Alan J. Ward
Let me start by explaining how I came to be here. For a number of years I have studied parliamentary governments in the Commonwealth and Europe, and in that process I developed an eight-part model which captured, I thought, the essence of parliamentary government. Then I came face to face with the Australian Senate, which seemed, at first glance, to play by different rules. So I had to spend some time investigating Australia, and that ultimately brought me to Canberra.
Today I want to talk about three things. First, I want to explain why I now think that Australia fits my parliamentary model. Second, I want to consider the role that upper houses play in Australian government. And third, I want to make some suggestions for constitutional reform that I expect you to find naive. I will necessarily paint with a very broad brush, given the scope of the topic, and for that I ask your forgiveness. As John Wayne once said, `I don't do nuance.' I also recognise that Australians have been debating these subjects since at least the Federation debates of the 1890s, and very intensively during the republic debates of recent years, but I thought it might be useful to have a foreign perspective on these matters.
First, then, why do I think Australia is a relatively orthodox parliamentary state, and what does that mean anyway? When I taught in Australia in the 1960s, federal politics was dominated by Robert Menzies, who listed heavily in the direction of things British, and so did academic interpretations of Australian government. In a 1964 article, for example, Gordon Reid wrote, `All seven units of government in Australia are based on the Westminster model … For Australians it was the sine que non of internal self-government'.  Since the 1960s, however, Australian political scientists have often described the federal government in non-Westminster terms. In 1980, Elaine Thompson coined the term Washminster to describe a hybrid American and British model of government.  When Dean Jaensch endorsed this notion in 1991 he wrote, `Any analysis of parliamentary democracy in Australia which starts from the premise that we applied a Westminster model is doomed to irrelevance'.  I think Jaensch was wrong, although I can appreciate why someone comparing Canberra and Westminster very specifically might be drawn to their dissimilarities: federalism, a formal constitution, judicial review, and the like. But if one were to focus on a generic model of parliamentary government, not the Westminster variant, British and Australian similarities would become very clear. Furthermore, as David Butler argued many years ago they illuminate each other in very useful ways. 
At the core of parliamentary government is the parliamentary executive, the principle that the government must have the support of a majority in Parliament. I should note at the outset that there are actually two kinds of parliamentary executive. In one, the members of the government sit as voting members of Parliament. We find this model in the Australian Commonwealth, Japan, and Germany, amongst other states. In the other model government members are not voting members of Parliament, although they may speak there. We find this model in many continental European countries, including Belgium, Austria, and the Czech Republic. I will primarily talk about the former, the internal executive model. It has eight core characteristics. Some of these are constitutional rules, and some are behaviours that follow from rules.
The first characteristic is what Walter Bagehot, in 1867, called fusion, the rule that members of the government sit in Parliament. Unusually for Australia, this rule is written into all Australian Constitutions, federal, state, and territory.
The second characteristic is also a rule, that the government must have the support of a majority in Parliament, or in the lower house of a bicameral Parliament, and it must resign if it loses that support. This is the essence of what colonial politicians called responsible government. At a minimum, as practiced in Britain, this rule requires the government to have a majority on confidence motions and major budget votes, but in many European constitutions it means much more. It can mean, for example, as in Spain, that the lower house appoints the Prime Minister, approves his choice of ministers, and approves his program before he takes office. Each decision is by a majority vote. 
This notion of majority rule came to the Australian colonies by order of the Colonial Office in the 1850s but it has been written into only one Australian Constitution, the ACT Act 1988. Nonetheless, with the notable exception of Sir John Kerr, who sacked a government with a lower house majority in 1975, it is observed everywhere in Australia as a convention.
The third characteristic is the one that gave me doubts when I first looked at Australia. It is the rule that in a bicameral parliament, one chamber has primacy. The parliamentary model rejects the proposition that a government can be responsible to two chambers, because they might be controlled by different majorities.
When I looked at what primacy means, it became clear to me that Australia satisfies this condition, at both the federal and state levels. Primacy is measured by four things, all of which are very clearly identified in a number of ways in modern parliamentary constitutions. First, the government is formed by the party or coalition which has a majority in the lower house. Second, the Prime Minister is a member of the lower house. Third, a majority of ministers sit in the lower house. And fourth, the lower house, or effectively the government that controls the lower house, possesses legislative initiative. Financial bills originate there, and most other legislation begins there too. Furthermore, legislation that originates in the upper house is most often government legislation, introduced there because of time constraints in the other house. In most parliamentary states, the upper house may only delay, not deny, legislation, but even where an upper house has the power to deny all, or certain, bills, as in the German and Indian federations, there is a presumption that the government will determine the bulk of the legislative program. This is certainly true of Australia, and I was intrigued to find that Senate Standing Orders provide for government senators to chair each of the legislation committees, even when the government is in the minority in the Senate. You certainly do not find this in Washington.
Australia, therefore, meets all four of these standards of lower house primacy, and this raises a very interesting question. The Washminster argument depends, above all, on the notion that the founders adopted an American federal Senate, and gave it American powers to reject any bill, including supply. Like its American counterpart, the Australian Senate lacks financial initiative. Unlike the American Senate, it may not amend a budget, but both senates may reject a budget. It is from this perspective that Elaine Thompson argues that the Australian Senate is not a house of review in the English sense. `The Senate is powerful in its own right. It is a second chamber in the American sense.' She adds that this `stand-off principle of Australian federalism is at odds with the idea that the government governs with the support of the lower house.'  But if the founders created a nearly co-equal upper house which is at odds, as Thompson says, with a government which is responsible only to the lower house, why does Australia satisfy the four standards of lower house primacy that I have identified?
It is clear that the founders considered the American Senate when they were devising the federal Senate, but we tend to overlook the fact that in 1900, when the Commonwealth Constitution was adopted, the House of Lords, the Canadian Senate, the New Zealand Legislative Council, and all the colonial legislative councils had the power to reject any bill, including supply, and this fact had absolutely nothing to do with American federalism. Thompson says the Australian Senate is an American upper house, but one could as easily have said its powers are those of a colonial Legislative Council. These powers were not unusual at the time, although they were in the process of being curtailed in the parliamentary practice of the time. However, the federal justification for the powers was new to Australia and the Senate was unique in being a popularly elected upper house. It was its legitimacy, therefore, and not only its powers, that John Quick, for one, at the Sydney Convention in 1897, believed would lead the Senate to assert itself in an American way. Hence, in his view, the need for constitutional procedures to adjust disputes between the two houses.  What the founders found difficult to predict, however, was quite how assertive the Senate might be. In particular would it act so as to make a government responsible to the other house unworkable? Winthrop Hackett was alluding to this when he said, at the 1891 Constitutional Convention, `[E]ither responsible government will kill federation, or federation, in the form which we shall, I hope, be prepared to accept it, will kill responsible government.'  For those who thought responsible government would fail, the decision to entrust the executive to the Governor-General and not spell out the rules of responsible government in law meant that the Governor-General would be free to adopt a non-British style of executive—the Swiss model was mentioned—if responsible government were to fail.  But the majority position at the Convention was to wait and see. Perhaps Australia could muddle through.
Once the federation settled down, Australia did muddle through. The government and the Senate learned to work together pretty much as state governments learned to work with Legislative Councils. This happened because parliamentary norms prevailed over American senatorial norms, and these led to the primacy of the lower house. It seems to me, furthermore, that parliamentary norms are growing stronger over time in Australia. No upper house has rejected supply since 1952, in Victoria. The Labor and Australian Democrat parties believe supply should never be denied, and the Liberal party makes only a very theoretical defence of this power. And finally, no-one appears to relish the thought of replaying the crisis of 1975, when the Senate blocked a vote on supply.
I will return to this subject a little later, but having dallied overlong on the primacy of the lower house, let me deal quickly with the remaining five characteristics of parliamentary government.
The fourth characteristic is a behaviour pattern that follows a rule. If a government can only take office, and remain in office, with the approval of a majority in the lower house, there is a huge incentive to organise disciplined parliamentary parties because without them governments would come and go every week. It is party discipline, of course, that gives the government its control of the lower house, and where parties are represented in both houses they are customarily directed from the lower house because that is where government is won or lost. Parties are not conspicuous in parliamentary constitutions, but they are creeping into provisions on casual vacancies in Australia, and they are very evident in Standing Orders.
Characteristics five and six of the parliamentary model follow from everything I have said so far. The fifth is that parliamentary government is Cabinet government. The executive is a committee of the majority in the lower house. And sixth, within the Cabinet, the Prime Minister or Premier has decisive influence. Between them, the Prime Minister and Cabinet control the resources and legal authority of office. They also control a disciplined parliamentary party or parties, national party organisations, and media attention.
If effective executive power lies with the Prime Minister and Cabinet, it follows, and this is my seventh characteristic, that there is no place in the model for a strong head of state. In fact some small Pacific Island states dispense with a separate head of state altogether by combining the jobs of Prime Minister and President, which proves, perhaps, that a state does not absolutely have to have a separate head of state.
Finally, all the characteristics I have identified so far—fusion, government with majority support, the primacy of the lower house, parties, Cabinet government, the special powers of the Prime Minister, and the weakness of the head of state—lead in one direction, to the concentration of power in the hands of the Prime Minister and Cabinet, who dominate the lower house through party discipline. Indeed, a perennial complaint levelled at parliamentary government is that it creates the `executive state', with governments that are too strong and legislatures that are too weak.
If you have followed me so far it will be obvious that I think Australia falls squarely within that group of states that one can call parliamentary. There is almost no `Wash' in the model and a great deal of `minister', or rather a great deal that is parliamentary in the sense that I have outlined. This is a model that can accommodate federal and unitary states, monarchies and republics, bicameral and unicameral states, and relatively powerful upper houses as well as weak ones. All can share these eight characteristics. That said, every parliamentary state is different from every other one in some way and I want, as my second topic, to say something favourable about the special way Australia operates parliamentary government, specifically the role of the upper house.
The upper house is always a problematical chamber in a bicameral system and many countries have chosen not to have one. Indeed, only fifteen British Commonwealth members have a second chamber. The fundamental problem is how to give the second chamber sufficient legitimacy and authority to avoid redundancy, but not so much as to make it impossible for a government selected from the lower house to govern.
It is always in the interest of a government to resist any growth in the authority of the upper house, but too many parliamentarians and academics have internalised the government's value that a strong upper house is unnatural in a parliamentary system. Recently an English friend, a constitutions specialist, sent me this message:
I went to a `seminar' at the Lords, on reform of the Lords, by Lords, last week. A better argument for abolition you'd go far to find; complacent not to say smug, with not an idea between them except to agree with the Labour Lord chairing the event that the future Lords must be more legitimate than at present but `less legitimate than the more legitimate House' so as not to be a source of potential `trouble'.
It is hard to believe that the `Labour Lord' referred to is a member of the party of upper house reform in the United Kingdom!
I found the same kind of timidity about upper house reform in Ireland. The Irish Senate has defeated a government motion only three times since the present constitution was adopted in 1937, but the Constitution Review Group that published a report in 1996 made no substantial suggestions for Senate reform. Indeed, the group would not be unhappy were the Senate to be abolished. Eamon de Valera, who as Irish Prime Minister was responsible for the present Irish Constitution, believed that Parliament exists to elect the government, and should then get out of the way, and the Review Group accepted this view implicitly. Everything it recommended about Dail Eireann (the lower house), the Senate, the President, and the government was designed to protect the government. 
There was no recognition in the Review Group report of the excessive concentration of power in the executive that Australians call `the executive state', even though Ireland had already begun a round of judicial inquiries into government misconduct that continues today and closely parallels the royal commissions into government corruption or impropriety conducted in four Australian states in the 1990s. What the Australian inquiries found was that parliamentary executives in the states do not adequately monitor themselves, that parliaments do not adequately monitor executives, and that systems of public service accountability are hopelessly flawed.  To date the Irish have not come to this conclusion.
If you accept, as I do, that the concentration of power in the executive is an unfortunate, and very often corrupting, characteristic of modern parliamentary government, you will probably accept that just about the only check on the executive comes from a house that it does not control. Sometimes it is the lower house that is not controlled, if the government finds itself dependent on independents or minor parties for its majority. This happened in New South Wales and Tasmania, for example, in recent years, where governments had to concede substantial reforms of Parliament in order to secure majorities.  In Australia it is generally to the upper house that one looks for a check on the executive. It would be redundant here to record what the Australian Senate has done in the past thirty years in this regard, but I want to stress that state upper houses have asserted themselves too since they were democratised between 1950 and 1978. For example, when Barbara Page examined periods between 1976 and 1989 when the government did not control the New South Wales Legislative Council, she found that ministers worked harder than hitherto to brief, consult and negotiate with members of the Council, and that members of the Council came to see themselves as full-time legislators. Their sitting hours doubled, their average attendance rose, they took an interest in a broader range of activities and formed new standing committees.  All very admirable in my view. Scott Bennett finds similar developments elsewhere. There is, he says, a new `upper house ethos', and Legislative Councils are `no longer ridiculed as retirement homes for geriatrics or attacked as the havens of ultra-conservative politicians.' 
The important lesson from Australia about the upper house is that, contrary to what is generally believed overseas, effective parliamentary government does not collapse when an upper house asserts itself. I know that Australian federal and state governments often have to tailor their budgets and bills to accommodate the upper house, which is no bad thing if the upper house is a legitimately representative body, but the evidence is strong that governments still hold the legislative initiative and get most of what they want, and certainly most of what they absolutely need, out of the upper house.  What we see at work in state and federal upper houses most of the time is what Senator Fred Chaney describes as `a degree of enforced reasonableness', enforced in the sense that unless the Senate restrains itself, governing is impossible.  Barbara Page writes, `For upper houses to perform an effective review function without undermining [lower house] primacy, they need to operate with a mixture of independence and restraint.'  Australian politicians understand what parliamentary government entails, and they do not push it to the point that it freezes up, which is exactly what has happened in the United States in recent years of divided government in the executive President model.
The last major review of House of Representatives procedures was, I think, the Blewett report of 1993 and the data it presented showed me, at least, that the Labor Governments of 1983 to 1996 undermined the House of Representatives by mauling question time and misusing the guillotine. But Blewett's own conclusion from this was not the one we hear most often in the northern hemisphere, that there should be reform of the lower house with lots and lots of new committees and expanded private members' opportunities. He knew the Australian federal government would not accept this. It already feels harassed by Senate committees and is not about to add House of Representatives committees to its problems. Instead, in his Gordon Reid lecture, published in 1994, Blewett said:
It may be ... that instead of paying attention to reform of the House of Representatives we should accept that chamber as essentially a debating forum between two party teams, and particularly their leaders, designed to clarify choices for a mass electorate, and concentrate on perfecting the Senate as a House of legislative review and as the body for effective scrutiny of the Executive. 
I think this was an awfully wise statement, one that probably only an Australian would make because the upper house is so discounted in other parliamentary countries. Furthermore, the statement applies to state parliaments as well as the Commonwealth. Indeed, the Royal Commission on WA Inc. recommended just such a role for the Western Australian Legislative Council in 1992. It also recommended that no ministers should sit in the Council in order to distance the upper house from the ministry. 
Turning to the upper house for effective review and scrutiny, as Blewett suggests, is not without risk. There is the risk of legislative stalemate, for example, because none of the conflict resolution procedures in the federal and state constitutions work adequately. There is also a risk of overloading members. There is only so much that the Tasmanian Legislative Council, for example, with nineteen members, can do, or even the Commonwealth Senate, with seventy-six members covering over forty committees. The question of giving upper houses additional staff support to do their jobs adequately has to be addressed. On balance, however, I think that Australian bicameralism is working quite well. Furthermore, it is being supported by a credible, if opportunistic, theory of democratic pluralism. Pluralists like Brian Galligan, Campbell Sharman, and Harry Evans argue that dispersing power away from the executive reflects the diversity of modern society and the fragmentation of modern party politics better than does party duopoly in the lower house.
Having said something nice about Australia I want to end with my third topic, on a critical note, by addressing reform. I realise that I will be talking about matters which will be decided, in large part, by the referendum on an Australian republic later this year , but what I have to say should be relevant whatever the voters decide. If the republic is approved under the terms proposed by the 1998 Constitutional Convention and the government, Australia will enter the new century with a pre-modern Constitution that no informed person I have met on this visit to Australia thinks is adequate. They see it as a very poor choice, and those who propose to vote for the republic will do so in hopes that the Constitution can be improved later. The record of constitutional reform in Australia should not give them much hope in that regard. If the republic proposal is rejected, reformers will gather their forces to try again, but they may have a very long wait.
I want to deal primarily with the executive and what we in the United States call reconciliation, meaning reconciling differences between the two houses. These subjects are related and I think that reform in both of them could be directed quite consciously at perfecting and clarifying the parliamentary model that I described earlier. Doing this would, I think, eliminate unnecessary tension and uncertainty in the Australian political system, and would make some sections of the Constitution intelligible for the first time to the average citizen.
Constitutional literacy is very important in a democracy and for that reason I find the calculated constitutional obfuscation practiced here to be irresponsible.  At the 1897 Constitutional Convention, Joseph Carruthers pleaded with his colleagues. `Here we are framing a written constitution … It is better to let that Constitution clearly express what it is intended to effect; do not let us have to back it up by quoting whole pages of Dicey.'  Dicey, you may recall, explained in 1887 how constitutional law in Britain was, and still is, qualified by customary rules, called conventions. In law the monarch ruled but in practice, by convention not law, the executive was in the hands of a prime minister and cabinet selected from the majority in the House of Commons. This was how the Australian colonies, now states, arranged the executive in the nineteenth century, and it was how the Commonwealth Constitution arranged the executive in 1900. All but two of the central rules of parliamentary government, that ministers must sit in Parliament and financial bills must originate in the lower house, were left to convention, not law, in the new federation, as in the colonies. What is astonishing to an outside observer, and would surely have astonished Carruthers, is that the Constitution for the republic will be little better. The 1998 Constitutional Convention's recommendations for an Australian republic, which will be written into the constitution if voters approve a republic in November 1999, make no serious attempt to clarify the executive in the Constitution. We will still be left having to quote whole pages of Dicey to explain the revised Australian Constitution.
This indicates that Australia is making very heavy weather of codifying or redefining constitutional conventions, the rules that define the executive and relations between upper and lower houses. This criticism applies at both the state and federal levels, but I will concentrate on the latter. As you know, executive power in the Commonwealth is vested in the Crown. The prime minister and cabinet are not created by the Constitution and there is no requirement that the government have the support of a majority in the lower house, or that the Governor-General accept ministerial advice. Although the proposed republican Constitution will identify the Prime Minister and other ministers in Section 59, the Constitution will not create or define those offices or require the President to appoint ministers having the support of a majority in Parliament. That will be the prerogative of the President, as it is of the Governor-General. Furthermore, while the President will be obliged to accept the advice of ministers and the Federal Executive Council in some matters, this obligation will not extend to all matters and the Federal Executive Council in Section 59 is not the same institution, in law, as the Cabinet.
One is left to wonder why a country would go to the immense trouble of modernising its Constitution to become a republic on the centenary of its Constitution without using the opportunity to write the model of government it actually practices into constitutional law. What seems widely unappreciated here is that virtually every parliamentary country, other than Britain and the three old dominions, has committed the rules of parliamentary government to constitutional law quite successfully. Latvia did it, for example, in 1922. Ireland went most of the way that year, with no objection from the British government of the time, and finished the job in its 1937 Constitution. Germany and Japan received detailed parliamentary constitutions after World War II, and the post-1947 parliamentary countries of the British Commonwealth, in Asia, the Caribbean, and the Pacific, identify parliamentary government in constitutional law. William Dale, Legal Adviser to the Commonwealth Office from 1961 to 1966, describes how Whitehall lawyers drafted at least thirty-three independence constitutions. Most of the new states adopted the British model, and all of them wrote British constitutional conventions into law. `And this,' Dale observes, `from almost the only country in the world to be itself without a written constitution.'  And finally, the reactivated constitutions of the newly democratised states of Eastern Europe, Bulgaria, the Czech Republic, Estonia, Hungary, Lithuania, Poland, and the new Slovak Republic, similarly identify parliamentary government in law, although most of them practice the external executive variant.
As early as 1922, therefore, and many times since 1947, states have been able to write the conventions of parliamentary government into constitutional law quite successfully. There is nothing so mysterious, so subtle, so nuanced about conventions that they cannot be written into law, once their purpose is clear. There have been several attempts to correct this deficiency in Australia. Sir Henry Parkes tried, for example, in resolutions introduced into the National Australasian Convention in 1891, but was rebuffed.  In 1983 and 1985, following the debacle of 1975, the perambulating Constitutional Convention that ran from 1973 to 1985 tried again. It identified an extraordinary list of forty-three constitutional conventions, far more than is necessary and some of which it invented, which is not quite how conventions are supposed to grow. But it could not agree to place them in the Constitution and it resolved that they should simply be respected as conventions. 
The republic debate provided an opportunity to revisit the question but the ball was dropped again. The 1998 Constitutional Convention accepted the view of the Republic Advisory Committee of 1993, better known as the Turnbull Committee, that the Governor-General has two kinds of powers, ordinary and reserve. Ordinary powers are those that, by convention, are always exercised on the advice of ministers: issuing writs for elections, appointing judges and ministers, summoning Parliament, assenting to legislation, entering into treaties, and declaring war and peace. Reserve powers are those that are ordinarily exercised on the advice of ministers but which may, in unspecified but exceptional circumstances, be exercised without, or contrary to, ministerial advice. They include refusing the royal assent to bills, appointing and dismissing a prime minister, and dissolving, or refusing to dissolve Parliament. The Turnbull committee recommended that the Constitution should recognise that ordinary powers may only be exercised on the advice of a minister, but it reluctantly recommended that a republican President should retain the reserve powers.  This is what Prime Minister Paul Keating recommended to Parliament in 1995,  it is what the 1998 Constitutional Convention also recommended,  and it is what will be adopted if the republic is approved. Section 59 of the Constitution will read:
The President shall act on the advice of the Federal Executive Council, the Prime Minister or another Minister of State; but the President may exercise a power that was a reserve power of the Governor-General in accordance with the constitutional conventions that related to the exercise of that power by the Governor-General.
Think carefully about what this says. It requires the President to act on advice except in the case of unspecified reserve powers that shall be exercised in accordance with unspecified constitutional conventions. Bear in mind that the reserve powers are unspecified because, over and over again, in constitutional debates going back at least to the Federal Conventions of the 1890s, Australians have determined that they cannot be defined. Bear in mind, too, that constitutional conventions are by definition non-legal rules, and you will see how odd Section 59 is. The President, it says, will be obliged to accept advice, except when he or she decides not to, and in those cases, when exercising totally legal, if unspecified, reserve powers, the President shall be bound by non-legal rules. Given this calculated ambiguity, how is the High Court supposed to rule on a government petition to annul an act of the President which was contrary to ministerial advice and, in the government's view, either did not involve a reserve power or violated a constitutional convention? The court would be obliged to determine matters which a host of Australian authorities have deemed to be indeterminable. I know that the proposed Section 59 is the outcome of a difficult process of bargaining, in which there had to be give and take, but it is still a constitutionally absurd provision. The fact that Section 38A of the New South Wales Constitution is very similar does not make it any less so.
Why do Australians cling like leeches to constitutional conventions and reserve powers? Prime Minister Paul Keating offered three very conservative reasons when he presented his republican proposals to the House of Representatives in 1995. First, he said that conventions offer a constitution the flexibility with which to respond to changing circumstances. However, as the Turnbull committee report points out, the reserve powers are used so rarely in Australia that they cannot possibly support a theory of constitutional evolution.  For example, only one Commonwealth government has been dismissed by a Governor-General, in 1975, and experts still disagree over whether he acted properly. Furthermore, why do so few countries, none with a modern constitution, recognise this kind of flexibility as a valuable constitutional attribute?
Second, Keating argued that codification would open the reserve powers to judicial review, thereby involving the High Court in political disputes. This strikes me as an essentially monarchist argument. Republics have no difficulty subjecting executive acts to judicial review because only sovereigns, not presidents, are beyond the law. Furthermore, I see nothing in the proposed Section 59 to preclude judicial review should a government decide to challenge a presidential decision. Indeed, the notions of reserve powers and conventions would be specifically fixed in the Constitution for the first time, and hence subject to court interpretation.
Third, Keating argued that because the reserve powers will only be exercised in unpredictable circumstances, they cannot be codified with precision. This was repeated time and again by members of the Constitutional Convention in 1998, but it is simply false. What Australia calls reserve powers are either eliminated from, or dealt with quite specifically in, a great many constitutions. Heads of state in parliamentary republics have few discretionary powers, and those they do have are carefully prescribed by law, not convention.
I would also add that the circumstances which might trigger the use of the reserve power are not really as unpredictable as Keating suggested. Four contingencies are usually cited, all of which can be satisfactorily addressed in constitutional law. The first is that the head of state must have a reserve power to refuse his assent to an unconstitutional act of Parliament. Why does this have to be a reserve power, in the Australian sense? Why cannot a head of state be authorised to refuse assent to an unconstitutional act, particularly if that decision is reviewable? In Ireland, for example, the President may refer a bill she thinks unconstitutional to the Supreme Court, after consulting the Council of State, and she need only sign it if the court finds the bill constitutional. Irish Presidents have referred nine bills to the court since 1937.
The second contingency is that a reserve power is needed in case the government acts illegally. I presume that this means that the government asks the President to act illegally, because other illegal government acts can be challenged in the courts. In this case, I would argue that a President could safely be given constitutional authority to refuse illegal advice, and even to dismiss a government for cause, perhaps after taking advice from some statutory body, a Council of State, for example, as in Portugal.  This would be a political act but the final decision would always rest with Parliament, which has to find a new government, or with the electorate if there is a dissolution. Furthermore, in a republic there are always procedures to impeach a President who acts capriciously. Impeachment to remove a President is a traditional feature of republican constitutions, but Section 62 of the proposed Australian Constitution will give the power of dismissal to the Prime Minister alone, a very odd way to insulate the presidency from politics.
The third contingency is that there might be a deadlock in the process of government formation requiring the intervention of the head of state. My view is that a modern parliamentary constitution should eliminate the false notion that the use of a reserve power can make this problem go away, and almost all of them do. Some permit the head of state to act as a broker between political leaders, or to nominate someone, a `formateur' in the European sense, to act as a broker, or even, as a last resort, to name a prime minister for the lower house to approve. But there is no escaping the fact that there are really only two ways to resolve a deadlock in government formation in a parliamentary democracy, by political accommodation in the lower house or at the ballot box. The most a President can do is jump-start a process that has stalled, and this power can safely be provided in law.
There is no need for an unspecified reserve power for any these three contingencies. There arguably is such a need for a fourth contingency, the one that occurred in 1975. The upper house might refuse supply to the government and the head of state might decide to dismiss a prime minister who refused to resign or request a dissolution. Ever since the Australasian Federal Conventions of the 1890s, this has been the critical argument for reserve powers, and it was very evident in the 1998 debates too. However, I do not think that the facts of the 1975 crisis support a reserve power. I think they show instead that no head of state would ever want to act as Sir John Kerr did then because it would start another almighty political and constitutional row, which nobody wants. Furthermore, if we set 1975 aside, it is clear that in practice Australian politicians in the Commonwealth and the states have resolved the possible conflict between a government and the upper house over supply in favour of the government. They might argue over the details of the budget, but no government is going to be driven out of office by being denied enough money to govern.
My conclusion, as an outsider, is that Australians should accept this question as settled in practice and amend the state and Commonwealth constitutions in two respects. First, they should recognise that governments must have the confidence of the lower house only. Second, they should provide that an upper house may not destroy a government by blocking the supply of money for the ordinary annual services of government, which is to say, for the day to day operation of government under existing law. This has been proposed time and time again, in the states and the Commonwealth. It was recommended in one form, for example, by the 1988 Constitutional Commission, which argued that supply should not be rejected in the first three years of a four year Parliament,  but in Australia only New South Wales has settled the issue by a constitutional reform.
I recognise that it will be argued by some that the absolute key to the power of Australian upper houses is their power over supply. Without that, the whole estimates process in the Commonwealth Senate, for example, which is so important to the Senate's oversight role, would be toothless. The power to reject supply serves as a kind of nuclear deterrent; a powerful weapon, even if never used. I think this argument is problematical, but if it is accepted, one can still amend a constitution in such a way that a crisis of supply can be resolved very quickly, without the use of reserve powers, and not drawn out in turmoil. There might be, for example, an automatic dissolution of both houses if supply is rejected or delayed beyond some statutory period. This would expose both the government and every member of Parliament to a general election, something few politicians relish. This formulation would depart from a pure parliamentary model by making the government accountable to the upper house in this one case, but it would be preferable to the uncertainty that now surrounds supply.
While dealing with supply, Australians might also turn their attention to the issue of conflict between the two houses on ordinary bills, and I know many of you have. As I have said, none of the procedures for resolving deadlock work adequately, either in the states or the Commonwealth. Like the power to block supply, the double dissolution followed by a joint sitting of a Parliament, which is provided in the Commonwealth and some state constitutions, is used primarily as a threat not a constructive constitutional device, which is not the best use of a Constitution. Foreign experience does not offer much guidance in this matter, because the lower house can override the upper in most cases, but there are a few precedents that might help. In India, for example, Parliament can move to a joint sitting without a double dissolution in a case of deadlock, and I know this has been suggested here.
But you do not absolutely have to solve this problem in constitutional law because the Goods and Services Tax (GST) legislation of this year shows that a government can negotiate a deal if it wants one. What seems odd to me, however, is that after fifty years of proportional representation in the Senate and the states, Australian governments have still not internalised the art of negotiation. We hear, for example, that the Coalition and the Australian Democrats did not meet to discuss the GST until the government lost the support of the Independent, Senator Harradine, and hence its majority for the budget, in the Senate. Negotiation is dragged out of governments here like pulling teeth. The experience of parliaments in Europe is that a proportional representation election generally precedes a period of negotiation. It is not a prelude to a slanging match between people who need each other, which is the Australian way because political relationships are dominated by the customary confrontational behaviour of government and opposition in the lower house. The good news on this front, however, may be Prime Minister Howard's interview with the ABC after he and Senator Lees, the Australian Democrat leader, had agreed on the terms of the GST bill to be presented to the Senate. When his interviewer, Barrie Cassidy, pointed out that the Democrats in the Senate were not, after all, `a threat to democracy,' the Prime Minister agreed, and noted that `in a sense a bit of the political paradigm has been altered today.' 
If those responsible for constitutional reform had been able to codify the executive, eliminate the reserve powers, clarify the issue of supply, and possibly review conflict resolution procedures, I am sure they would have been able to give the Australian people what they really want in the republic referendum, even if they really should not, which is a popularly elected President. This would have been so because there would have been no significant powers for an elected President, claiming a popular mandate, to abuse. Australia could have had an Irish President, which would be no bad thing. About half the parliamentary republics in the world have popularly elected presidents and half do not, and it does not seem to make a dime's worth of difference, because every elected President operates within a framework of clearly understood, and very limited, powers. This was understood by many speakers in the 1998 Constitutional Convention, but the necessary reform of the Constitution to eliminate reserve powers eluded them. Instead, the 1998 Constitutional Convention agonised to create a presidential selection process that would produce a `non-political' President who would not abuse the reserve powers. And what a mess they made of it. The nominating committee the Convention recommended is not to be trusted to nominate the best person for the job because, in an indictment of all politicians, it will be obliged to nominate a political eunuch, someone who is a member of neither a political party nor a parliament. Then the Prime Minister will make a recommendation from the committee's short list, which will be confidential. This recommendation must be seconded by the Leader of the Opposition and approved by a two-thirds majority of both houses of Parliament.
This process will produce someone who is not associated with partisan politics, it is true, but the nomination and dismissal procedures for the President provide an extraordinary role for one of the most partisan people in the country, the Prime Minister. He or she will nominate half of the thirty-two person presidential nominating committee and the Prime Minister's supporters, those who form a majority in the House of Representatives, will appoint another three or four. And once the President is installed, the Prime Minister alone will be authorised to dismiss the President, without a stated cause. The only constitutional restraint on the Prime Minister is that within thirty days, the House of Representatives may vote against the dismissal, which would constitute a vote of no confidence in the Prime Minister. But that chamber is controlled by the Prime Minister's party or coalition, and even if you can imagine it censuring a Prime Minister, this act would not bring back the President automatically. He or she will have to be reappointed by the same process as the original election. All this strikes me as cumbersome and unnecessary. The only benefit for Australia is that the process will probably not produce an Australian version of Jessie `the Hulk' Ventura, the professional wrestler who was elected Governor of Minnesota in November 1998.
This discussion of the appointment and dismissal of the President reminds me that the Prime Minister is the one certain winner in the Australian constitutional debate. John Howard is a monarchist who would rather see no change to the Constitution, but if it happens, the enormously powerful position of the Prime Minister already present in the Australian monarchy will be confirmed. As I have argued above, parliamentary government has the effect of concentrating power in the centre, in the Cabinet and the Prime Minister. This has made for a particularly powerful government in Britain and the old dominions because the Prime Minister and Cabinet control the royal prerogatives through their advice to the head of state: the appointment of public servants, military officers, and judges, declarations of war, and the signing of treaties. Republican constitutions are not so generous to governments and some, at least, of these responsibilities are invariably shared with the legislature. This will not be the case in Australia. In their determination to rock as few boats as possible by adopting a minimalist approach to constitutional change, constitutional reformers have simply rolled over the prerogative from the Governor-General to the President, and that means that the extraordinary powers of Australian prime ministers and their ministerial colleagues have been confirmed. That, surely, is another opportunity missed.
Question — In the lower house, if the Premier or the Prime Minister was elected by the lower house—rather than the way it is now—would this change the power of the Prime Minister?
Alan Ward — It might. I think it has changed the power of the Chief Minister in the ACT because the election is secret. I think it might not, if it's public. If it's an open election, then the Prime Minister who is the leader of the majority party or coalition will always get elected. I don't think that, given the present party sanctions and the norms of party behaviour, an open election in Parliament is going to change the powers of the Prime Minister. I don't think in general that it would make very much difference.
What would make an enormous difference would be to codify the powers of the Prime Minister and codify the powers of the Government. I don't think any codified Constitution would give a republican Prime Minister the kinds of powers that you're proposing to give the Prime Minister here through his advice to the President, which are extraordinary, such as the appointment of judges and all sorts of public officials, declaring war and peace, and entering into treaties. Once you start to spell it out, its absurdity becomes evident. That is certainly Campbell Sharman's view. So one of the reasons why Australians don't want to spell out the government's powers is that, as soon as you start to codify, you realise that you can't give the Prime Minister all those royal prerogatives, which will become presidential prerogatives. So I think that starting to spell things out and open up the process will lead you down a different road.
Prime ministers do behave differently in the European model where the government is external to parliament. Parliament takes on some independence from government, even though the government has to have the support of the majority there. So prime ministers in, say, the Hungarian Parliament, do not enforce the same sort of discipline that you would find in a parliament where the Prime Minister and the ministers sit in, and are involved in the management of the place.
There are two kinds of selections of prime ministers. There is one where the Prime Minister is nominated or elected by the lower house—that's the Irish system and most European systems. The other is that the head of state selects someone who in his or her judgement has the support of the majority in the house. That's the monarchists' way of moving in a republican direction. I think that's eminently desirable.
Question — You mentioned party discipline and you mentioned South Australia a couple of times. I sat under a professor of political science and history in Adelaide—W.G.K. Duncan—and he was passionate on parliamentary democracy and, if I recollect correctly, one of his principles was a two party system. Could that not be your eighth principle?
Alan Ward — Not without narrowing the model to the point that it starts to exclude countries like Ireland. There are too many parliaments around the world which have some form of proportional representation in the lower house. Once you've accepted that, then you've accepted that two party politics is unlikely. You end up with the Irish system. I don't think there has been an Irish government for about twenty years which has not been a coalition government. I don't think it's two party politics once you've got proportional representation. So I wouldn't put that in the model because it restricts the model far too much.
Question — In your model would you include the present government of South Africa? And what would you do about Israel, where they're electing the Prime Minister?
Alan Ward —No, I wouldn't include South Africa, though I honestly don't know enough and I would have to take that question on notice.
Until recently Israel had a parliamentary system in which the lower house was elected by pure proportional representation, and the threshold was effectively one or one and a half percent of the vote, so it had lots of little parties. This meant that every government had to be a very tortuous coalition and in fact every government was dependent upon the votes of tiny religious parties in the end.
So what the Israelis in their wisdom decided to do was to get stronger government by electing the President—for example, as they do in France—where you allow anybody to run, but you end up with a run-off between the two leading candidates. It was thought that this would force the system into duopoly. But the actual result is that they elect the new President that way, but they continue to elect their Parliament the old way. They therefore elect a prime minister who has a majority of the popular vote, but his party does not have a majority in Parliament, and there is a clash between a presidential system and majority government. I know a few political scientists in Israel who called this lunacy.
They've now run two elections under this system—I don't think they'll run a third under it. It's very easy to amend the Israeli constitution by just an ordinary act of Parliament. They were already talking about amending it before the second election, and I suspect they will amend it before the next one, unless things work out reasonably smoothly. But the changes to the rules have not solved the problem of government formation thus far.
I wouldn't put the Israeli system in my model. To me, the Israeli system is sui generis, it's on it's own. It's like the Finnish system. No one else has it.
Question — I was pleased to hear you demolish Elaine Thompson's use of `Washminster', because I think this has been too easily accepted and repeated for a number of years. I would draw attention to the fact that in the 1950s, Percy Partridge said that any nation that thought the combination of responsible government and federalism would produce a new creature was quite wrong—that our government had developed very much on British lines. When you stressed that the Senate had roughly the powers of a colonial legislative council, you hit the nail upon the head because, although we copied the American Constitution a great deal, the point was that in the debates there were constant references to examples of the then upper houses, particularly South Australia.
Alan Ward — Yes.
Question — In Australia we have a three-level system of government, the federal, state and local government, with the exception of the ACT. Do you have any comments on the desirability of trying to reduce three levels to two levels?
Alan Ward — No, I don't see any need. I'm a pluralist who does not like excessive central control. But let me use this question to say something about federalism. One of the things I'm rather unhappy about here—and you see it described in Brian Galligan's work—is the way the levels of government interact. A lot of what goes on is not really federalism. It is dominated by a kind of state to federal government diplomacy, which is conducted secretly, covertly. You don't know really what's going on. You see the state and federal governments caucusing somewhere to sort things out, in ways that we don't see in the USA. I think this is, in large part, because the states have handed over too much of their power to the Commonwealth.
I think the decision to hand the federal government control of universities, for example, was a disastrous one. The states have also handed over their budget power. Now, because of a High Court decision on excise taxes, the Commonwealth controls just about all the taxing powers in the country. There's not much left but for the states and the federal government to bargain, in a kind of diplomatic way, for their shares. This is non-parliamentary and I don't like it very much. I don't know how you can reverse it, but you should try to stop it going any further.
* This paper was presented as a lecture in the Senate Occasional Lecture Series at Parliament House on 18 June 1999.
Gordon S. Reid, `Australia's Commonwealth Parliament and the “Westminster Model”', in Colin Hughes (ed.), Readings in Australian Government, Queensland University Press, St. Lucia, Qld, 1968, p. 109.
Elaine Thompson, `The “Washminster” Mutation', in Patrick Weller and Dean Jaensch (eds), Responsible Government in Australia, Drummond, Richmond, Vic., 1980, pp. 32–40. Thompson makes no mention of the Washminster model in her recent book chapter, `The Constitution', in Rodney Smith (ed.), Politics in Australia, Allen and Unwin, St. Leonards, NSW,1997, pp. 85–103.
Dean Jaensch, Parliament, Parties, and People: Australian Politics Today, Longman Cheshire, Melbourne, 1991, p. 214.
David Butler, The Canberra Model, Cheshire, Melbourne, 1973.
Spanish Constitution, Articles 99–100.
Thompson, `The “Washminster” Mutation', pp. 34, 36–7.
Australasian Federal Convention, Debates, Sydney, 15 September 1897, p. 552.
J.A. La Nauze, The Making of the Australian Constitution, Melbourne University Press, Melbourne, 1972, p. 41.
See Sir Samuel Griffith at the National Australasian Convention, Sydney, 1891.
Alan J. Ward, `The Constitution Review Group and the “Executive State” in Ireland', Administration, vol. 44, no. 4, winter 1996–97, pp. 60–61.
In Queensland, South Australia, Victoria, and Western Australia. See Alan J. Ward, `Responsible government and recent constitutional change in Australia and New Zealand', Adelaide Law Review, vol. 15, no. 2, p. 1.
Alan J. Ward, `Minority government and the redefinition of parliaments in Australia and New Zealand', Legislative Studies, vol. 11, no. 2, Autumn 1997, pp. 3–5.
Barbara Page, `Developments in the Legislative Council of New South Wales since 1978', Legislative Studies, vol. 5, no. 2, summer 1991, pp. 24–27.
Scott Bennett, Affairs of State: Politics in the Australian States and Territories, Allen and Unwin, Sydney, 1992, p. 71.
See the publication, Business of the Senate, published annually by the Table Office, Department of the Senate, Commonwealth Parliament.
F.M. Chaney, `Bicameralism Australian style: governing without control of the upper house', The Parliamentarian, vol. 69, no. 3, July 1988, p. 170.
Page, op. cit., pp. 28–29.
Neal Blewett, `Parliamentary reform: challenge for the House of Representatives', Australian Quarterly, vol. 65, no. 3, spring 1995, p. 12.
Bruce Stone, `Accountability reform in Australia: the WA Inc Royal Commission in context', Australian Quarterly, vol. 65, no. 2, winter 1993, pp. 25, 27.
Campbell Sharman addresses this issue in `The Senate and Good Government', Papers on Parliament, No. 33, May 1999, pp. 153–170.
Australasian Federal Convention, Debates, Adelaide, 19 April 1897, p. 913.
William Dale, `The making and remaking of Commonwealth constitutions', International and Comparative Law Quarterly, vol. 42, January 1993, p. 67.
La Nauze, op. cit., p. 38.
Australian Constitutional Convention, Adelaide, Proceedings, 1983, vol. 1, pp. 319–22, and Australian Constitutional Convention, Brisbane, Proceedings, 1985, pp. 7–45, 389–91.
Republic Advisory Committee, An Australian Republic, Canberra, 1993, vol. 2, pp. 241–95.
Sydney Morning Herald, 8 June 1995.
Report of the Constitutional Convention, Canberra, 2-13 February, 1998, vol. 1, p. 45.
Republic Advisory Committee, An Australian Republic, Canberra, 1993, vol. 1, p. 99.
Portuguese Constitution, Article 195.2.
Report of the Constitutional Convention, Canberra, 2-13 February, 1998, vol. 1, p. 17.
ABC transcript, 28 May 1999.