Separation of Powers and the Judicial Branch

Cheryl Saunders

“A society in which the observance of the law is not assured, nor the separation of powers defined, has no constitution at all.”1

1. Introduction

Much of the debate in this country over the changes to the office of Lord

Chancellor and the creation of a Supreme Court was cast in terms of separation of

powers; although more cautiously on the part of the government, at least in those

terms.2

The purpose of this paper is to explore the meaning of the doctrine of separation

of powers in its application to the judiciary in countries in the British or (perhaps

more suitably) the Commonwealth constitutional tradition, with particular reference to

questions of a broadly constitutional kind that many such countries are facing. It is

thus intended to be comparative, rather than directed specifically to the circumstances

of the United Kingdom, although the topic clearly is prompted by developments, here.

The comparison might have been less productive before these changes were set in

train. The institutional arrangements for the protection of judicial independence and

the rule of law in most Commonwealth countries are broadly similar to the

arrangements that were broadly assumed to be in place in the United Kingdom, or at

least in England and Wales. Relevantly, in most Commonwealth countries they are

justified by a doctrine of separation of powers; often as a constitutional requirement.

In fact, however, the arrangements in this country were quite different in significant

respects. This does not mean that they were worse. It helps to explain, however, why

Goodhart Professor of Legal Science 2005-2006, University of Cambridge; Professor of Law, University of Melbourne1 Declaration of the Rights of Man and of the Citizen, Article 16, http://www.hrcr.org/docs/frenchdec.html 2 See for example, Judges’ Council, Response to the Consultation Papers on Constitutional Reform http://www.dca.gov.uk/judicial/pdfs/jcresp.pdf : “One of the reasons given by the government for the abolition of the Office of Lord Chancellor and the creation of a Supreme Court is to increase the separation of powers”, at [33]. See also Andrew Le Sueur, “Judicial Power in the Changing Constitution”, in Jeffrey Jowell and Dawn Oliver (eds) The Changing Constitution 5th ed, 2004, 323, 336

- 1 -

the doctrine of separation of powers has had here what Trevor Allan rightly describes

as “unsympathetic treatment”.3 It also explains why the changes that are in train align

the position of the judiciary in this country more closely with countries elsewhere,

removing some obstacles to effective comparison and bringing the United Kingdom

under what I will argue is a distinctive separation of powers umbrella.

2. Separation of powers in the British constitutional tradition

The concept of separation of powers, as used in western constitutionalism,4

assumes that there are three categories of public power, legislative, executive and

judicial, that are vested in three distinct groups of institutions, similarly described.5

There are notorious difficulties with the concept, which once led Geoffrey

Marshall to dismiss it as “infected with so much imprecision and inconsistency that it

may be counted little more than a jumbled portmanteau of arguments for policies

which ought to be supported or rejected on other grounds”.6 For present purposes the

difficulties may be categorised in the following way.

First, every constitutional system that purports to be based on a separation of

powers in fact provides, deliberately, for a system of checks and balances under

which each institution impinges upon another and in turn is impinged upon. There is

an open question whether this qualifies a system of separation of powers7 or

completes it “by giving each department the necessary constitutional means…to resist

3 Trevor Allan, Law, Liberty and Justice, Clarendon Press, Oxford, 1993, 50; see also Eric Barendt “Separation of Powers and Constitutional Government” [1995] Public Law 599, to much the same effect.4 For a general introduction see John Morrow, History of Western Political Thought Palgrave, 2nd ed 2005, chapter 95 The definition here is deliberately cast in general terms, for reasons that will become obvious. Some, relatively standard definitions are more prescriptive, however. Barendt cites as the “classic formulation…three distinct functions of government…which should be discharged by three separate agencies…and that no individual should be a member of more than one of them.”6 Geoffrey Marshall, Constitutional Theory Clarendon Press, Oxford, 1971, 1247 It might qualify a separation of powers institutionally (one institution affects the operations of another) or functionally (in doing so, the institution exercises a part of the power of the other): see, on this distinction, Marshall, op.cit., 102. Marshall also notes that the notion of checks and balances comes from the theories of mixed government.

- 2 -

encroachment by the others”.8 The answer must depend in part on the actual checks

and balances in each case.

Secondly, there is some disagreement about what is separated pursuant to the

doctrine. The two principal options are:

A separation of institutions; which might, for example, preclude overlapping

membership

A separation of functions, empowering each institution to exercise the

function for which it is designed (and perhaps, by extension, no other

function)

In reality, however, these are not mutually exclusive options. Any system of

separation of powers must involve at least a measure of both.

And this in turn leads to the third area of uncertainty: the degree of separation

that such a doctrine requires. Complete separation is impossible. There must be a

point, however, at which partial separation is not worthy of the name. Where, then,

should the lines be drawn? And how should the lines that are drawn be enforced?

It is possible to approach the answers to these questions through a consideration of

the purposes of a separation of powers. Received wisdom has it that there are two

such purposes:

One is to prevent the abuse of public power through the concentration of

power. Thus James Madison in Federalist 47: “The accumulation of all

powers, legislative, executive and judiciary, in the same hands, whether of

one, a few, or many, and whether hereditary, self-appointed or elective, may

justly be pronounced the very definition of tyranny”.

The other is to enhance the efficiency of government. In its original form this

rationale was used to justify the institution of a strong presidency in the United

States,9 but it might be turned to more general purposes. It might be argued,

for example, that separation of powers in this respect recognises that each of

8 James Madison, Federalist Paper 51 in Alexander Hamilton, James Madison and John Jay, The Federalist, Everyman Library, London, 265, 2669 Geoffrey R Stone, Louis M. Seidman, Cass R. Sunstein, Mark V. Tushnet, Constitutional Law 2nd ed 1991, 363

- 3 -

the branches is peculiarly well equipped to exercise the particular functions

assigned to it. Alternatively, it might be argued that the collective effect of a

particular configuration of separated powers is a form of government with

attributes best suited to the needs of a particular state.

Both of the two purposes traditionally assigned to a separation of powers were

originally formulated by reference to the separation of legislative and executive

power; indeed this was also the genesis of the theory in the writings of Montesquieu

and Locke. Both purposes could be adapted to the separation of the judicial branch.

As a generalisation, however, the doctrine of separation of powers has not been well

developed in relation to the judiciary, at least in a way that is useful here. The

explanation lies in the distinctive constitutional arrangements of both the two states

that have been the flagships of the separation of powers doctrine.

One is the United States, in which the relationship between the legislature and

the executive as part of a balanced three-way separation of powers dominates

the discourse and influences the doctrine;10 incidentally suggesting caution in

borrowing conceptions of “deference” from that source.

The other is France, where the separation of powers places emphasis on the

judiciary; but with a view to precluding it from impinging upon the other

branches of government.11

By contrast, the constitutional arrangements in the United Kingdom are quite

different, in key respects. Unlike the United States, this country has a

parliamentary system with at best a weak institutional separation of the

legislative and executive branches. Unlike France, judicial review of executive

action, at the hands of the general courts, is a core feature of the constitutional

system.

For reasons that are easy to understand, moreover, the doctrine of the separation of

powers has had only muted and partial recognition in the United Kingdom.12There is

10 See for example, Stone et al, op.cit, where even the examination of the significance of the doctrine for the rule of law is undertaken by reference to the relationship between the legislature and the executive. 11 Marshall, op.cit. 99. Marshall also quotes Hans Kelsen: “the judicial review of legislation is an obvious encroachment upon the principle of separation of powers”, at 99, from Hans Kelsen, General Theory of Law and State trans. A. Wedberg (1961), 26912 Paul Craig has said that it is “regarded as a central construct in our constitutional structure”; in a context, however, that is confined to the scope of judicial power vis-à-vis the other branches: Paul

- 4 -

of course a strong tradition of judicial independence, given teeth through a range of

guarantees enshrined in statute and long-standing constitutional convention.13 There is

a long tradition of limited government, involving the division of power in various

ways, through the “mixed” and “balanced” constitutions.14It is even possible to argue

for a significant degree of functional separation of powers in reality, if not in outward

form.15 And the great William Blackstone himself adapted Montesquieu and Locke to

argue for a “distinct and separate existence of the judicial power in a peculiar body of

men…separated from both the legislative and the executive power”.16

But the intermixture of institutions is such that it is almost impossible to

describe it as a separation of powers, without becoming entangled in endless,

unproductive debate. This is most obviously the case with the relationship between

the legislature and the executive, famously claimed by Walter Bagehot as a fusion of

powers. For present purposes, however, the triple hats of the Lord Chancellor and the

dual functions of the House of Lords are a further obstacle to describing the system as

characterised by a separation of powers. In an additional twist, it is even possible to

argue that these mixed institutions have been key mechanisms for protecting judicial

independence. Clearly, such a case can be made for the multiple functions of the Lord

Chancellor. In a speech in this university just over 2 years ago Lord Woolf observed

that it had not been “appreciated sufficiently that the Lord Chancellor played a pivotal

role in co-ordinating the three arms of government” nor that he “was able to act as a

lightening conductor at times of high tension between the executive and the

judiciary”.17 In time it may also come to be appreciated that a similar case could have

been made for the Appellate Committee of the House of Lords.

Countries with a system of government most similar to that of the UK are, not

surprisingly, found in the Commonwealth. Many of them have parliamentary systems.

These include Australia, Canada, India and South Africa, which I use as occasional

Craig, “Fundamental Principles of Administrative Law”, in David Feldman (ed) Public Law OUP 2004, 689, 70613 For detailed description and analysis see A.W. Bradley, “The Constitutional Position of the Judiciary” in David Feldman (ed) Public Law OUP 2004, 333, 339-36314 Marshall, op.cit., 101. 15 Barendt, 615. Trevor Allan has argued also for a de facto institutional separation of powers that relies on the “firm conventions governing…[the] various roles” of public officers: Trevor Allan, op.cit 52 16 William Blackstone, The English Constitution 7th ed, 1984, 10, quoted in Allan, op.cit., 5017 Lord Woolf, “The Rule of Law and a Change in the Constitution”, the Squire Centenary Lecture, 3 March 2004, http://www.dca.gov.uk/judicial/speeches/lcj030304.htm

- 5 -

examples in the next part. All four have a judiciary that reviews executive action and,

for that matter, legislation as well. All place great store on judicial independence. In

all of them, however, the judiciary is institutionally and to a large extent functionally

separate from the other branches of government. And in all of them, with the possible,

but by no means certain exception of Canada, the relationship of the judiciary to the

other branches is conceived in terms of the separation of powers. Traditional

mechanisms for protecting judicial independence, of which guarantees of tenure are

an example, can be understood as features of the separation of judicial power. The

points of necessary interface with the other branches in relation to, for example, the

appointment and removal of judges and the funding of courts can be analysed within a

framework of checks and balances. This is, however, a latter-day rationalisation. In

historical reality, most of the arrangements in these countries for the constitution of

the judiciary were modelled as far as possible on comparable arrangements here.

The limits of the possible were reached with the position of Lord Chancellor

and the institution of the Appellate Committee of the House of Lords,18 which could

not be replicated elsewhere. The transformation of the first and the pending

replacement of the second by a Supreme Court thus bring Britain closer to

Commonwealth experience in this regard.19 There is considerable interest throughout

the Commonwealth in what has been considered necessary here to provide protection

for judicial independence, once those two institutions are gone. Changes that attract

particular attention include the creation of a position of President of the courts of

England and Wales, with representational and management responsibilities;20 a

Judicial Appointments Commission; statutory protection of the principle of judicial

independence; and the use of a “concordat” to record aspects of the arrangement less

susceptible to codification, including the structures for administrative support.21

18 For the role played by the Lord Chancellor in relation to judicial appointments see Sir Thomas Legg, “Judges for the New Century” [2001] Public Law 62. See in particular the following: “it is important to understand the real nature of the present system. Otherwise, it might stand in the catalogue as simply appointment by the executive. Appointment is indeed on the advice and personal decision of a government minister. But he is a very unusual kind of minister….”: at 65.19 For a detailed exposition of the process by which the changes were made see Lord Windlesham, “The Constitutional Reform Act 2005” Parts 1 and 2, [2005] Public Law 806; [2006] Public Law 3520 Constitutional Reform Act 2005 section 721 The Concordat, January 2004, http://www.dca.gov.uk/consult/lcoffice/judiciary.htm

- 6 -

Many of the problems that have been anticipated in the course of negotiations

over the Constitutional Reform Act also currently are problems elsewhere. In

consequence, there will also be Commonwealth interest in the effectiveness of the

solutions that have been put in place. And the converse, presumably, also is true.

Assuming that this is so, the next part canvasses some issues that have arisen in

broadly comparable countries in connection with the separation of judicial power,

making due allowance for relevant differences including, critically in this context, the

impact of entrenched constitutions elsewhere. Those experiences in turn assist to

throw light on the effects and utility of the doctrine of separation of powers in its

application to the judicial branch.

3. Commonwealth experience with the separation of judicial power

For convenience, I divide the issues to which I propose to draw attention into two

groups: those connected with the institutional separation of powers (including the

impact of checks and balances) and those connected with a functional separation of

powers.

a. Institutional arrangements

At one level, the implications of an institutional separation of powers for the

constitution of the judiciary are straightforward. Judges may not be members of other

branches; members of other branches cannot be appointed to courts. Some significant

questions nevertheless arise at the margin. One is whether judges can perform other

public functions that are not judicial in character in their personal capacities. At the

Commonwealth level in Australia, where the separation of powers is constitutionally

protected, this question has generated a substantial body of case law, distinguishing

permissible from non-permissible functions by reference to a standard of

incompatibility and the purpose of protecting the perception of the independence of

the courts.22 Practice varies in the Australian states, where the doctrine receives

significantly less constitutional protection. In at least one state, Victoria, such

appointments will not be accepted by the judiciary at all.

22 Wilson v Minister for Aboriginal and Torres Strait Islander Affairs (1996) 189 CLR 1

- 7 -

The principal area of concern under the broad rubric of institutional separation

on which I wish to focus, however, arises from the points of institutional interface

between the judiciary and the other branches, representing checks on the judicial

branch. These are, of course, numerous, but I will focus on three in particular: the

appointment of judges; judicial remuneration; and the administration of courts. All

have implications for judicial independence. All necessarily and properly involve one

or other or both of the other branches, with their somewhat different preoccupations

and accountability mechanisms. To balance these potentially opposing interests, all

rely on inter-branch respect and self-restraint. Hope once was placed in the office of

Attorney-General to this end; but political realism has now prevailed. In all three

areas there is considerable movement, with no clear resolution in sight.

i. Appointment of judges

In all four countries, appointment of judges once lay with the executive

branch. Exercised with wisdom, such a mechanism is capable of producing good

results, questions about transparency and inclusiveness aside. Used as a source of

patronage, however or, worse, as a means by which the overall predilections of a

court might be influenced by the executive branch, it detracts from the quality and

independence of the courts.

Questions about transparency and inclusiveness can no longer so readily be

put aside, however, and the gains from patronage too often prove tempting. The

difficulty has been to find new arrangements that preserve at least some of the

strengths of the old, while overcoming their weaknesses. There is a range of variables,

all controversial in different degrees: the institution(s) through which appointment is

made; the degree of openness in the processes of application and selection; and the

extent of lay involvement in the appointment process.

The result for the moment is a smorgasbord of different procedures, on which

the Constitutional Reform Act may now also be placed.23 At one end is Australia, still

clinging to executive appointment to the judiciary, after some inconclusive inquiries 23 For other common law processes, see Ministry of Justice, “Appointing Judges: A Judicial Appointments Commission for New Zealand?” April 2004, http://www.justice.govt.nz/pubs/reports/2004/judicial-appointment/index.html x

- 8 -

into options for change.24 At the other is India, where since 1993, as MP Jain

describes it, in a masterly understatement: “the effective power to appoint Supreme

Court judges has …passed from the Executive to the Judiciary itself which has greatly

strengthened judicial independence”.25 Canada was until recently closer to Australia;

following changes announced on 20 February this year, however, nominees to the

Supreme Court must appear before a (televised) parliamentary committee hearing,

although the final decision remains with the Prime Minister, in the form of advice to

the Governor-General.26 In South Africa, the reconsideration de novo of the

constitutional settlement following the fall of apartheid provided an opportunity for

experimentation with new mechanisms for judicial appointment. The result is the

Judicial Services Commission, with a mixed membership of 23, which is consulted in

relation to some key appointments and identifies a short list for others, from which the

President must choose.27 The hearings of the JSC are public; a feature that can be seen

as important in principle in post-apartheid South Africa, whether or not is has the

effect of allowing South Africans to “get to know” the appointees that Canadian

Prime Minister recently claimed for his new appointment procedure.28

ii. Judicial terms and conditions

It is a commonplace for common law countries to adopt the Act of Settlement

guarantee against reduction of judicial salary while in office. It is equally a

commonplace that protection of this kind is symbolically significant but practically

insufficient. It takes no account of inflation or the relativities between the salaries of

24 Attorney-General Philip Ruddock “Selection and Appointment of Judges”, 2 May 2005, http://www.ag.gov.au/agd/WWW/MinisterRuddockHome.nsf/Page/Speeches_2005_Speeches_2_May_2005_-_Speech_-_Selection_and_Appointment_of_Judges . The Attorney noted, inter alia, his opposition to the notion of a Judicial Appointments Commission along United Kingdom lines: “…it represents an abrogation of the Executive’s responsibility…you just move the debate from who is being appointed to the bench to who is being appointed to the appointments commmission”.25 MP Jain, Indian Constitutional Law 5th ed 2003, Wedhwa and Company Nagpur, 332. See in particular SC Advocates on Record Association v Union of India AIR 1994 SC 26826 CBC News, The Supreme Court http://www.cbc.ca/news/background/supremecourt/ . The first Supreme Court Justice to be appointed following the process is Mr Justice Marshall Rothstein, on 1 March 2006. Justice Rothstein appeared before an “Ad Hoc Committee”, with a non-parliamentary chair, pending finalisation of the procedures. Announcing the appointment, the Prime Minister observed that it “marks an historic change in how we appoint judges in this country.  It brought unprecedented openness and accountability to the process.  The hearings allowed Canadians to get to know Justice Rothstein through their members of Parliament in a way that was not previously possible.” http://pm.gc.ca/eng/media.asp?category=1&id=1041 27 Constitution of South Africa, section 178 and implementing legislation28 Ibid

- 9 -

judges and other public office-holders, much less of the levels of remuneration

available at the bar. It is ambiguous in its protection of other terms and conditions of

judicial office including pension or superannuation entitlements.

It may be that a more effective guarantee can be fleshed out by reference to

broader constitutional principles, such as Canadian Charter section 11(d), protecting a

right to a hearing by “an independent and impartial tribunal”. In this event, however,

the fleshing out is done by the judges themselves, creating an unseemly impression of

self-interest that it is likely to be difficult to dispel. But the alternative is that the

adequacy of judicial remuneration lies almost entirely in the discretion of the

government and parliament of the day. A familiar checks and balances problem thus

arises. Granted, these are important questions for both the independence and quality

of the courts. But the provision of resources, the control of public spending and the

management of the economy lie, properly, with the other branches requiring, once

again, a mediating principle that recognises the responsibility of the elected branches,

while meeting the appropriate needs of the courts.

One solution, widely adopted, is the establishment of an independent body, to

advise on judicial salaries and other terms and conditions. Such a solution distances

deliberation on judicial salaries from the political process, but cannot remove it

altogether. A body of which I am a member illustrates the point. The Judicial

Remuneration Tribunal for the Australian state of Victoria for had authority, for a

period that proved startlingly short, to make determinations about judicial salary

increases that would come into effect through a special appropriation unless

disallowed by either House. The first such recommended increase survived

disallowance – just: the second failed to do so. An event of this kind arguably is

more damaging to the standing of the judiciary than leaving the decision to the normal

political process in the first place.29 In the course of the furore that followed, the

authority of the JRT over salary levels was removed, in favour of a system that tied

29 In her statement in response to the government’s decision, the Chief Justice of Victoria described it as “a direct interference with the independence of the courts”: http://www.supremecourt.vic.gov.au/CA256902000FE154/Lookup/MediaReleases/$file/Media-StatetCJJudSalApr04.pdf.

- 10 -

the salaries of Victoria judges over time to those of their federal counterparts,30

themselves dependent on determinations by the federal remuneration tribunal.

In Canada, these issues have given rise to a significant degree of litigation,

exploring the limits of the Charter. In an initial skirmish over judges’ pensions, in

which the validity of a contributory requirement was upheld, the Supreme Court of

Canada observed that judges must “bear their fair share of the financial burden of

administering the country.”31 Shortly afterwards, the Court was asked to advise on the

validity of legislation reducing the salaries of provincial judges. The Court held the

legislation invalid, because it was made “without recourse to an independent,

objective and effective process for determining judicial remuneration”.32

Commissions were duly established. In due course, however, their recommendations

for increases in salaries were not accepted and the issue returned to the Supreme

Court. The Court held that governments are not bound by the recommendations of

such Commissions as long as they justify their departure from such recommendations

with “rational reasons”, relying on a “reasonable factual foundation,” in a manner that

respects the purposes of the creation of the Commissions in the first place:

“preserving judicial independence and depoliticising the setting of judicial

remuneration”.33 Most of the challenged provincial decisions met these criteria; but

that of the government of Quebec did not. There is currently some media interest in

whether a recent decision by the Government of Canada not to accept in full the

recommendations of the federal commission ultimately will go to the Court, requiring

it to determine a question concerning the salaries of all federal judges, including their

own.34

iii. Administering the courts

30 Supreme Court, Victoria, Annual Report 2004 http://www.supremecourt.vic.gov.au/CA256902000FE154/Lookup/AnnualReport/$file/AR0204Review.pdf31 R v Beauregard [1986] 2 SCR 5632 Reference re Remuneration of Judges of the Provincial Court (P.E.I.), [1997] 3 S.C.R. 3; for consequential litigation concerning the validity of decisions made by courts in these circumstances, see Reference re Remuneration of Judges of the Provincial Court (P.E.I.), [1998] 1 S.C.R. 333 Provincial Court Judges' Assn. of New Brunswick v. New Brunswick (Minister of Justice); Ontario Judges' Assn. v. Ontario (Management Board); Bodner v. Alberta; Conférence des juges du Québec v. Quebec (Attorney General); Minc v. Quebec (Attorney General), 2005 SCC 44, [2005] 2 S.C.R. 28634 Canadian Bar Association E-News June 2006, http://www.cba.org/CBA/newsletters/news4-06/news.aspx

- 11 -

The constitutional dimensions of court administration have not, until quite

recently, become apparent. For much of the 20th century courts in common law

countries have been funded and administered, more or less effectively, through

Departments of Justice or their equivalents.

By the end of the 20th century, however, extending into the first part of the 21st,

this had become yet another issue between the courts and the other branches, raising

questions about the separation of powers. The reason for its sudden prominence may

lie in the equally sudden rise of new public sector management practices, with their

requirements for productivity measures, outcome based reporting and efficiency

dividends, with consequential implications for the administration of the courts. There

can be no dispute about Parliament’s role in appropriating funds for court services and

about the need for accountability for expenditure. Equally, there can be no dispute

about the interests of the executive in the efficiency and effectiveness of court

performance. On the other hand, as the Supreme Court of Canada has now

emphasised on several occasions, “administrative” independence is a core

characteristic of judicial independence, which may affect either its individual or its

institutional dimension.35 The scope of administrative independence for this purpose

has not been fully explored. The Court has said that it includes, at least, “the

assignment of judges, sittings of the court, and court lists — as well as the related

matters of allocation of court rooms and direction of the administrative staff engaged

in carrying out these functions”.36

The issue thus is the positioning of the dividing line between the functions of the

courts and the executive branch so as to safeguard that aspect of judicial independence

that concerns court administration while ensuring that the courts comply with

accountability and other public standards. For the moment, there is a range of

approaches to this issue, with Australia illustrating both ends of the spectrum. For the

most part the Australian states have retained the traditional model. The federal courts, on

the other hand, administer themselves, within a framework imposed by their constituent

statutes. Thus section 17 of the High Court of Australia Act 1979 empowers the court to

35 Reference re Remuneration of Judges of the Provincial Court (P.E.I.), [1997] 3 S.C.R. 3, Lamer CJ, [118]-[119]. See also Valente v. The Queen, [1985] 2 S.C.R. 673. 36 Valente v. The Queen, [1985] 2 S.C.R. 673. In the PEI case Lamer referred to this list as “narrow”: [117]

- 12 -

“administer its own affairs” including the power to contract and to manage court

precincts. The matters covered by the Court’s annual report suggests that the

management burden is significant.37 It seems, however, to have held the line in some

respects. Its 2004-5 report identifies for the Court “a single output, High Court business,

contributing towards the overall outcome, which is ‘interpreting and upholding the

Australian Constitution and performing the functions of the ultimate appellate Court in

Australia”.38 The price of the outcome was $14.3 million.

An illustration of the difficulties that can arise is provided by recent events in

South Africa. Section 165 of the Constitution of South Africa relevantly provides that:

The judicial authority of the Republic is vested in the courts

The courts are independent…

No person…may interfere with the functioning of the courts

Organs of state…must assist and protect the courts to ensure [their]

independence.

In 2005, extending into 2006, the government proposed an amendment to the

Constitution which would have added two new sub-sections, specifying that the Chief

Justice was “head of the judicial authority” and that the “Cabinet member responsible

for the administration of justice exercises authority over the administration and budget

of all courts”.39 What did this mean; and did it matter? Many people thought that it

did; and their concerns were exacerbated by the terms of a proposed Superior Courts

Bill creating, for example, a position of Executive Secretary to the Chief Justice as an

officer in the department, with general responsibility to administer the office of the

Chief Justice, admittedly under his direction. The government for its part argued that

it was doing no more than giving effect to what it described as “the Commonwealth

model of the separation of powers between the Executive and Judiciary”. In the

context of court administration that is something of an exaggerated claim. For the

moment I am told that the proposals are shelved. It seems likely however that they

will in time re-emerge, in this or an altered form.

37See for example the requirement for “ Monthly reporting to the Department of Finance and Administration, in accordance with the Budget Estimates and Framework Review”, High Court of Australia, Annual Report 2005, 3138 Ibid, 31-3239 Constitution Fourteenth Amendment Bill 2005

- 13 -

iv. Conclusion

At this point some brief concluding remarks about institutional separation are

in order. The subject raises a wide range of issues. I have focussed only on three, in

which there is a significant interface between the judiciary and the other branches. In

each case the arrangements in question are long-standing. In each, for a variety of

reasons, the traditional forms are undergoing change. The process of change is

irreversible. It has implications for the composition and operation of the courts and,

by extension, for judicial independence. As may be inevitable at such a time, the

changes are taking place in isolation from each other. There is a still-open question

about the extent to which a doctrine of separation of powers can be developed to give

them direction and coherence.

b. Questions about power

I now move to explore, albeit more briefly, the issues that arise in connection

with a functional separation of powers, with its emphasis on the isolation of judicial

power itself, rather than of the institution by which it is exercised.

i. Scope of judicial power

The doctrine necessarily has two dimensions. First, it identifies and, under a

system of controlled constitutionality prescribes, what courts should do (and other

branches should not do). Viewed in this light, the doctrine has implications for due

process and civil liberties. These can best be illustrated by experience in Australia,

where the doctrine of separation of judicial power is highly developed and formal

rights protection is almost entirely absent.40 It is clear in Australia, for example, at

least at the federal level, that only a court can adjudge a person guilty of an offence

and impose punishment accordingly41 or exercise other “core” judicial functions. The

doctrine also inhibits other branches, at least to a degree from directing the courts in

40 Justice Deane once described the separation of judicial power as the ‘most important’ of the guarantees of rights and immunities, express or implied, under the Australian Constitution: Street v Queensland Bar Association (1989) 168 CLR 461, 521 41 Polyukhovich v Commonwealth (1991) 172 CLR 501, per Mason CJ

- 14 -

the manner of the exercise of their powers.42 For a time, it also appeared that the

doctrine might limit the authority of the Parliament to order detention without trial

(obviously exceptional circumstances, such as quarantine, aside) although it no longer

seems likely that this will prove to be so.43

ii. Limitations on judicial power

In its second dimension, however, a functional separation of powers serves to

identify what courts may not do. In a system that recognises constitutional rather than

parliamentary supremacy, the doctrine thus precludes conferral on courts of non-

judicial power, requiring some fine lines to be drawn.44 More significantly, perhaps,

for present purposes, the doctrine may be used to suggest limits that the courts should

impose on themselves. In Australia, for example, a separation of powers argument has

been used to explain why courts should hesitate to supply the deficiencies of statutes,45

review decisions on the merits,46 overturn previously established decisions,47 invalidate

legislation prospectively,48 or apply unincorporated international treaties.49 Paul Craig

has argued that in these respects “The concept operates as a source of judicial

legitimacy; with the courts defending their role as the rightful interpreters of

legislation, and of the legality of executive action. It serves also as the foundation for

judicial restraint…”50

This is, of course, nothing new.51 But it may have new significance, at a time of

expansion in both the depth and breadth of judicial review. Expansion in depth tends

to break down the distinction between judicial and merits review. Expansion in

42 Chu Kheng Lim v Minister for Immigration (1992) 176 CLR 1, in which the majority of the Court invalidated a section that they interpreted as a direction from the Parliament not to release specified persons from custody, even if they were unlawfully held43 The possibility was based on a reading of Lim. It finally faded with the decision in Al-Kateb v Godwin [2004] HCA 37. See generally Simon Evans “Australia” (2006) 4 International Journal of Constitutional Law 51744 In Australia, for example, federal courts cannot be given arbitral power, which has been held to be non-judicial45 Strickland v Rocla Concrete Pipes Ltd (1971) 124 CLR 46846 Minister for Aboriginal Affairs v Peko-Wallsend Ltd (1986) 162 CLR 2447 Queensland v Commonwealth (1977) 139 CLR 58548 Ha v New South Wales (1997) 189 CLR 465.49 Minister for Immigration and Ethnic Affairs v Teoh (1995) 183 CLR 273, 28750 Craig, op.cit., 70751 Duport Steels Ltd v Sirs [1989] 1 All ER 529, 541; R v Secretary of State for the Home Department; ex p Fire Brigades Union [1995] 2 AC 513, 567; Bradley, op.cit., 340-341

- 15 -

breadth brings the judiciary face to face with the executive in unfamiliar and

sometimes sensitive areas. Both have the potential to lead to claims that the judiciary

is trespassing into the sphere of the other branches. It is possible to refute some such

claims by reference to the separation of powers. As David Feldman has recently

observed: “One can respect the role of the executive as makers of policy while

upholding the obligation of the courts to ensure that the policy is not unlawful…”52 At

some point, however, separation of powers places limits on judicial review. The

courts are unlikely to be the only branch to have a view about where these limits are

properly drawn.

4. Evaluation

A doctrine of separation of powers now provides the principal framework within

which the relationship of the courts to the other branches of government is resolved in

this and many other Commonwealth countries. The courts were not originally the

principal focus of the doctrine; clearly enough, however, it can be adapted to the

purpose.

In its present form, however, the doctrine is somewhat too general to provide

much assistance in resolving a range of existing and emerging difficulties at the

interface between courts and the other branches. A degree of tension is inevitable,

representing checks and balances in action. It may also be that what is perceived as a

difficulty by one branch may be viewed in a different light by others.53 Elaboration

nevertheless would be useful, in the interests of both persuasion and prediction. At

the very least the separation of powers in relation to the judiciary needs to be

explained in terms of purposes specifically applicable to the judiciary: protecting

judicial independence, grounding legitimacy and securing the rule of law. Those

values in turn might be better understood and accepted by placing judicial review in

the context of a broader constitutional vision; as an integral part of a tradition in

which the virtues of limited government have managed to survive an ever-closer

union between the legislative and executive branches.

52 David Feldman, “Human Rights, Terrorism and Risk: The Roles of Politicians and Judges” [2006] Public Law at 37653 David Feldman, ‘None, one or several? Perspectives on the UK’s constitution(s)’ [2005] CLJ 329

- 16 -