2. The constitutional design

Navigation: Previous PageContents | Next Page

Constitutions explain only a fraction of how democratic governments actually work, but they do provide the organizational and procedural framework for government action. There are two aspects of the Australian Constitution that make it particularly interesting. One is the way in which it attempts to combine responsible government with strong bicameralism. The other is the number of critically important provisions that cannot be found in the Constitution—or that can be found only by implication, and then only by those who know where to look and how to read between the lines.

A ‘Federal Commonwealth’

What is explicit in the Constitution is that Australia is a federal system. The preamble announces that ‘the people of New South Wales, Victoria, South Australia, Queensland, and Tasmania [later joined by Western Australia] ... have agreed to unite in one indissoluble Federal Commonwealth under the Crown of the United Kingdom of Great Britain and Ireland ... ’ The Commonwealth Constitution was the product of prolonged negotiations during the 1890s among representatives of colonies that had enjoyed self-government for decades and now were uniting voluntarily in a federation.

Not surprisingly, therefore, the powers of the Parliament, and consequently those of the Commonwealth, are enumerated in much the same manner as the legislative powers of the Congress are enumerated in the US Constitution. In addition, and unlike the American arrangements, sec. 51 of the Commonwealth Constitution authorizes one or more states to refer (or transfer) other matters to the Parliament in Canberra. The enumerated subjects on which the Parliament may legislate include:

[M]atters referred to the Parliament of the Commonwealth by the Parliament or Parliaments of any State or States, but so that the law shall extend only to States by whose Parliaments the matter is referred, or which afterwards adopt the law.

The Commonwealth Constitution also contains, in sec. 109, a provision comparable to the ‘Supremacy Clause’ of the US Constitution. Section 109 states that ‘When a law of a State is inconsistent with a law of the Commonwealth, the latter shall prevail, and the former shall, to the extent of the inconsistency, be invalid.’

Also like the American Constitution, room for expansion of Commonwealth power has been found in the Australian Constitution, perhaps in excess of what its authors had anticipated or would have approved. In the United States, it is found particularly in the ‘Commerce Clause,’ giving Congress the authority to regulate ‘Commerce with foreign Nations, and among the several States, and with the Indian Tribes,’ that has been interpreted to expand the reach of the federal government. In Australia, one place it is found is in the authority of Parliament to make laws respecting ‘external affairs.’ The High Court, exercising a power of constitutional interpretation much like that exercised by the US Supreme Court, has held that the Commonwealth Parliament may legislate to implement the terms of any valid treaty or other international agreement to which Australia is a party, even if the Parliament otherwise would lack the constitutional power to enact laws on the subject of that international compact.

In a well-known case, the Court upheld the Commonwealth’s authority to pass legislation preventing construction of a dam in Tasmania, a matter that otherwise would have been within the exclusive authority of that state, because the Commonwealth was acting to implement an international convention. The result is an open-ended opportunity for the federal government to expand its legislative jurisdiction at the expense of the states. Whenever the Commonwealth enters into an international obligation, it also receives the power to legislate in order to satisfy that obligation. (It should be mentioned that, in Australia, the government can enter into a treaty or other international agreement without the consent of the Parliament, including the Senate in which all states are represented equally.) A cynic might even imagine the possibility of the Commonwealth deciding to become a party to some treaty or international agreement primarily because of the added domestic legislative power that would accompany it.

Another provision of the Commonwealth Constitution probably has affected federal-state relations over an even broader array of issues. Sec. 96 authorizes the Parliament to ‘grant financial assistance to any State on such terms and conditions as the Parliament thinks fit.’ Under this authority, the Parliament makes grants available to states for purposes within the states’ jurisdiction, but sometimes these grants have been given only if the states met certain conditions. By this means, the Commonwealth has been able to influence policies that are beyond its constitutional purview by influencing how the states legislate with respect to those matters. The basis for the Commonwealth’s influence, obviously enough, is the states’ desire for the funds that they can receive only if state policies satisfy federal conditions.

The executive government and Parliament

Of greater interest for our purposes are the constitutional provisions establishing the executive and legislative institutions of the Commonwealth, assigning powers to them, and defining the relations among them. It is on these matters that the Constitution is remarkably incomplete and misleading, and deliberately so.

Anyone who read and believed chapter II, on ‘The Executive Government,’ would be bewildered by the practical operation of Australia’s government. Consider secs 61–64:

61. The executive power of the Commonwealth is vested in the Queen and is exercisable by the Governor-General as the Queen’s representative, and extends to the execution and maintenance of this Constitution, and of the laws of the Commonwealth.

62. There shall be a Federal Executive Council to advise the Governor-General in the government of the Commonwealth, and the members of the Council shall be chosen and summoned by the Governor-General and sworn as Executive Councillors, and shall hold office during his pleasure.

63. The provisions of this Constitution referring to the Governor-General in Council shall be construed as referring to the Governor-General acting with the advice of the Federal Executive Council.

64. The Governor-General may appoint officers to administer such departments of State of the Commonwealth as the Governor-General in Council may establish.
Such officers shall hold office during the pleasure of the Governor-General. They shall be members of the Federal Executive Council, and shall be the Queen’s Ministers of State for the Commonwealth.
After the first general election no Minister of State shall hold office for a longer period than three months unless he is or becomes a senator or a member of the House of Representatives.

From reading these provisions, we learn that Australia is indeed a monarchy. All executive power of the Commonwealth is vested in the Queen (and her successors) acting through her appointed agent, the Governor-General.[4] The Governor-General is advised by the Federal Executive Council and sometimes is required to seek the Council’s advice (but not its consent). However, he appoints the members of the Council and may dismiss any of them if and when he chooses. The Governor-General also determines the organization of the executive government by establishing ministries (‘departments of State’). He appoints ministers to head these departments from among members of the Executive Council, and the Governor-General may dismiss any minister just as he may remove any member from the Council itself. The only restriction on the Governor-General’s discretion in selecting ministers is that they must be (or within three months, must become) members of the Senate or the House of Representatives. However, this requirement applies only to ministers, not to all members of the Federal Executive Council.

Now consider what we have not learned from these provisions. If we relied on their plain meaning, we would not know that, in practice, the Governor-General exercises exceedingly little discretionary power (with some ill-defined reserve powers, such as the power that was at the heart of the 1975 crisis discussed in Chapter 4). We would not know that it is the majority party or coalition in the Parliament, or its leader, and certainly not the Governor-General, that selects the members of the Federal Executive Council, one of whom is designated the prime minister; that it is the prime minister, and certainly not the Governor-General, who decides which minister will head which departments; that all ministers hold their offices at the discretion of the prime minister or his party or coalition in the Parliament, and certainly not ‘during the pleasure of the Governor-General’; that the only active members of the Federal Executive Council are the Representatives and Senators selected by the current prime minister or his party caucus in the Parliament; and that the Governor-General is most unlikely to ignore the advice his ministers give him. As Brian Galligan (1980b: 266) has put it, ‘In normal circumstances ministers are not his advisers; they are his masters.If the Governor-General can do almost anything according to law, he can do virtually nothing according to convention.’

Nowhere does the Constitution mention the prime minister, the Cabinet, or the concept or practice of responsible government by which the prime minister and Cabinet continue in office only so long as they continue to enjoy the confidence of a majority of the Members of the House of Representatives. The only hint of such things is the requirement that each minister must be, or soon become, a member of the House or Senate. Instead, the cardinal principles of responsible government that Australia inherited from Great Britain, and to which it intended to adhere, are conventions. These conventions are shared understandings of what the Constitution really means, not what it actually says. As we shall see when we look briefly in Chapter 5 at the constitutional debates of the 1890s, some thought it was unnecessary to spell out intentions and expectations that were universally shared; others thought the conventions of responsible government were too subtle and nuanced to be captured adequately in flat assertions of constitutional text.[5] We shall return to the subject of conventions in Chapter 4 and again in Chapter 10.

What the Constitution has to say about the location and exercise of legislative power does little to cast doubt on the power of the monarch, acting through the Governor-General. The Australian Parliament comprises the monarch as well as both houses. The Governor-General summons Parliament to meet; he may prorogue it (thereby ending a parliamentary session and terminating all pending legislative business); and he may dissolve the House of Representatives (and under certain conditions, the Senate as well) before the expiration of the term for which its Members are elected. When Parliament passes a bill, the Governor-General may exercise a veto that Parliament cannot override, or he may propose his own amendments to the bill, or he may ‘reserve the law for the Queen’s pleasure.’ In the last case (in theory) the monarch has two years to decide whether to give her assent, just as she may, within one year, disallow any law to which the Governor-General has assented.

In short, the Governor-General’s legislative powers are nominally greater than those of the American President. Contrary to the American notion of separation of powers (or in Richard Neustadt’s more accurate formulation, separated institutions sharing powers), the Governor-General is an integral component of the Parliament. For example, the Parliament cannot even consider a spending proposal unless the Governor-General recommends it (sec. 56). He also has constitutional authorization to propose amendments to any bill that Parliament already has approved, unless he chooses instead to veto that bill absolutely (sec. 58).

Again, of course, there is little connection between these constitutional formalities and the operations of Australian government. The Governor-General summons and prorogues Parliament, and dissolves the House of Representatives, when the government asks him to do so. Likewise, when the Governor-General does propose amendments to bills that Parliament has sent him for his assent, they are the government’s amendments that he sends to Parliament House at the government’s request. Since 1901, Governors-General have returned with amendments a total of 14 bills, only three of them since 1948 (House of Representatives Practice 2001: 805).

So in the definition of the Commonwealth’s legislative and executive institutions, and in the allocation of legislative and executive powers between them, there is a striking disjunction between what the Constitution says and what it was intended and understood to mean.[6] As Kirby observes (2001: 593), ‘If one were to read the Australian Constitution, without knowledge of the conventions by which it operates, one could be forgiven for concluding that Australia was a kind of personal fiefdom of the British monarch [acting through her agent, the Governor-General].’ Yet notwithstanding the explicit terms of the Constitution, there is no question that its authors considered the conventions of cabinet responsibility and responsible government to be Australia’s great political inheritance from Great Britain, an inheritance that they fully intended to honour and continue.

The Senate and its powers

It also was understood that responsible government meant responsibility not to Parliament but to one-half of Parliament, the House of Representatives, just as in London it meant responsibility only to the House of Commons, not to the House of Lords as well. Just as in the United States in the 1780s, however, the agreement among the Australian states in the 1890s required the creation of a bicameral Parliament.

Like American Senators, most Australian Senators are elected for six-year terms, compared with the two-year terms of American Representatives and the maximum three-year terms for Members of Australia’s House of Representatives.[7] In both bodies, the terms of Senators are staggered. In the US, one-third of the Senate is elected every two years. In Australia half of the Senators usually are elected every three years at what are called half-Senate elections. At the request of the government, the Governor-General regularly dissolves the House before the end of its maximum three-year term of office; the Senate, by contrast, can be dissolved only in the case of a double dissolution (which is a constitutional possibility discussed below).

Also as in the United States, each state has the right to elect the same number of Senators, regardless of the differences in their populations. Each of the original Australian states (and so far there are no others) is guaranteed not two but a minimum of six Senators, a number that was increased to the current number of 12 by the Representation Act 1983. Furthermore, the Commonwealth Constitution of 1901 provided for direct popular election of Senators, a development that would not come to the United States until the US Constitution was amended in 1913. Finally, Australia’s Constitution includes what has become known as the ‘nexus’ provision of sec. 24: ‘the House of Representatives shall be composed of members directly chosen by the people of the Commonwealth, and the number of such members shall be, as nearly as practicable, twice the number of the senators.’ Thus, any increase in the membership of the House—to reflect population growth, for example—requires a corresponding increase in the membership of the Senate.

As we shall see, the nexus between the size of the House and that of the Senate gives the House an important advantage if and when the Constitution’s procedures for resolving legislative disagreements are invoked. On the other hand, members of the constitutional Conventions who supported the Senate’s influence could well have felt that the nexus was to be preferred to leaving the size of the houses to later legislation. It was reasonable to surmise that Parliament would enact legislation to increase the House’s membership in order to keep pace with Australia’s increasing population, but also that the House (and governments) would not have much incentive to support legislation making comparable increases in the membership of the Senate. In fact, in 1948 and again in 1983, when the number of Senators per state was increased, it was not because there was a felt need for more Senators. It was the size of the House that governments of the day wanted to expand, and increasing the size of the Senate was the constitutional cost of doing so.

In the Introduction, I referred to Lijphart’s concept of strong bicameralism, characterized by two chambers that are symmetrical, in that they have more or less comparable powers, but that are incongruent, in that they are selected in significantly different ways. I will defer discussion of how Australia’s House and Senate are elected and how their modes of election have changed, and focus here on the Senate’s constitutional powers, especially compared with those of the House of Representatives.

The controlling provisions are in sec. 53 of the Constitution which states that, ‘Except as provided in this section, the Senate shall have equal power with the House of Representatives in respect of all proposed laws.’ So the two houses are equal partners in the legislative process, with three exceptions relating, not surprisingly, to financial legislation:

Proposed laws appropriating revenue or moneys, or imposing taxation, shall not originate in the Senate.

The Senate may not amend proposed laws imposing taxation, or proposed laws appropriating revenue or money for the ordinary annual services of the Government.

The Senate may not amend any proposed law so as to increase any proposed charge or burden on the people.

As we might expect, the meaning of these prohibitions has required some interpretation and involved some negotiation over the years.[8] What, for instance, constitutes ‘the ordinary annual services of the Government’ or a ‘proposed charge or burden on the people’? We will touch on these questions later. For the moment, what is important is the general principle that financial legislation, both taxing and spending, is the primary responsibility of the House and, through it, the government.[9]

Emblematic of the government’s primacy in financial matters is sec. 56, which provides that ‘A vote, resolution, or proposed law for the appropriation of revenue or moneys shall not be passed unless the purpose of the appropriation has in the same session been recommended by message of the Governor-General to the House in which the proposal originated.’ In explanation, Moore (1910: 138D) argues that ‘It is an essential part of our Parliamentary system that every grant of money for the public service shall be based upon the request or recommendation of the Crown.’ He goes on to quote Erskine May that ‘The foundation for all Parliamentary taxation is its necessity for the public service as declared by the Crown through its Constitutional advisors.’

However, the effect of these restrictions on the Senate regarding financial legislation is mitigated by the provisions of secs 54 and 55, which are intended to prevent the House of Representatives from taking undue advantage of the prerogatives it enjoys under sec. 53. With regard to spending bills, sec. 54 requires that ‘The proposed law which appropriates revenue or moneys for the ordinary annual services of the Government’—a bill that the Senate cannot amend—‘shall deal only with such appropriations.’ This condition is primarily intended to protect against what is known in Canberra as ‘tacking’: including in the appropriation bill a non-appropriation provision (what in the Washington vernacular would be called a legislative ‘rider’) to prevent the Senate from being able to amend it.

With regard to revenue bills, sec. 55 provides that:

Laws imposing taxation shall deal only with the imposition of taxation, and any provision therein dealing with any other matter shall be of no effect.

Laws imposing taxation, except laws imposing duties of customs or of excise, shall deal with one subject of taxation only; but laws imposing duties of customs shall deal with duties of customs only, and laws imposing duties of excise shall deal with duties of excise only.

The first clause again protects the Senate against ‘tacking’—in this context, being presented with a bill containing non-tax provisions that the Senate cannot amend because they have been included in a tax bill. The second clause prevents the House from sending to the Senate a bill that deals with more than one aspect of Australia’s Commonwealth tax system, except that there can be omnibus customs bills and omnibus excise bills so long as those bills do not contain provisions on other subjects, tax-related or otherwise. To the Senate the Constitution says that initiating financial legislation is a prerogative of the House; to the House the Constitution says that it must not abuse its privileged position regarding that legislation.[10]

Even more important, the Senate is far from being powerless with respect to financial legislation. First, when the Senate cannot amend a bill from the House, it can request that the House agree to the amendments that the Senate would have made if sec. 53 did not prevent it from doing so:

The Senate may at any stage return to the House of Representatives any proposed law which the Senate may not amend, requesting, by message, the omission or amendment of any items or provisions therein. And the House of Representatives may, if it thinks fit, make any of such omissions or amendments, with or without modifications.

Thus, the Senate need not stand mute when it receives a spending or tax bill from the House. In fact, when the Senate agrees to request that the House make one or more amendments to such a bill, the Senate does so before the third reading of the bill (which marks its passage). So the two houses must dispose of the request in a mutually agreeable way before the bill reaches the third reading stage in the Senate, which it must do before it can become law.[11] In other words, the House may resist Senate requests for amendments, but the House cannot ignore them nor can it reject them summarily unless it is prepared to allow the bill to die. Second, even though the Senate cannot amend certain financial bills, it does not have to pass them, and it may reject them either by direct vote or by its refusal to bring them to a vote.[12]

These constitutional authorities that the Senate enjoys have led it to reject any notion that the House enjoys a general primacy over money-related bills.[13]

The provisions of section 53 are usually described as limitations on the power of the Senate in respect of financial legislation, but they are procedural limitations only, not substantive limitations on power, because the Senate can reject any bill and can decline to pass any bill until it is amended in the way the Senate requires. In particular, the distinction between an amendment and a request is purely procedural; in one case the Senate amends a bill itself, in the other it asks the House of Representatives to amend the bill. In both cases the bill is returned to the House of Representatives for its agreement with the proposed amendment. In the absence of agreement the Senate can decline to pass the bill.

The provisions of section 53 therefore have a purely procedural application, to determine whether amendments initiated by the Senate should take the form of amendments made by the Senate or requests to the House of Representatives to make amendments. The only effect of choosing a request instead of an amendment is that a bill makes an extra journey between the Senate and the House ... .

While appropriation bills and bills imposing taxation may not originate in the Senate, this does not mean that the Senate is not an equal partner with the House of Representatives in actually making appropriations. (Odgers’ Australian Senate Practice 2001: 292)

Not surprisingly, some commentators disagree. For example, Rydon (1985: 68) contends that ‘The Senate was made directly subordinate to the House in regard to money bills—which it could not originate or amend but could reject—and indirectly subordinate in all legislation through the provisions for the settlement of disputes between the houses.’

When the Commonwealth Constitution was written, the British House of Lords still enjoyed more than a suspensive veto over legislation; its veto power was limited by the Parliament Act 1911, which was enacted ten years after the first Commonwealth Parliament convened. Perhaps if Federation had come a decade later, the Australian Senate also might have been denied the power to block passage of tax and spending bills, not just to delay them and suggest amendments. Perhaps not, however. Colin Hughes quotes Redlich as having written (in his The Procedure of the House of Commons) in 1908, the year before the events that precipitated the 1911 law, that:

Amendment of the single money bill was constitutionally impossible. For two hundred years the House of Lords had ceased to claim any such right. In the face of the alternative presented to them, the Lords could do nothing else than accede to the aggregate of financial proposals without exception. They could not bring themselves to reject the whole financial scheme of the year. And so the matter ended. For more than a generation now the Commons’ right to sole management of the country’s finance has been asserted in this way; it is now both true in fact and accepted as a principle of constitutional law that the House of Lords is excluded from influence on money matters and it can never expect to reassert a claim to possess any. (Hughes 1980: 45)

The implication is that the authors of the Commonwealth Constitution surely would have been aware that, although the Lords had not (yet) been denied the power to amend or defeat supply bills, it was well-established that they did not do so. In addition, however, the American example was readily at hand; Bryce’s The American Commonwealth was popular reading at the time, though by no means the only source of information available to delegates about American constitutional arrangements and their practical operation. In any event, and as we shall see in Chapter 4, the Senate’s discretion with regard to money bills eventually gave rise to the greatest political and constitutional controversy in Australian history.

Pressing requests

There also has been an ongoing disagreement about how insistent on its requested amendments the Senate can and should be (Edwards 1943; House of Representatives Practice 2001: 433–438; Odgers’ Australian Senate Practice 2001: 325–327).

Both houses accept that the Senate may request that the House make amendments to money bills; sec. 53 leaves no doubt on that score. However, there have been disagreements about the interpretation and application of this section (House of Representatives Practice 2001: 428–433). As early as 1903, questions arose as to whether a particular Senate proposal could be made as an amendment or whether it needed to be embodied in a request. And as recently as 1995 and 1996, the two houses received committee reports on the appropriate interpretation of this section. The two reports, however, were less than compatible. Since then, ‘the preference in the House has been to avoid delaying the business of the Parliament with debates on the matter. On occasions when the Chair has drawn the attention of the House to Senate amendments where the position was unclear, the House has thought it appropriate not to take any objection.’ (House of Representatives Practice 2001: 431, 436–437)

This issue was at the heart of an early test of the Senate’s legislative strength, which took place barely a year after the Commonwealth Parliament was inaugurated in May 1901. In April of the following year, the House sent the Senate the Customs Tariff Bill, certainly the most contentious measure the Parliament had tackled to date (Souter 1988: 69–72). The Senate was constitutionally barred from amending the bill but not from recommending amendments and requesting that the House concur in them. After debating the bill for more than a month, the Senate requested 93 amendments. The House responded by accepting 33 of them, amending 11 others, and rejecting the remaining 49 Senate amendments.[14] The Senate then ‘pressed’ its request that the House concur in 26 of the 49 amendments that the House had rejected.

There was some uncertainty and disagreement about whether the Senate had exceeded its constitutional rights in pressing some of its amendments once the House had rejected them. The issue never has been resolved in principle. In 1902, Senator Symon argued for the Senate’s right to press a request:

Surely, when a person is given the power to make a request—unless the contrary is expressly stated—he is not debarred from civilly and courteously repeating it a second time. Power to request means to request as often as necessary till the request is granted ... (Commonwealth Parliamentary Debates, 9 September 1902: 15824)

However, the Attorney-General argued to the contrary in 1933, that ‘Repetition of the requests converts it into a demand’, and concluded that:

The Senate should recognize that the only practical way in which effect may be given to the words of the section which draw a distinction between making a request at any stage of a bill, and amending a bill, is by taking the view that a request can be made only once, and that, having made it, the Senate has exercised all the rights and privileges allowed by the Constitution. (Commonwealth Parliamentary Debates, 30 November 1933: 5249)

It is an interesting debate, the kind that constitutional scholars relish, but life and the work of the Parliament must go on. So in 1901, rather than risk delaying what was considered to be essential legislation, the House acceded to the Senate’s requests for some of the remaining amendments, and again refused to agree to others of them, but the House reserved the constitutional issue for another day:

Having regard to the fact that the public welfare demands the early enactment of a Federal tariff, and pending the adoption of Joint Standing Orders, the House of Representatives refrained from the determination of its constitutional rights or obligations in respect of the Senate’s Message of 3rd September, 1902, and resolved to receive and consider it forthwith. (Journals of the Senate, 4 September 1902: 545)

The Senate was not to be outdone. While agreeing to the House’s latest message, the Senate also approved a motion asserting that ‘the action of the House of Representatives in receiving and dealing with the reiterated requests of the Senate is in compliance with the undoubted constitutional position and rights of the Senate.’ (Journals of the Senate,9 September 1902: 552)

So it did not take long for the two houses to confront each other over a problem that at least some authors of the Constitution knew they had left embedded in it.[15] Much the same sequence of events took place in 1908, when the Senate requested 238 amendments to another customs tariff bill. Once more the House chose not to engage in a constitutional dispute, but instead stated that it was considering the Senate amendments without prejudice. The Senate responded with its assertion that the House simply was acting in recognition of the Senate’s constitutional powers.

By 1933, when the Parliament undertook a major tariff revision, the two houses evidently had reached an uncomfortable but mutually-understood modus vivendi on this matter.[16] Souter (1988: 294–295) reports that the Senate requested amendments to 47 of the 1800 tariff items in the bill. The House agreed to make 33 of the amendments, made seven others with modifications, and rejected the remaining seven of the requested amendments. When the Senate pressed three of the seven amendments—affecting rabbit traps, dates, and spray pumps—‘the House of Representatives responded in accordance with the unwritten rules of the game.’

After resolving that public interest demanded early enactment of the tariff, and carefully refraining ‘from the determination of its constitutional rights or obligations’, the House agreed to the pressed requests, with modifications. On receipt of this message the Senate resolved that the House’s dealing with its reiterated requests was ‘in compliance with the undoubted constitutional position and rights of the Senate’, and agreed to the Bill as amended.

The issue persists to the present, and the current state of play is aptly summarized in House of Representatives Practice (2001: 434): ‘There has been a difference of opinion as to the constitutionality of the action of the Senate in pressing requests. However, the House, while passing a preliminary resolution refraining from determining its constitutional rights or obligations, has on most occasions taken the Senate’s message into consideration.’[17] However the House is anxious to reject any implication (drawn by the Senate, for example) that it has, by usage, accepted the Senate’s right to press requests. Instead, House of Representatives Practice (2001: 436) quotes approvingly the observation that ‘a government has often been prepared to forfeit constitutional niceties for the sake of getting its legislation made,’ especially when the alternatives are to lose the bill or use it to begin satisfying the requirements for a double dissolution. In the 1933 case, one Member concluded ‘that the three items rabbit traps, spray pumps, and dates, however important they may be, hardly justify a double dissolution.’

House of Representatives Practice responds to a summary of the arguments advanced in Odgers’ Australian Senate Practice with a quotation from Quick and Garran’s seminal The Annotated Constitution to the effect that pressed requests have no constitutional standing. ‘A House which can make an amendment can insist on the amendment which it has made; but a House which can only “request” the other House to make amendments cannot insist upon anything.’ In their view, if the House decides not to make an amendment the Senate has requested, ‘the Senate must take the full responsibility of accepting or rejecting the bill as it stands.’ (Quick and Garran 1901: 672)

One of the other arguments offered in support of Quick and Garran’s position is that ‘the consequence of the opposite view [is] that the distinction between the power to request and the power to amend [is] merely formal.’ (House of Representatives Practice 2001: 435)[18] As we have seen, that is precisely the view that the Senate has taken. The discussion of this subject in Odgers’ Australian Senate Practice (2001: 327) concludes that:

Section 53 being ... a procedural section, prescribing procedural rules for the Houses to observe, it is for the Houses, in their transactions with each other, to interpret those rules by application. It is suggested that, in their dealings with Senate requests over the years, the Houses have supplied the required interpretation so far as the pressing of requests is concerned, and that interpretation is that requests may be pressed.

This is precisely the argument of agreement by usage that the House has been at pains to refute. Elsewhere, in insisting on the ‘Effective equality of the Senate and the House in the making of laws and the performance of all other parliamentary responsibilities’, Odgers’ Australian Senate Practice (2001: 3–4) notes simply that ‘The only qualification is that certain types of financial legislation must originate in the House of Representatives, and in some cases the Senate is limited to suggesting and, if necessary, insisting on amendments.’ (emphasis added)

In effect, the two houses have agreed to disagree.[19] Should the House ever decide to stand and fight on this ground, I expect that the ensuing battle would be bloody indeed.

Double dissolutions and joint sittings

When the Commonwealth Constitution was being designed, it required little imagination to anticipate that Parliament could encounter legislative deadlocks. At the 1897 Sydney Convention, Deakin stressed the powers of the Senate and the prospects for deadlock:

[W]e must take into account the different quality of these two houses, and the enormously greater power of resistance we are giving to the second chamber in this federal constitution, far greater than any second chamber possesses in our several colonies. It is on the broadest franchise. Representing the people in every sense of the term, that chamber will be a far more formidable opponent of the chamber of representatives than any [colonial or state] legislative council could possibly be. Under this constitution we are creating on the one side a senate and on the other side a house of representatives with its executive—and the executive is the important element in most of these considerations. We are creating in these two chambers, under our form of government, what you may term an irresistible force on the one side, and what may prove to be an immovable object on the other side, and the problem of what might happen if these two were brought into contact. (Convention Debates,[20] 15 September 1897: 582)

The Constitution’s provisions

To resolve such problems, the Constitution’s authors provided, as a last resort, an elaborate procedure that involves a ‘double dissolution’ of both houses of Parliament under sec. 57, which states in part that:

If the House of Representatives passes any proposed law, and the Senate rejects or fails to pass it, or passes it with amendments to which the House of Representatives will not agree, and if after an interval of three months the House of Representatives, in the same or the next session, again passes the proposed law with or without any amendments which have been made, suggested, or agreed to by the Senate, and the Senate rejects or fails to pass it, or passes it with amendments to which the House of Representatives will not agree, the Governor-General may dissolve the Senate and the House of Representatives simultaneously. But such dissolution shall not take place within six months before the date of the expiry of the House of Representatives by effluxion of time [i.e., within six months of the end of the three-year term for which Representatives are elected].

Then, if after the House and Senate elections following a double dissolution, the House passes the bill for a third time and the two houses still are unable to reach agreement on it, the Governor-General may convene a joint sitting of the two houses, also under provisions of sec. 57:

If after such dissolution the House of Representatives again passes the proposed law, with or without any amendments which have been made, suggested, or agreed to by the Senate, and the Senate rejects or fails to pass it, or passes it with amendments to which the House of Representatives will not agree, the Governor-General may convene a joint sitting of the members of the Senate and of the House of Representatives.

The members present at the joint sitting may deliberate and shall vote together upon the proposed law as last proposed by the House of Representatives, and upon amendments, if any, which have been made therein by one House and not agreed to by the other, and any such amendments which are affirmed by an absolute majority of the total number of the members of the Senate and House of Representatives shall be taken to have been carried, and if the proposed law, with the amendments, if any, so carried is affirmed by an absolute majority of the total number of the members of the Senate and House of Representatives, it shall be taken to have been duly passed by Houses of the Parliament, and shall be presented to the Governor-General for the Queen’s assent.

Thus, before Parliament can decide the ultimate fate of a bill at a joint sitting, first the two houses must reach a deadlock over it. This deadlock can arise if the Senate defeats a House-passed bill, or if the Senate fails to vote on passing it, or if the Senate passes the bill after making amendments to it (or requesting amendments in the case of a bill that the Senate is barred from amending) that are unacceptable to the House. Then, after an interval of at least three months following the point at which deadlock was reached, and whether during the same or the subsequent session of Parliament, the same process must be repeated with the same result.[21] The House again must pass the same proposal, with or without any amendments that the Senate had made or requested or to which the House had agreed before the first deadlock was reached; and the Senate again must defeat the proposal, fail to vote on passing it, or insist on amendments that the House refuses to accept.

Only after the House and Senate have reached a second deadlock over the same proposal may the Governor-General, acting at the request of the government, dissolve both houses simultaneously (a double dissolution), leading to new elections for all seats in both the House and the Senate.[22] After the new Parliament convenes following those elections, and if the same deadlock then occurs for a third time, the Governor-General may convene the two houses in a joint sitting. At this joint sitting, there are to be votes on the bill and on any amendments that one house has approved and the other has not. An absolute majority of the membership of both houses is required to approve any amendment and to pass the bill, if and as amended.[23]

It bears emphasizing that a joint sitting can consider only a bill that satisfies the requirements of sec. 57 and only those amendments to it that in the US Congress would be called ‘amendments in disagreement’—i.e., amendments that one house has proposed and that the other house has taken action on that constitutes an unwillingness to agree to them. Neither house can propose additional amendments at the joint sitting, nor may any compromises be proposed. The joint sitting may only choose among alternatives that already had been defined and considered by the two houses acting separately.[24]

Clearly, then, this procedure cannot be invoked quickly, and those who designed it cannot have expected that it would be used frequently.[25] In devising it, the Constitution’s authors could not look for inspiration to either America’s written constitutions or Britain’s constitutional conventions. The US Constitution requires bicameral differences to be resolved if a law is to be enacted, but it is silent on the procedures for doing so. And when the Australian Constitution was written, the British Parliament had no formal procedures for resolving the legislative deadlocks that could occur before passage of the Parliament Act 1911.

Odgers’ Australian Senate Practice (2001: 80) characterizes sec. 57 as ‘a concession of federalism to democracy’:

Provided that the whole process set out in section 57 is followed, the normal double majority for the passage of laws may be dispensed with, only for the legislation causing the deadlock, and laws may be passed in accordance with the wishes of the majority of the representatives of the people as a whole, if that majority is not too narrow. In cases of significant disagreement, democratic representation prevails over the geographically distributed representation of the people provided by the Senate.

If and when push finally comes to shove, the Constitution favours the ultimate legislative supremacy of the House of Representatives. In light of the ‘nexus’ requirement of sec. 24 that ‘the number of [Representatives] shall be, as nearly as practicable, twice the number of senators,’ the procedure for voting in joint sittings all but ensures that the House and, therefore, the government eventually can prevail in a legislative dispute with the Senate if each house is united in support of its position.[26] In a House of Representatives document intended to explain Parliament to the Australian public, double dissolutions are characterized as an opportunity for the voters to break the deadlock by changing the composition of the Senate to more closely conform with that of the House. ‘In effect, the legislation may be put to the people, presenting the electorate with the opportunity to change the composition of the Senate following a full Senate election.’ It is also noted, however, that ‘There is also, of course, the possibility of a change in the composition of the House—the deadlock may be broken in either direction.’[27]

In practice, however, any differences between the two houses that might emerge from the difference in their bases of representation or in their modes of election—both of which are discussed in the next chapter—have been overwhelmed by the strength of party discipline in both houses. The possibility of sec. 57 coming into play now depends almost entirely on whether the government enjoys majority control of the Senate. Party discipline now trumps any sense of obligation to support the position of one’s chamber. What matters is the voting strength of government and non-government forces in the two houses combined.

Four double dissolutions

In more than a century, there have been only six double dissolutions: in 1914, 1951, 1974, 1975, 1983, and 1987—but only one joint sitting to consider legislation—in 1974.[28] The events of 1974 and 1975 merit extended discussion in a later chapter. A summary of the causes and consequences of the other four double dissolutions will bring the double dissolution procedures to life and highlight some of the questions that have arisen in interpreting and implementing sec. 57.


As a result of the 1913 elections (for the entire House and half the Senate), the Liberal Government of Prime Minister Joseph Cook had a one-vote majority in the House but held only 7 of 36 seats in the Senate. The government found this situation untenable; new elections to both houses either would strengthen its position or put it out of its political misery.

To that end, according to Souter, the Government Preference Prohibition Bill

was introduced in October [1913] for the specific purpose of provoking a disagreement between the houses and in due course providing constitutional grounds for a dissolution of them both ... . By no stretch of the imagination was this [bill] central to the Cook Government’s programme; but it was certain to be rejected by the Senate a second time when re-submitted after an interval of three months. That would give [Prime Minister] Cook his grounds for going to the Governor-General. (Souter 1988: 133)

That the procedural requirements of sec. 57 were met was not in question. However, there was a dispute as to whether the bill giving rise to the deadlock justified a double dissolution. Should the Governor-General take into account the significance of the legislation in question in passing on the government’s request for a double dissolution? (Sawer 1956: 115–117, 121–124; Zines 1977: 218–222)

In his letter to the Governor-General requesting the double dissolution, the Prime Minister explained that the dearth of Liberal Senators ‘has for two successive sessions made the parliamentary machine unworkable’ (quoted in House of Representatives Practice 2001: 448), implying that the situation would not change until new elections took place. However, the Prime Minister did not contend that the fate of the Commonwealth hung on the fate of the bill in question. In fact, it was the uncontested insignificance of the bill that led the Senate to advise against granting the double dissolution.

The Senate expressed its position in an Address to the Governor-General, arguing in part that:

The Constitution deliberately created a House in which the States as such may be represented, and clothed this House with co-ordinate powers (save in the origination of Money Bills) with the Lower Chamber of the Legislature. These powers were given to the Senate in order that they might be used; but if a Senate may not reject or even amend any bill because a Government chooses to call it a ‘test’ bill, although such bill contains no vital principle or gives effect to no reform, the powers of the Senate are reduced to a nullity. We submit that no constitutional sanction can be found for the view, which is repugnant to one of the fundamental bases of the Constitution, viz., a Legislature of two Houses, clothed with equal powers, one representing the people as such, the other representing the States.

And we respectfully submit that the dissolution of the Senate ought not to follow upon a mere legitimate exercise of its functions under the Constitution, but only upon such action as makes responsible government impossible, e.g. the rejection of a measure embodying a principle of vital importance necessary in the public interest, creating an actual legislative dead-lock and preventing legislation upon which the Ministry was returned to power. (Journals of the Senate, 17 June 1914: 3)

The Chief Justice, Sir Samuel Griffith, took essentially the same position in his advice to the Governor-General, in which he argued that the power that sec. 57 gives to the Governor-General should be regarded as ‘an extraordinary power’:

to be exercised only in cases in which the Governor-General is personally satisfied, after independent consideration of the case, either that the proposed law as to which the Houses have differed in opinion is one of such public importance that it should be referred to the electors of the Commonwealth for immediate decision by means of a complete renewal of both Houses, or that there exists such a state of practical deadlock in legislation as can only be ended in that way. (quoted in Odgers’ Australian Senate Practice 2001: 88–89)

Both the Senate and the Chief Justice could find support for their position in an argument that had been made to those engaged in writing the Constitution by the Leader of the Convention, Edmund Barton:

‘[D]eadlock’ is not a term which is strictly applicable to any case except that in which the constitutional machine is prevented from properly working. I am in very grave doubt whether the term can be strictly applied to any case except a stoppage of the legislative machinery arising out of conflict upon the finances of the country. ... a stoppage which arises on any matter of ordinary legislation, because the two houses cannot come to an agreement at first, is not a thing which is properly designated by the term ‘deadlock’, because the working of the constitution goes on—the constitutional machine proceeds notwithstanding a disagreement. ... it is only when the fuel of the machine of government is withheld that the machine comes to a stop and that fuel is money. (Convention Debates, 15 September 1897: 620)

Notwithstanding such arguments, the Governor-General granted the double dissolution. In doing so, he made no reference to the legislation at issue or to the prospects for future legislation. Evidently he thought it unnecessary or inadvisable either to weigh those factors or to acknowledge what part, if any, they played in his decision. Furthermore, he did not address how much discretion a Governor-General should exercise in deciding whether or not to grant a requested double dissolution. In contrast to the opinion of the Chief Justice, quoted above, that the Governor-General should give his ‘independent consideration’ to the importance of the bill or the parliamentary situation more generally, the Prime Minister had asserted that the Governor-General’s discretion under sec. 57 ‘can only be exercised by him in accordance with the advice of his Ministers representing a majority in the House of Representatives’ (quoted in House of Representatives Practice 2001: 448), implying that it also would be inappropriate for the Governor-General to declare a double dissolution unless advised to do so.

The Cook Government was defeated at the ensuing elections, so the bill died and no joint sitting took place. However, the precedent had been established ‘that sufficient cause for double dissolution could be deliberately engineered.’ (Souter 1988: 137) Subsequent prime ministers have stressed the significance of the legislation giving rise to their requests for double dissolutions and, as Cook had, the likelihood that similar problems would arise again if the composition of the Parliament remained unchanged. On occasion, governors-general have referred to such considerations in announcing double dissolutions. However, no Governor-General has refused to grant a double dissolution that the government of the day has requested if the requirements of sec. 57 have been satisfied.


It was not until 37 years later that the next double dissolution occurred. When it did, it was under different political circumstances and it raised a different issue about the application of sec. 57 (Whitington 1969: 152–159).

In 1950, the Menzies Government, comprising a coalition of the Liberal and Country parties, held a 74–48 majority in the House (with one Independent) but was in the minority, 26–34, in the Senate. In May of that year, the House passed the Commonwealth Bank Bill. In June, the Senate passed it with amendments, but the House disagreed with the Senate amendments and asked the Senate to reconsider them. Instead of withdrawing its amendments, the Senate insisted on them. In response, the House insisted on its disagreement to the amendments and the Senate then reaffirmed its insistence on them. At that point, the House failed to take further action. Instead, and a week before the House received a message of the Senate’s final action, an identical bill was introduced in the House. The House passed this second bill on the same day in October on which it received the Senate’s message of its final action on the first bill. In the Senate, the second bill was referred to a select committee with instructions to report in four weeks. Several days later (and well before the four-week period expired), the Prime Minister requested a double dissolution, which the Governor-General proceeded to grant. (For the chronology of events, see Odgers’ Australian Senate Practice 2001: 90–94, and House of Representatives Practice 2001: 449–450.)

In Menzies’ advice to the Governor-General, the Prime Minister addressed the basis for his request and justified the need for a double dissolution less in terms of the specific bill at issue than in terms of the more general situation in Parliament:

... the Government, with a new mandate from the people, has been in major affairs, constantly delayed and frustrated by the facts that the two Houses are of opposite political complexions and that in consequence the legislative machine, except in respect of relatively minor matters, has been materially slowed down and rendered extremely uncertain its operation.

Under these circumstances, if the only condition upon which a Double Dissolution could be granted was, broadly expressed, that a serious conflict between the two Houses ought to be ended by the votes of the electors, then I would have no doubt whatever that as Prime Minister I should be more than justified in asking you to take the necessary steps to have determined by those electors a disagreement which tends so strongly against the giving of prompt expression to the public will. (quoted in Nethercote 1999: 12)

To appreciate the reason for Menzies’ argument, it helps to understand that, as we shall discuss in the next chapter, these events occurred during the first Parliament after enactment in 1948 of the law that provided for Senators to be elected thereafter by proportional representation. One reason that the Labor Government of the day had proposed the change was to ensure that it would retain a majority in the Senate if, as expected, it lost control of the House, as it did, to Menzies and the Coalition. In 1950, consequently, the Coalition Government was, for one of the very few times in the Federation’s first half-century, faced with a Senate that it did not control. Securing a double dissolution, therefore, gave Menzies and his Government the opportunity to gain control of the Senate while retaining control of the House of Representatives. With these same possibilities in mind, the Australian Labor Party (ALP), which did control the Senate, had to think twice before creating the grounds for a double dissolution and an election that might leave it in the minority in both houses.

Although the ALP platform called for abolition of the Senate, the tactical value of the upper house was undeniable at times like the present. But careful judgment was required as to how that advantage could be used against the Government without provoking a double dissolution election at which Labor was likely to be savaged again. Some unpalatable measures would therefore be allowed through the Senate ... (Souter 1988: 411)

The Commonwealth Bank Bill, however, was not allowed through, and Labor’s worst fears were realized. At the ensuing election, the Liberals were returned with majorities in both houses and the ALP was banished to the political wilderness. With respect to the banking bill, no third deadlock occurred, no joint sitting was necessary, and a different bill on the same subject subsequently became law.

In connection with these events, the question arose as to whether the Senate’s decision to refer the second bill to a select committee constituted a ‘failure to pass’ it within the meaning of sec. 57. As the Solicitor-General argued at the time (quoted in House of Representatives Practice 2001: 451–452), ‘The expression “fails to pass” is clearly not the same as the neutral expression “does not pass”, which would perhaps imply mere lapse of time.’ So ‘Perhaps the principle involved can be expressed by saying that the adoption of Parliamentary procedures for the purpose of avoiding the formal registering of the Senate’s clear disagreement with a Bill may constitute a ‘failure to pass’ within the meaning of the section.’ That was precisely the Prime Minister’s contention. Menzies argued that the Senate had demonstrated sufficiently its intent to procrastinate so that its inaction constituted conclusive evidence of its determination not to pass the bill:

[T]here is clear evidence that the design and intention of the Senate in relation to this Bill has been to seek every opportunity for delay, upon the principle that protracted postponement may be in some political circumstances almost as efficacious, though not so dangerous, as straight-out rejection. Since failure to pass is, in section 57, distinguished from rejection or unacceptable amendment, it must refer, among other things, to such a delay in passing the Bill or such a delaying intention as would amount to an expression of unwillingness to pass it. (quoted in House of Representatives Practice 2001: 450)

When the Senate rejects a bill, its ‘failure to pass’ it is obvious. But when the Senate either takes no action or takes some other action, such as referring a bill to a select committee, it becomes more a matter of judgment as to whether the ‘failure to pass’ requirement has been met. In this case, the Senate averred that referring the bill to a committee did not imply an unwillingness to consider the bill further, or even to pass it. However, the Governor-General granted the double dissolution, as the government had requested. So the government’s arguments prevailed in practice, and the High Court did not have occasion to rule on their merits.


There were other bills on which the two houses had disagreed in 1950–1951, but the government did not seek to have any of the others satisfy the requirements of sec. 57 so that they could have been eligible for consideration if there had been a joint sitting following the 1951 double dissolution and elections. In 1983, Parliament confronted, albeit in a different form, the issues that had arisen in connection with the 1914 and 1951 double dissolutions, as well as additional questions surrounding a double dissolution that involved multiple bills.

The 1980 elections had produced a House in which the Fraser Liberal-National Party coalition had an 82–66 majority, but was narrowly in the minority in the Senate. As a result of legislative actions and inactions beginning in August 1981, the government requested a double dissolution in February 1983. In so doing, the government asserted that a total of 13 bills had completed the procedural stages laid out in sec. 57 and so might become eligible for consideration at a joint sitting if one were to take place after the intervening election (House of Representatives Practice 2001: 461–463).

Of particular interest were nine of the bills that were Sales Tax Amendment bills that the Senate could not amend. Instead, the Senate had requested amendments that the House had resolved not to make. ‘The Senate considered the House’s position and declined to pass a resolution “that the requests be not pressed”, the effect of which was to press the requests’ (Odgers’ Australian Senate Practice 2001: 109), an action that, the government argued, constituted a ‘failure to pass’.

In dissolving both houses, the Governor-General took note of the Prime Minister’s assertions regarding the importance of the bills in question and the implications of the deadlocks for the ability of the sitting Parliament to function effectively in the future. However, the Governor-General was unwilling to grant the double dissolution when the government first requested it, asking instead for additional evidence that Parliament had in fact become ‘unworkable’ and that there was no effective alternative to the double dissolution. The government was able to satisfy the Governor-General on this score. However, the Governor-General’s request and the government’s compliance with it strengthened the contention that the Governor-General can and even should make an independent determination as to whether requests for double dissolutions should be granted.

For Uhr, there were cautionary lessons to be drawn from this incident by both the government and Opposition. It implied that there were limits on the ability of an Opposition-controlled Senate to force a double dissolution and new elections,[29] though no one in Canberra seems anxious to test those limits after having experienced the events of 1975, which we will review in Chapter 4. What may prove more important in practice is a message to governments not to assume that they can artificially create the basis for a double dissolution by passing one or more non-money bills that they know the Senate will not accept, and do so primarily for the purpose of being able to achieve a double dissolution at a subsequent time of the government’s choosing—that is, whenever obstruction or opposition in the Senate becomes too inconvenient. However, the issue has yet to arise again (it was not an issue in 1987), so we cannot know whether a future Governor-General will be prepared to refuse a government’s request for a double dissolution when there is no alternate government available to replace it.

The elections replaced Fraser’s Liberal-National Government with an ALP majority of 75–50 in the House and a plurality of 30–28 over the Coalition in the Senate, with five Senate seats in other hands. Consequently, the new government did not pursue passage of the bills in question and no joint sitting was convened.

Two other issues arose in connection with this double dissolution, issues on which the two houses evidently do not see eye-to-eye to this day. One was what Odgers’ Australian Senate Practice (2001: 110), more than a quarter-century later, calls the ‘stockpiling’ of bills in anticipation of a double dissolution so that they might be salvaged by passage in a joint sitting. The author editorializes that, ‘At least in circumstances where there is no withholding of supply by the Senate, such a use of stockpiled bills, perhaps stale and unrelated to a particular situation, does not appear to be within the intent of section 57 of the Constitution.’ This position is not surprising since this practice so obviously works to the advantage of the government and the House it controls, and to the corresponding disadvantage of the Senate.

The second issue was whether the two houses had reached the required impasse on the sales tax bills—the House having decided not to make the requested Senate amendments and the Senate having decided not to not press them. The House did not address this question directly; instead, it took the position that the Senate should not have pressed its requests in the first place. When the House received the message relating that the Senate had done so:

Mr. Speaker made a statement on the constitutional issues involved, noting that the right of the Senate to repeat and thereby press or insist on a request for an amendment had never been accepted by the House. The House then agreed to a resolution inter alia endorsing the statement of the Speaker in relation to the constitutional questions raised by the Senate message and declining to consider the message in so far as it purported to press amendments contained in the earlier message. (House of Representatives Practice 2001: 461)

The Senate’s authoritative treatise on its procedures emphasizes instead that ‘the initial parliamentary consideration of these bills ended in the House, not the Senate,’ and argues that ‘The fault lay with the House in deliberately and wrongly breaking off communication with the Senate and shelving the bills.’ (Odgers’ Australian Senate Practice 2001: 111)

Neither issue was adjudicated because the bills died with the defeat of the Fraser Government at the 1983 elections. Should either issue arise again, the differing positions of the two houses, which seem to have persisted for so long, might well be argued again.


Four years later, there was no doubt that the House had twice passed the Hawke Government’s Australia Card Bill 1986 and that the Senate had twice rejected it by refusing second reading (Sugita 1997: 163-166). The elections that followed the double dissolution left the political complexion of Parliament essentially unchanged: the government was in a solid majority in the House and in a solid minority in the Senate. The government did have enough votes to prevail in a joint sitting. In preparation for a joint sitting to pass the bill, therefore, the House passed it for a third time.

During Senate debate, however, a convincing argument was made that implementation of the bill, if enacted, would require regulations that the Senate, acting unilaterally, without the concurrence of the House, could vote to disallow. Furthermore, an equally compelling argument was made that the Senate would do just that, given the non-government majority in the Senate. At the government’s instigation, therefore, the Senate eventually took action on the bill that surely constituted ‘failure to pass’. But then, instead of requesting a joint sitting, the government let the bill die. It knew that it could anticipate victory in a joint sitting, but that its victory would be fruitless because of the likelihood (or virtual certainty) that the Senate would veto the necessary implementing regulations. Also, it was too late to amend the bill in a way that would have circumvented this problem because sec. 57 permits a joint sitting to vote only on the bill and any amendments that one house or the other already has passed (and the other has not accepted).

Implications and interpretations

As we shall discover in Chapter 5, the Constitution’s authors laboured long and hard to decide whether to include provisions to resolve legislative deadlocks and, if so, how to design those provisions. Yet there was only one double dissolution in the first half-century of Federation, and a total of only six in a century. Why?

Double dissolutions rarely have been necessary because governments almost always have had enough votes in the Senate to see their legislation enacted. As we shall see in the next chapter, governments usually had majorities in the Senate from the formation of the party system until the mid-1950s. Even when governments have faced Opposition majorities or, in recent decades, non-government majorities, non-government Senators have been reluctant to press their legislative powers out of a combination of respect for the principles of responsible government as well as a desire to avoid having to face the electorate before the natural expiration of their six-year terms. Furthermore, governments have had at least two reasons for preferring to reach compromises with the Senate rather than remaining adamant and resorting to double dissolutions: first, a recognition that Senate amendments often have improved government legislation, and have even been made by the Senate at the government’s initiative or with its support or acquiescence; and second, a calculation that compromise with the Senate is preferable to the risk of a new election at which its own majority in the House would be at risk.[30]

Of the four double dissolutions we have just reviewed, none led to a joint sitting and none led to enactment of the specific bill in question. In 1914 and 1983, the elections brought the defeat of the sitting government and, therefore, the demise of the legislation at issue. Governments must exercise caution in invoking sec. 57; double dissolutions and the elections that follow involve risks as well as potential rewards. In 1987, the Hawke Government, which was returned to office, did not pursue the bill that led to the double dissolution when it concluded that doing so ultimately would prove futile. In 1951, the Menzies Government, which also remained in office, dropped the specific bill in favour of other legislation on the same subject (though if the government had been determined to enact the same bill that gave rise to the double dissolution, presumably it could have done so after having won control of the Senate).

As we have seen, these four double dissolutions triggered several disagreements about how sec. 57 is to be interpreted and applied. One issue concerns what constitutes the Senate’s failure to pass a bill, which is an essential ingredient of deadlock. In 1951, it was established that the Senate did not have to defeat a bill in order for that bill to qualify under sec. 57. But uncertainty remains about what other Senate actions (such as referring a bill to a select committee) do satisfy the constitutional requirement. In Chapter 4, we will examine another double dissolution that occurred in 1974. In that context, the High Court ruled that the Senate had not ‘failed to pass’ a bill when, on the same day in December that it received the bill from the House, it voted to adjourn debate on it until the first sitting day in February of the following year. ‘The Senate has a duty to properly consider all Bills and cannot be said to have failed to pass a Bill because it was not passed at the first available opportunity; a reasonable time must be allowed.’ (Victoria v Commonwealth 1975 7 ALR 1, quoted in Odgers’ Australian Senate Practice 2001: 82) But what constitutes ‘a reasonable time’?

In the same decision, the Chief Justice commented that when the Senate has amended a House bill, the equivalent of what in congressional parlance is known as the ‘stage of disagreement’ should be reached before the ‘failure to pass’ threshold has been crossed. In other words, it is not sufficient for the Senate to have amended a bill it has received from the House. The House should disagree to the amendments, and the Senate should insist on its amendments instead of receding from them at the House’s request. Only then can it be said that deadlock has occurred; only then should the three-month clock begin to run. Although this issue was not before the High Court in the 1975 case, the Chief Justice’s comments (quoted in Odgers’ Australian Senate Practice 2001: 87) still are instructive:

At the least, the attitude of the House of Representatives to the amendments must be decided and, I would think, must be made known before the interval of three months could begin. But the House of Representatives, having indicated in messages to the Senate why it will not agree, may of course find that the Senate concurs in its view so expressed, or there may be some modification thereafter of the amendments made by the Senate which in due course may be acceptable to the House of Representatives. It cannot be said, in my opinion, that there are amendments to which the House of Representatives will not agree until the processes which parliamentary procedure provides have been explored. (emphasis in original)

The same reasoning could be applied to determining when there is a second deadlock for the purpose of declaring a double dissolution and then a third deadlock for the purpose of convening a joint sitting. The question which the Chief Justice suggests but does not address is whether the processes to which he referred must be exhausted, or whether it suffices for each house to have made known its rejection of the position taken by the other with respect to the Senate’s amendments. Unsettled questions remain.

We also have seen how governments can provoke double dissolutions, or control when they take place, for their political advantage. In requesting a double dissolution, the government of the day naturally emphasizes the importance of the bill or bills that have been blocked by the Senate’s ‘failure to pass’ (Odgers’ Australian Senate Practice 2001: 84). However, this is done for political, not constitutional, reasons. As noted above, the first double dissolution, in 1914, was the result of a deadlock that the government deliberately created over relatively minor legislation when, according to Prime Minister Cook, it had become ‘abundantly clear’ that the Opposition had taken control of the Senate. Cook explained that the government then ‘decided that a further appeal to the people should be made by means of a double dissolution, and accordingly set about forcing through the two short measures for the purpose of fulfilling the terms of the Constitution.’ (quoted in Odgers’ Australian Senate Practice 2001: 83)

Although any government will deny that it would even think of requiring a new election solely for political reasons, the Senate still must recognize that its failure to pass any bill twice might be used by the government as grounds for calling new Senate and House elections. Moreover, once the requirements for a double dissolution have been satisfied, it falls to the government to decide if and when the Governor-General declares the double dissolution, which gives the government the flexibility to choose a politically advantageous moment. Sec. 57 states that the Governor-General ‘may,’ not ‘must,’ declare a double dissolution, leaving open the possibility that he or she could reject a government’s advice to do so. In practice, however, I think it quite unlikely, in the foreseeable future and especially in light of the events of 1975, that a Governor-General would exercise this discretion and thereby enmesh himself in a highly charged partisan political controversy.

The use of double dissolutions for electoral advantage at propitious moments is linked to another application of sec. 57 that has inspired controversy: basing a double dissolution on the Senate’s ‘failure to pass’ more than one bill (Zines 1977: 222–224). In 1983, as many as 13 bills were said to have satisfied the requirements of sec. 57. So if a government waits until each of two or more bills has twice reached deadlock, and then calls for a double dissolution, each of those bills then is eligible for consideration and passage at a subsequent joint sitting (assuming a third, post-election, deadlock also occurs) at which the position of the House and the government is likely to prevail. According to the High Court, ‘a joint sitting of both Houses of Parliament convened under s. 57 may deliberate and vote upon any number of proposed laws in respect of which the requirements of s. 57 have been fulfilled.’ One Justice put it nicely: ‘One instance of a double rejection suffices but if there be more than one it merely means that there is a multiplicity of grounds for a double dissolution, rather than grounds for a multiplicity of double dissolutions.’ (Cormack v Cope 1974 131 CLR 432, quoted in Odgers’ Australian Senate Practice 2001: 83)[31]

The Senate has objected, especially because of the opportunities governments may be able to create that enable them to ‘stockpile’ bills in order to trigger a double dissolution, even if many months, or even years, later. If the House can construct any bill that the Senate is certain to reject, and reject again, it gives the government the ability to secure a double dissolution, not just a dissolution of the House, whenever it chooses and regardless of the merits or importance of the bill.

More generally, the ways in which sec. 57 has been interpreted and applied has caused the Senate heartburn for several reasons. The criticisms and suggestions made on behalf of the Senate in Odgers’ Australian Senate Practice 2001: 117 deserve quotation at length:

Section 57 of the Constitution was intended to provide a mechanism for resolving deadlocks between the two Houses in relation to important legislation. By judicial interpretation, and by the misuse of the section by prime ministers over the years, it now appears that simultaneous dissolutions can be sought in respect of any number of bills; that there is no time limit on the seeking of simultaneous dissolutions after a bill has failed to pass for the second time; that a ministry can build up a ‘storehouse’ of bills for simultaneous dissolutions; that the ministry which requests simultaneous dissolutions does not have to be the same ministry whose legislative measures have been rejected or delayed by the Senate; that virtually any action by the Senate other than passage of a measure may be interpreted as a failure to pass the measure, at least for the purposes of the dissolutions; and that the ministry does not need to have any intention to proceed with the measures which are the subject of the supposed deadlock after the elections. By putting up a bill which is certain of rejection by the Senate on two occasions, a ministry, early in its life, can thus give itself the option of simultaneous dissolutions as an alternative to an early election of the House of Representatives. This gives a government a de facto power of dissolution over the Senate which it was never intended to have, and greatly increases the possibility of executive domination of the Senate as well as of the House of Representatives. (emphasis added)

Consideration should be given to a reform of section 57 to restrict the power of a ministry to go to simultaneous dissolutions as a matter of political convenience. In order to restrict section 57 to its intended purpose, a limitation should be placed on the number of measures which may be the subject of a request for dissolutions, time limits should be placed upon such dissolutions in relation to the rejection of the measures in question, and a prime minister should be required to certify that the measures in question are essential for the ministry to carry on and that it is the intention of the ministry to proceed with the measures should it remain in office, and the Governor-General should be required to be satisfied independently as to those matters.

Although a double dissolution puts members of both houses—and, of course, the government—at risk, Uhr suggests that governments may calculate that the elections following a double dissolution may well be worth that risk. Referring to the 1951 double dissolution, he argues (1992: 100) that ‘it introduced into the armoury of prime ministers the threat of a double dissolution in parliamentary circumstances judged as ‘unworkable’ by an ambitious executive.’[32]

So the House of Representatives can try to confront Senators with the choice between capitulation—approving government legislation that a majority of them may oppose—and double dissolution—facing the electorate well before the expiration of their six-year terms of office. There is a certain irony to this argument, as we shall see, because at the heart of the events of 1974 and 1975 were attempts by Senators to use their authority to ‘fail to pass’ government legislation in order to force Representatives to face the electorate before the expiration of the terms for which they had been elected. Before turning to those events, however, we first need to review the party and electoral systems that have done so much to shape the relations between the Parliament and the government and, within the Parliament, between the House of Representatives and the Senate.

Navigation: Previous PageContents | Next Page

Back to top