly lies. But in relation to statutory

as distinguished from constitutional limitations, restrictions or restraints on the author-
ity of a federal tribunal, the provision may be taken into account in ascertaining what
the apparent restriction or restraint actually signifies when it is necessary to determine
whether the situation is one in which prohibition properly lies.66

In the result, Dixon CJ and the rest of the Court noted that the result was
‘unfortunate’ insofar as the impact was to entrench an order made by the Arbi-
tration Court that was only ever intended to be a stop-gap measure. However,
the High Court gave effect to the privative clause by holding that the dispute
was the province of the Arbitration Court and ‘in the Supreme Court no other
question could be considered except invalidity on the ground of complete lack
of jurisdiction falling for that reason or otherwise outside the protection of
At the federal level, where for most of the twentieth century the only review-
ing court was the High Court, that court scrutinised more closely the federal
Labour Court and Commission. As noted by Dixon CJ above, the High Court was
consonant of its powers under s75(v) of the Constitution. It was in the federal
jurisdiction that Dixon J (as he then was) enunciated the famous bona fides test in
R v Hickman; Ex parte Fox and Clinton.68 As noted earlier, his Honour stated that
privative clauses that purport to exclude even certiorari can validly restrict the
scope for judicial intervention provided that three criteria are met. The protected
decision must constitute a bona fide attempt to exercise the power conferred on
the decision maker; it must relate to the subject matter of the legislation; and it
must be reasonably capable of reference to the power given to the body.69 It is
noteworthy, however, that in spite of this deferential dictum, the High Court did
intervene in Hickman’s case, ruling that the Commission order in that instance
was invalidated by jurisdictional error.
In more recent years, controversy has arisen again over the use of privative
clauses in state (as distinct from federal) industrial relations legislation. The
most important such privative clause is s179 of the Industrial Relations Act 1996
(NSW). It provides:

(1) A decision of the Commission (however constituted) is final and may not be
appealed against, reviewed, quashed or called into question by any court or
(2) Proceedings of the Commission (however constituted) may not be prevented from
being brought, prevented from being continued, terminated or called into question
by any court or tribunal.
(3) This section extends to proceedings brought in a court or tribunal in respect of a
decision or proceedings of the Commission on an issue of fact or law.
(4) This section extends to proceedings brought in a court or tribunal in respect of
a purported decision of the Commission on an issue of the jurisdiction of the
Commission, but does not extend to any such purported decision of:
a. the Full Bench of the Commission in Court Session, or
b. the Commission in Court Session if the Full Bench refuses to give leave to appeal
the decision.

(5) This section extends to proceedings brought in a court or tribunal for any relief or
remedy, whether by order in the nature of prohibition, certiorari or mandamus,
by injunction or declaration or otherwise.

Thus, s179 of the Industrial Relations Act 1996 (NSW) not only protects decisions
from actual or purported jurisdictional error on the part of the Industrial Relations
Commission. It also covers the Industrial Relations Commission in court session.
Again, this body is a superior court of record whose judges have the same status
as judges of the Supreme Court of New South Wales. Interestingly, jurisdiction
is bestowed upon the Industrial Relations Commission in Court session to try
defendants for criminal breaches of the Occupational Health and Safety Act 2000
(NSW). In most instances there is no right of appeal to the superior courts.
However, the privative clause means that aggrieved litigants are unable to seek
judicial review of these criminal proceedings. The statute confers only limited
rights to appeal such cases to the New South Wales Court of Appeal, but the
privative clause prevents the judicial review of decisions. This has caused some
disquiet in the community.70
This provision was considered by the High Court in Fish v Solution 6 Holdings
Limited.71 This is an extraordinary case that is very difficult to understand unless
considered within the very particular context of state-judicial politics in New
South Wales. The case was one in which the plaintiffs attempted to bring an action
in the New South Wales Industrial Relations Commission for orders relating to an
allegedly unfair contract of employment and seeking a declaration that a related
share purchase agreement was unfair, harsh and unconscionable. The defendants
fought the action by seeking prohibition in the New South Wales Court of Appeal
to stop the Commission from hearing the case at all on the basis that it did
not have the jurisdiction to entertain an application in a matter that involved a
commercial dispute. (The action had to be brought in the Court of Appeal because
the Commission in court session has the status of the Supreme Court.) The New
South Wales Court of Appeal found unanimously that the privative clause in s179
did not apply where, as in this case, no ‘decision or purported decision’ of the
Commission had yet been made.72 Extraordinarily, the Court of Appeal proceeded
to grant prohibition on the basis that the Commission had no jurisdiction to rule
on matters of commercial disputation. It made the ruling whilst acknowledging
that the privative clause in question would have prevented judicial review of the
decision had the Commission entered upon the inquiry in question. The Court
acknowledged that the Commission would have been within its jurisdiction in
deciding to hear the case because the subject matter related in part to an unfair
contract of employment.
The High Court, by majority, dismissed the appeal.73 The majority held that
the privative clause did not apply in this matter because they held that the Com-
mission did not have jurisdiction to hear commercial disputes and the application
to the New South Wales Court of Appeal was instituted before the Commission
had had the opportunity to conduct a hearing or make a decision.74 Kirby J, in

dissent, argued that the Court of Appeal’s approach was inconsistent with a pre-
vious line of authority that prohibition would be refused unless the jurisdictional
objection had first been advanced and determined before the Commission.75
Heydon J also registered a spirited dissent. His Honour argues persuasively that
the majority ruling sits uneasily with precedent suggesting that a superior court
of record should always be called upon to make its own rulings on matters going
to jurisdiction before the intervention of a judicial body of similar or superior
Perhaps the most surprising aspect of the majority ruling is the underlying
presumption that the privative clause in question would have protected the pro-
ceedings from judicial review had the Commission proceeded to hear the case.
The High Court upholds the apparent force of the privative clause, but then per-
mits the complete subversion of the legislative measure by allowing prohibition
to issue in the form of a pre-emptive strike! It is an approach that finds more than
a few resonances with that adopted in the context of another privative clause,
this time enacted in the domain of federal administrative law (for further details
see pp. 361–3).
More generally, the joint majority judgment in Fish noted that there were a
number of presumptions that operate to urge a narrow interpretation of privative
clauses enacted in state legislation, including that ‘it must . . . be presumed that
a State parliament does not intend to cut down the jurisdiction of the Supreme
Court of that State over matters of a kind ordinarily dealt with by the State
Supreme Courts and which, if dealt with by those Courts, are amenable to the
appellate jurisdiction of [the High] Court under s73 of the Constitution’.77 On
the broader question of the effect of Parliament including ‘purported’ decisions
within the ambit of the privative clause, Kirby J said (in dissent):

The statutory inclusion of reference to a ‘purported decision’ could not protect from
supervisory orders of the highest court of the State action by the Commission that
did not reach the fundamental requirements contemplated by Parliament in protecting
‘decisions’ and also ‘purported decisions’.78

With respect, this approach seems to be one that is more consistent with tradi-
tional understandings of the role of superior courts in overseeing the operation
or application of the rule of law.

Privative clauses in immigration and refugee law
Industrial relations aside, the most extensive debates about the respective roles
of the courts and the executive arms of government have occurred in cases involv-
ing immigrants and asylum seekers. The conflict between the executive and the
judiciary dates back at least to the mid 1980s, when the newly-established Fed-
eral Court began to explore the supervisory powers vested in it by the Adminis-
trative Decisions (Judicial Review Act) 1977 (Cth).79 Faced with unprecedented
incursions into the previously closed world of immigration decision-making,80

successive federal governments have attempted to reassert the control of the
executive. Changes to law and policy in the immigration field include some of
the most extreme examples of privative clauses ever created in Australia. Indeed,
it is fair to say that every device for restricting the power of the courts (and
for maximising the power of the executive) has been trialled in this field. If the
migration cases provide the starkest examples of courts struggling to ‘read down’
privative clauses, this is undoubtedly a reflection of the fact that contentious
migration cases almost invariably involve serious issues of human rights. By the
same token, the uncertain status of many migrants and the recent exponential
increase in the federal courts’ migration case load have combined to generate
reluctance in the courts to deny governments the policy outcomes they desire.
Many of the victories won by migrants and refugee claimants have been pyrrhic.81
In the 1980s, the open-ended nature of the migration legislation, which vested
sweeping powers in administrators, was found to make decision makers particu-
larly susceptible to judicial review. The courts used the legal principles of procedu-
ral fairness or natural justice and notions of legal relevance and reasonableness to
overturn decisions and question the way in which decision makers made their rul-
ings. Perhaps the most dramatic response to the blossoming of the ‘new adminis-
trative law’ in the immigration field was the decision in 1989 to ‘codify’ migration
decision-making by replacing the broad discretions with closely circumscribed
regulations. While the Minister for Immigration has always retained the power
to intervene, ordinary decision makers – and any bodies reviewing their rulings –
have been subjected to tighter and tighter controls. In fact, all sorts of devices
have been employed to bolster the power of the minister. These include the
personalisation of decision-making with the insertion of various clauses that
turn on subjective factors such as the ‘satisfaction’ or ‘opinion’ of the minister,82
and the creation of what are known as ‘non-compellable, non-reviewable
The first attempt to constrain expressly the supervisory power of the courts
occurred in 1992 in response to judicial rulings that challenged governmental
policies aimed at the mandatory detention of certain unlawful non-citizens. The
original targets of the laws were ‘boat people’ from Cambodia. A section that
purported to prohibit any court from ordering the release of such persons was
struck down by the High Court. The Court held that the effect of the provision
was to place the decision to detain solely in the hands of the administration – to
the point that it operated to usurp the judicial power vested by the Constitution
exclusively in Australia’s federal courts.84
The controversy surrounding the Cambodian (and other) boat people led in
due course to the amendment of the Migration Act 1958 (Cth) to create a special
system for the judicial review of migration decisions in Part 8 of the Act. The first
Part 8 of that Act came into force on 1 September 1994, denying migrants access
in the Federal Court to either the Administrative Decisions (Judicial Review) Act
1977 or the Judiciary Act 1903 (Cth). The curious regime then created spelled
out the grounds on which the Federal Court could review migration decisions,

carefully excluding the broader grounds of procedural fairness, relevancy and
reasonableness that were thought to be giving the Federal Court too great a
licence to intervene in migration cases.85
The first Part 8 was designed specifically to limit what was perceived to be
abuse of the judicial review process by unmeritorious applicants.86 However,
despite these legislative measures, the number of migration appeals continued to
mount. Cases brought before the High Court in its original jurisdiction increased
as plaintiffs sought alternative avenues of redress to compensate for the reduced
grounds of review available at the Federal Court.87 More surprisingly, the Fed-
eral Court also experienced an increase in applications for review, despite the
restrictive legislative provisions.
In Abebe v Commonwealth,88 the applicant argued that the provisions
amounted to an unconstitutional attempt to constrain the powers of the Fed-
eral Court. Ms Abebe also sought prerogative relief under section 75(v) of the
Constitution, on the ground that the tribunal’s decision in her case was unlawful
by reason of unreasonableness. By a narrow majority of four to three,89 the High
Court ruled that Parliament could legislate to prevent the Federal Court from
reviewing part of a legal ‘matter’, confining its jurisdiction to deal with only parts
of a legal problem.90 The majority of the Court further held that it was within
Parliament’s ability to narrow the exercise of judicial power by the Federal Court
through Part 8 of the Migration Act, and to restrict those decisions available for
judicial review.
Despite the judicial deference evidenced in Abebe, there appeared still to be
significant scope for judicial review of migration decisions. Accordingly, the gov-
ernment repealed its first Part 8 experiment, replacing it with a privative clause
regime modelled on the legislation at the heart of the Hickman case. Section 474
of the Migration Act 1958 (Cth) states:

A privative clause decision:
a. is final and conclusive, and
b. shall not be challenged, appealed against, reviewed, quashed or called into question
in any court; and
c. is not subject to prohibition, mandamus, injunction, declaration or certiorari in any
court on any account.

A ‘privative clause decision’ is defined in s474(2) as a decision of an administra-
tive character made under the Act or the Regulations. Most decisions relating to
migration were intended to be privative clause decisions. The intent of the legis-
lation was to exclude review not only by the Federal Court, but also by the High
Court, notwithstanding its Constitutionally-protected review powers. Although
the government cannot entirely oust the jurisdiction of the High Court,91 it can
signal its preference that the Court not intervene in certain classes of dispute.
It is clear from the construction of the privative clause that the government had
believed it would be interpreted in light of the Hickman principle, as it understood
that principle to operate.92 The government probably had good reason to be

confident in its new privative clause scheme, having regard to the High Court’s
deferential approach to the former Part 8.
The High Court’s interpretation of the new Part 8 in Plaintiff S157 is a curious
mix of deference and assertion. While holding that the privative clause regime
was indeed constitutional, the Court nonetheless stated that any tribunal decision
evidencing jurisdictional error would fall outside the privative clause scheme and
therefore be open to review by either the Federal or High Courts.93 The Court
stated firmly a failure to exercise jurisdiction or an excess of the jurisdiction
conferred by the Act is ‘regarded, in law, as no decision at all . . . Thus, if . . . the
question cannot properly be described in the terms used in s474(2) as “a decision
made under this Act” . . . [it] is, thus, not a “privative clause decision” as defined
in ss474(2) and (3) of the Act’.94
The effect of this pronouncement is to expand the concept of jurisdictional
error and to hold that such an error vitiates a decision.
The deferential strain in the judgment is illustrated by the fact that the privative
clause scheme was held to be constitutional. Therefore, Part 8 and the privative
clause remain in place in the Migration Act, though the High Court has severely
reduced their applicability.95 Subsequent decisions by that Court suggest that a
particularly hard line is being taken against decisions in which there has either
been a departure from the strict terms of the legislation and/or a failure to follow
fair procedures.96

The future of privative clauses?

The desire in governments to make administrative decisions genuinely impervi-
ous to the meddling of courts has not diminished. Although never enacted,97 the
United Kingdom Parliament was asked to consider a clause in 2003 that sought
to extend the prohibition on judicial review to:

. . . prevent a court, in particular, from entertaining proceedings to determine whether
a purported determination, decision or action of the Tribunal was a nullity by reason of–
(i) lack of jurisdiction,
(ii) irregularity,
(iii) error of law,
(iv) breach of natural justice, or
any other matter . . . 98

Critically, the proposed subsection provided that the privative clause would
cover not just administrative decisions or conduct, but also decisions or conduct
affected by fundamental legal error – hence the reference to ‘purported decisions’.
This was an entirely new development and caused great consternation because
courts in the United Kingdom (and, it might be interpolated, in Australia) have
traditionally held that a decision that is infected by jurisdictional error is, in
reality, no decision at all. The result is that a privative clause designed to immunise

‘decisions’ from judicial review does not apply to a failed attempt to make a
decision.99 In light of this, Professor Jowell observed that ‘[t]hose who drafted
[the privative clause] must have studied Anisminic and other cases very carefully,
as there is no room to doubt that this ouster clause is judge-proof’.100 In practical
terms, therefore, this privative clause (if interpreted literally) would completely
immunise the specified class of administrative action (except on the very narrow
grounds provided, viz. bad faith).
Shortly after this proposal was made in the United Kingdom, a new Bill came
before the Commonwealth Parliament, proposing a similar amendment to the
privative clause in s474 of the Migration Act 1958 (Cth).101 The Australian Bill
proposed to expand the definition of ‘privative clause’ to include:
(b) a purported decision that would be a privative clause decision within the meaning
of subsection 474(2) if there had not been:
(i) a failure to exercise jurisdiction; or
(ii) an excess of jurisdiction;
in the making of the purported decision.102

The key term is, again, ‘purported decision’. The Explanatory Memorandum
stated: ‘A “purported decision” is a decision that would be a privative clause
decision, had it not been affected by jurisdictional error.’103 If accepted, this
would clearly expand the ambit of administrative action falling within the scope
of the privative clause. The accompanying explanatory material also sought to
make clear that this amendment was aimed at closing the perceived loop-hole
identified in Plaintiff S157,104 by bringing jurisdictional errors within the scope
of the privative clause.105 This Bill was never fully debated, nor was it put to a
vote. However, some aspects were included in the Migration Litigation Reform
Act 2005 (Cth), which came into force on 1 December 2005. The references to
purported decisions remain in the context of denying jurisdiction to the inferior
federal courts.
Why is this new form of privative clause, which seeks to bring within its ambit
‘purported’ decisions, so significant? Such provisions are a manifestation of a new
type of privative clause that parliaments are starting to consider in order to make
particular classes of administrative decision ‘judge-proof ’. It is more draconian
than anything previously proposed, in the sense that it seems intended to hive
off entire categories of administrative action from judicial supervision in respect
of almost every conceivable administrative error.
Had the United Kingdom privative clause been enacted, it would, almost cer-
tainly, have caused a constitutional crisis. Lord Woolf (as he then was) described
this provision as ‘so inconsistent with the spirit of mutual respect between the
different arms of government’, and such an affront to the rule of law, that the
courts may have found it ineffective in achieving its aims.106 In making this state-
ment, Lord Woolf seemed to be alluding to the United Kingdom courts asserting
a Marbury v Madison107 type of power to strike down legislation. Legal orthodoxy
has always posited that no such power exists in the United Kingdom courts.108

Of course, the situation in Australia has always been different: Section 75 of
the Commonwealth Constitution clearly gives the High Court power to invalidate
legislation that is found to be inconsistent with the Constitution itself. What
Australian administrative lawyers can learn from the debate in the UK relates to
how the High Court might be likely to approach this nascent breed of privative
It appears that this new type of privative clause is intended to force the courts
to confront the will of Parliament more directly. As we stated earlier in this chap-
ter, one of the central justifications used by courts for their narrow interpretation
of privative clauses has been that they are vindicating the will of Parliament. By
explicitly stating that ‘purported decisions’ are also included within the ambit of
the relevant privative clause, it would be very difficult for the courts to state –
as they have on numerous occasions – that an administrative decision infected
with jurisdictional error is no real decision at all and, therefore, such an admin-
istrative decision does not fall within the scope of the privative clause. To take
that approach, courts would need to depart even further from the natural and
ordinary meaning of the privative clause itself.
One method of dealing with such a privative clause is to avoid the privative
clause in the manner encouraged by the New South Wales Court of Appeal (and
endorsed by the majority of the High Court) in the Solution 6 Holdings case.109 In
that case, the Court of Appeal granted relief in the form of prohibition, finding
that the privative clause (which included ‘purported’ decisions) did not operate
prior to a decision being made. This ‘pre-emptive strike’ approach neatly circum-
vents the privative clause. However, such a course of action would presumably
only be available before the relevant tribunal has had the opportunity to reach
its decision. In the vast majority of cases for which judicial review is sought, this
would be of little assistance.
We would argue, therefore, that the High Court must eventually confront such
a privative clause head on. When this occurs, it will face a stark choice: either it
must give the privative clause its full application, thereby enfeebling the Court’s
own role and its constitutional position; or it must invoke some higher principle
to refuse to give full force to the terms of the privative clause. Four fundamental
points need to be made here.


ńňđ. 14
(âńĺăî 19)