Separation of Powers: Doctrine and Practice


Author: Graham Spindler
(This article, suitable for Year 11-12 Legal Studies, originally appeared in the publication Legal Date in March 2000.)

In 1989 at the Fitzgerald Enquiry into corruption in Queensland, Jo Bjelke-Petersen - then Premier of Queensland for 19 years - was asked to define the doctrine of the separation of powers in the Westminster system. He could not answer the question. The Queensland Premier certainly understood ‘power’, but his ignorance of what is usually regarded as one of the most fundamental tenets of liberal democracy demonstrated not only that he was no student of the theory of government, but that the democratic process in Queensland had become precarious.

In fact, the doctrine is not exemplified in the constitutions of the Australian states. However, the practice is usually evident, and if the object of separation of powers is to develop mechanisms to prevent power being overly concentrated in one arm of government, then state processes do eventually have that effect.

Defining the Doctrine
The doctrine of the separation of powers divides the institutions of government into three branches: legislative, executive and judicial: the legislature makes the laws; the executive put the laws into operation; and the judiciary interprets the laws. The powers and functions of each are separate and carried out by separate personnel. No single agency is able to exercise complete authority, each being interdependent on the other. Power thus divided should prevent absolutism (as in monarchies or dictatorships where all branches are concentrated in a single authority) or corruption arising from the opportunities that unchecked power offers. The doctrine can be extended to enable the three branches to act as checks and balances on each other. Each branch’s independence helps keep the others from exceeding their power, thus ensuring the rule of law and protecting individual rights.

Obviously under the Westminster System – the parliamentary system of government Australia adopted and adapted from England – this separation does not fully exist. Certainly in Australia the three branches exist: legislature in the form of parliaments; executive in the form of the minsters and the government departments and agencies they are responsible for; and the judiciary or the judges and courts. However, since the ministry (executive) is drawn from and responsible to the parliament (legislature) there is a great deal of interconnection in both personnel and actions. The separation of the judiciary is more distinct.

Origins of the Doctrine
States throughout history have developed concepts and methods of separation of power. In England, parliament from its origins at least seven centuries ago was central to an struggle for power between the original executive (the monarch) and the councils of landowners, church leaders and commons. Similarly judges, originally representing the executive, developed increasing independence. Parliament was a significant force in an increasingly mixed form of government by the time of the Tudors and soon afterwards was directly challenging the doctrine of the divine right to power of the Stuart monarchs. The English Civil War (1642-60) between parliament and monarchy resulted in the monarchy continuing but under an arrangement which established not only parliament’s legislative authority but also opened the way to the development of cabinet government.

In his Second Treatise of Civil Government, English philosopher John Locke (1632-1704) noted the temptations to corruption that exist where “... the same persons who have the powers of making laws to have also in their hands the power to execute them ... “. Locke’s views were part of a growing English radical tradition, but it was French philosopher, Baron de Montesquieu (1689-1755), who articulated the fundamentals of the separation doctrine as a result of visiting England in 1729-31. In his The Spirit of Laws (1748), Montesquieu considered that English liberty was preserved by its institutional arrangements. He saw not only separations of power between the three main branches of English government, but within them, such as the decision-sharing power of judges with juries; or the separation of the monarch and parliament within the legislative process.

Locke and Montesquieu’s ideas found a practical expression in the American revolution in the 1780s. Motivated by a desire to make impossible the abuses of power they saw as emerging from the England of George III, the framers of the Constitution of the United States adopted and expanded the separation of powers doctrine. To help ensure the preservation of liberty, the three branches of government were both separated and balanced. Each had separate personnel and there were separate elections for executive and legislature. Each had specific powers and some form of veto over the other. The power of one branch to intervene in another through veto, ratification of appointments, impeachment, judicial review of legislation by the Supreme Court (its ability to strike down legislation or regulations deemed unconstitutional), and so forth, strengthened the separation of powers concept, though inevitably involving each branch in the affairs of another and to some extent actually giving some of the powers of one branch to another

It was a high water mark in institutionalising individual liberty through the separation of powers and one embedded even further by early judgements of the Supreme Court but, as the struggles, inefficiencies and political gamesmanship illustrated by the recent Clinton impeachment attempt or by Congress’s delaying of budgets, it also made government harder. This had been partly the intention. Few subsequent democracies have fully adopted the American approach, but the concept is widely aspired to, though taking varying forms amidst the complex interplay of ideas, interests, institutions and Realpolitik that are part of each system of government.

The Doctrine in Australia - the Commonwealth
While certainly not the American model, a form of the doctrine operates in the Australian versions of the Westminster model, most notably in the Federal Constitution. The writers of the Australian Constitution in the 1890s retained the Westminster cabinet system. Unlike the Americans of the 1780s, they had several working democratic federal constitutional models to examine, along with well-established democratic traditions of their own, and wanted to maintain strong ties with Britain, not create a revolution. Their interest in the U.S. Constitution was more in its mechanisms of federation such as the Senate, than in the checks and balances between branches of government. Nevertheless some elements can be found.

The Australian Constitution begins with separate chapters each for the Parliament, Executive and Judiciary, but this does not constitute a separation of powers in itself. Executive power was nominally allocated to the Monarch, or her representative the Governor-General (Section 61), while allocating it in practice to the Ministry by requiring the Governor-General to act on the Government’s advice (subject, of course to the Governor-General’s controversial ‘reserve powers’). This was the Westminster model and it relied on convention as much as the words of the Constitution. However, the specific requirement for Ministers of State to sit in Parliament (Section 64) clearly established the connection between Executive and Parliament and effectively prevented any American-style separate executive.

The situation with the judiciary, however, was different. The whole of Chapter III of the Constitution (Judicial Power of the Commonwealth) and Section 71 in particular, has been used by the courts to establish a strict separation of powers for Federal Courts from the ministry and parliament. In New South Wales v. Commonwealth (1), the High Court ruled that this part of the Constitution does embody the doctrine of separation of judicial powers. This also applies to tribunals and commissions set up by Federal Parliament which, unlike some of their equivalents in the states, can only recommend consequences. The Federal Parliament itself, however, has the rarely used privilege of being able to act as a court in some circumstances, primarily where it may regard a non-member as acting “in contempt” of parliament.

However, the Courts have found that the separation that exists for the judiciary does not strictly apply to the relationship between executive and legislature (2). In Victorian Stevedoring and General Contracting Co. v. Dignan (3), it was found that legislative power may be delegated to the executive. The same case, however, reconfirmed the separation of judicial powers. Thus, while the courts are separate and the High Court can rule on legislative and constitutional questions, the executive is not only physically part of the legislature, but the legislature can also allocate it some of its powers, such as of the making of regulations under an Act passed by Parliament. Similarly, the legislature could restrict or over-rule some powers held by the executive by passing new laws to that effect (though these could be subject to judicial review).

The Constitution does provide for one form of physical separation of executive and legislature. Section 44, concerning the disqualifications applying to membership of Parliament, excludes from Parliament government employees (who hold “an office of profit under the crown”) along with people in certain contractual arrangements with the Commonwealth. This was demonstrated in 1992 after Independent MP, Phil Cleary, had won the Victorian seat of Wills. Cleary, on leave without pay from the Victorian Education Department at the time of his election, was held to be holding an office of profit under the Crown and disqualified (4). The Court noted that that Section 44’s intention was to separate executive influence from the legislature. This requirement does not apply to state elections.

Elections themselves, in recent years, have reflected voter concern with separation of powers-related issues. In 1995, NSW voters overwhelmingly endorsed a referendum proposal clarifying the independence of judges. In the 1999 Victorian election, voters appeared to reject a perceived concentration of power by the Premier, particularly in his gagging of fellow party members and changes to the role of the state Auditor General.

Even though the Australian Constitution says little about political parties, parties have an important impact on the relationship of powers between executive and legislature. The existence of varied political parties is a feature of the freedoms of opinion essential to a liberal democratic system and the contest between them is a factor in controlling the potential excesses of any one group. However, the system can have other effects. Since by convention the party controlling the lower house forms the government, then the ministry (being also the party leaders) also exerts authority over the lower house. The exceptional strength of Australian party discipline ensures that, within the house, every member of the numerically larger party will almost always support the executive and its propositions on all issues. Despite debates and the best efforts of the Opposition and Independents (particularly in Question Time), this inevitably weakens the effective scrutiny of the executive by the legislature.

Party domination in Australia thus further reduces the separation between executive and legislature, although Parliamentary processes do usually prevail. However, robust democratic systems have a capacity to self-correct, as has been demonstrated by the Senate. Because of the party system, the Senate failed to ever be ‘the states’ house’ originally intended by the Constitutional framers. However, the adoption of a proportional system of voting in 1949 created a new dynamic and the Senate in recent decades has rarely been controlled by Governments. Minor parties have gained greater representation and Senate majorities on votes come not from the discipline of a single party but from a coalition of groups on a particular issue. This happens in most democracies but in Australia is often regarded (particularly by supporters of the major parties) as an unnatural aberration. As a result the role of the Parliament as scrutineer of executive government, immobilised to some extent in the Lower House by the party system, has been expanded by the Upper House.

The Doctrine in Australia - the States
In the case of the Australian states, where the basic governmental structures were in place before the Australian Constitution, separation of powers has little constitutional existence even though it is generally practiced. This has been shown in cases such as Clyne v. East (5) for NSW and the doctrine extensively discussed in cases such as Kable v. The Director of Public Prosecutions (6). In these and other judgments it was noted that a ‘general’ doctrine of separation of powers operates as accepted practice in the state through constitutional convention. That the position is similar in other states has been confirmed in cases in Victoria (7), Western Australia (8) and South Australia (9).

In practice, there is far more crossing of responsibilities in the states than Federally. As with the Commonwealth, Ministers have powers to make regulations (in effect, legislating) and are, of course, Members of Parliament and responsible to it. Again, the rigid party system increases the domination of at least the lower house by the executive from the majority party and there are often complaints that the executive is manipulating parliament or treating it with contempt. In some cases upper houses have increased their roles of scrutiny of the executive, though this varies according to the electoral systems used for upper houses – where they exist.

Parliamentary scrutiny of the executive and, in particular, by the NSW Upper House, was tested in 1996-99 when Treasurer Michael Egan, on behalf of cabinet, refused to table documents in the Legislative Council of which he is a member. The documents related to several controversial issues, and the reasons given for this refusal included commercial confidentiality, public interest immunity, legal professional privilege and cabinet confidentiality. The Council, determined to exercise its scrutiny of the executive, pressed the issues and eventually adjudged the Treasurer in contempt, suspending him from the house twice. The matters were disputed in three cases in the High Court and the Supreme Court of NSW (10-12). The results upheld that the Legislative Council did have the power to order the production of documents by a member of the House, including a minister, and could counter obstruction where it occurred. However, the question of the extent of the power as regards cabinet documents, will be subject to continued court interpretation.

In relation to the judiciary, traditionally the most separated and independent arm, the separation so clearly established in the Commonwealth does not exist in the state constitutions. Nevertheless, certain state courts, having had jurisdiction to deal with Federal laws conferred on them by the Commonwealth Parliament, have in effect a Federal Constitutional basis for separation of their powers. The general separation of state courts is practiced, but the issue of tribunals set up by state parliaments is different since such bodies sometimes exercise both executive and judicial power through being able to impose fines or penalties. The Administrative Decisions Tribunal in NSW is one such example.

In NSW, the issue of judicial independence was recently raised in a rare Australian instance of a legislature exercising scrutiny over a judge. The power of removal of a judge in NSW lies with the Governor on Parliamentary recommendation, the possible grounds being proved misbehaviour or incapacity. In 1998 the Judicial Commission recommended Parliament consider removal of a Supreme Court Judge on the grounds of incapacity. In the Court of Appeal (13) the Judge, Justice Bruce, argued that this contradicted the concept of the independence of the judiciary. The Supreme Court agreed that, despite the lack of any formal separation of powers in the NSW Constitution, the Commonwealth Constitution did significantly restrain Parliamentary interference with the judiciary. Nevertheless, the court held that nothing had occurred that would impinge on the integrity of the judicial system and that Parliament could consider the case. Justice Bruce appeared before the Legislative Council but removal was not recommended.

In Conclusion
While the doctrine of the separation of powers and its practice will not necessarily be the same thing, the purpose behind the doctrine can be seen to be embedded in democracies. In the Westminster system, as practiced in Australia, discussion of the doctrine is riddled with exceptions and variations. Certainly, in its classical form it exists here only partially at best; but in practice mechanisms for avoiding the over-concentration of power exist in many ways – through constitutions and conventions; the bicameral system; multiple political parties; elections; the media; courts and tribunals; the federal system itself; and the active, ongoing participation of citizens. The doctrine is part of a simultaneously robust and delicate constant interplay between the arms of government. A tension between separation and concentration of powers will always exist, and the greatest danger will always lie with the executive arm – not judges or legislatures – because in the executive lies the greatest potential and practice for power and for its corruption. Preventing this in our system relies as much upon conventions as constitutions and the alarm bells should ring loudly when government leaders dismiss or profess ignorance of the concept.


Case References
(1) New South Wales v. Commonwealth (1915) 20 CLR 54
(2) Attorney-General (CTH) v. R. (1957) 95 CLR 529
(3) Victorian Stevedoring and General Contracting Co. Pty Ltd v. Dignan (1931) 46 CLT 73
(4) Sykes v. Cleary and Others (1992) 176 CLR 77
(5) Clyne v. East (1967) 2 NSWLR 483
(6) Kable v. The Director of Public Prosecutions for New South Wales FC 96/027 Commonwealth Constitution
(7) Collingwood v. Victoria (No. 2) (1994) 1 VR 652
(8) Nicholas v. Western Australia (1971) WAR 168
(9) Gilbertson v. South Australia 919780 AC 772
(10) Egan v. Willis and Cahill (1996) 40 NSWLR 650
(11) Egan v. Willis (1998) 73 ALJR 75
(12) Egan v. Chadwick and Others (1999) NSWCA 176 revised 04/08/99 CA 40828/98
(13) The Honourable Justice Vince Bruce v. The Honourable Terrence Cole and Others, Matter No CA 40337/98 NSWSC 260 (12 June 1998)