Michael Sloane

The two largest mainland Commonwealth territories, the Northern Territory and the Australian Capital Territory (ACT), have both enjoyed representation in the Senate since 1975. The 40th anniversary of their first election falls in December 2015. The legitimacy of the presence of territory representatives in the Senate—often characterised as a ‘states’ house’—no longer attracts much comment. However, the legislation to enable this representation was the subject of great rancour, only passing the Commonwealth Parliament following affirmation at a joint sitting of the two houses and subsequently surviving two High Court challenges.

The extraordinary means required to pass this legislation in 1974 can in part be attributed to the fine balance of Senate numbers and the tense political atmosphere between the major parties at the time. However, it must also be recognised that the introduction of territory representation to the Senate was felt by many to endanger the ‘federal balance’ of the Constitution in a way that the earlier introduction of territory members in the House of Representatives did not. The Northern Territory gained representation, in a limited form, in the House of Representatives in 1922, as did the ACT in 1948. Although voting restrictions were initially imposed on these representatives, these restrictions were gradually removed.

At the heart of the controversy over the legitimacy of allowing territory representation in the Senate, leaving aside allegations of potential party-political advantage, lay popular conceptions of the Senate as designed to protect the interests of the original six states, in particular the interests of the less populous states against those of the more populous states. Under this view, it was thought that the presence of non-state representatives in the Senate would undermine this function. However, the actual provisions of the Constitution concerning the composition of the Senate are ambiguous, appearing both to limit representation in the Senate to the states while also providing the parliament with total discretion as to the extent and nature of representation it may grant to territories in both houses. The High Court was ultimately asked twice to rule on this matter, which it did in favour of allowing territory representation, although the reasoning underpinning these decisions reveals the difficulty the court had in resolving the issue.

While the senators for the Northern Territory and the ACT are the most prominent form of territory representation in the Senate, more indirect representation is also afforded to Australia’s inhabited external territories. This takes the form of allowing eligible voters in these external territories to vote in Senate and House of Representatives elections in the Northern Territory and the ACT.

The fact of Commonwealth territory representation in the Senate, while no longer a matter of controversy, is interesting both in terms of the legislative process by which it was brought about, which exhausted the dispute-resolution mechanism under section 57 of the Constitution and was tested twice in the High Court, and in terms of its implications for our understanding of the constitutional limitations on the composition of the Senate.

Constitutional provisions

The number and distribution of representatives in the Senate is governed by several sections of the Constitution, as well as by legislation that has been passed pursuant to these sections. Section 7 of the Constitution states:

The Senate shall be composed of senators for each State, directly chosen by the people of the State, voting, until the Parliament otherwise provides, as one electorate.

...

Until the Parliament otherwise provides there shall be six senators for each Original State. The Parliament may make laws increasing or diminishing the number of senators for each State, but so that equal representation of the several Original States shall be maintained and that no Original State shall have less than six senators. The six original states at the time of

The six original states at the time of Federation are therefore each guaranteed a minimum of six senators. The original states are also guaranteed that, although parliament may increase the number of senators for each state above this original allocation of six per state, it must maintain parity in the representation of the original states. The parliament has in fact increased the number of senators chosen by each state twice since Federation, from six to 10 in 1948, and from 10 to 12 in 1983.[1]

Section 7 sets a foundational limit for the composition of the Senate which must be met until the parliament otherwise provides and, as noted, also sets limits within which the parliament may vary this original arrangement for the original states. However, the Constitution also contains another provision which allows the parliament to vary the composition of the Senate. Section 122 provides:

The Parliament may make laws for the government of any territory surrendered by any State to and accepted by the Commonwealth, or of any territory placed by the Queen under the authority of and accepted by the Commonwealth, or otherwise acquired by the Commonwealth, and may allow the representation of such territory in either House of the Parliament to the extent and on the terms which it thinks fit.

Read in isolation, this section appears to allow the parliament unlimited power to determine how Commonwealth territories may be represented in either of its houses, if at all, and on what terms this representation will take place. The parliament relied on this section to allow the election of two senators each from the Northern Territory and the ACT in 1975. Whether section 122 in fact provides such an unlimited scope to the parliament to determine the representation of Commonwealth territories, or whether it is rather to be read as allowing the parliament to determine territory representation only such that it does not disturb the balance outlined in the provisions of part II of chapter I, is a matter that the High Court clarified in 1975 and 1977.

Through the effects of legislation passed in accordance with sections 7 and 122 of the Constitution, the size of the Senate has stood at four distinct levels since Federation: 36 senators from 1901 to 1949, 60 senators from 1950 to 1975, 64 senators from 1976 to 1984, and 76 senators from 1985 onwards.[2] The present complement of 76 senators can be divided according to their distinct constitutional foundations—ultimately, 72 senators can be attributed to section 7 of the Constitution, while the four territory senators can be attributed to section 122.

Commonwealth territories—internal and external

By population, the Northern Territory and the ACT are, by a great margin, the largest of the Commonwealth’s territories and they are the only territories to have been allocated representatives in either house of the parliament.[3] Nevertheless, the remaining Commonwealth territories are not entirely excluded from representation in parliament, albeit they enjoy it only in attenuated form.

The Constitution mentions three methods by which the Commonwealth may acquire a territory. As section 111 states, a state may surrender territory to the Commonwealth:

The Parliament of a State may surrender any part of the State to the Commonwealth; and upon such surrender, and the acceptance thereof by the Commonwealth, such part of the State shall become subject to the exclusive jurisdiction of the Commonwealth.

Section 122, quoted earlier, expresses the Commonwealth’s law-making powers for any territory that is surrendered to it by a state, as in section 111, or placed under its authority by the Queen, or ‘otherwise acquired’.[4]

Commonwealth territories are commonly divided into external and internal territories.

Internal territories

The Commonwealth currently possesses the following internal territories: Northern Territory, Australian Capital Territory and Jervis Bay Territory. The Northern Territory was formally transferred from South Australia to the Commonwealth under the Northern Territory Acceptance Act 1910.[5] In accordance with section 125 of the Constitution, the Australian Capital Territory, originally called first the Territory for the Seat of Government and then the Federal Capital Territory, was transferred to the Commonwealth from New South Wales by the Seat of Government Acceptance Act 1909.

It is notable that the populations of both territories, although small at that early stage, were stripped of all political representation in the process of transferral to the Commonwealth. Whereas they had previously been represented in the parliaments of South Australia and New South Wales respectively, they subsequently had no representation in the federal parliament and no form of local government. They were also unable to vote in constitutional referendums until this right was granted by a constitutional amendment in 1977.[6]

Jervis Bay Territory is included as part of the ACT federal electorate of Fraser and its residents are counted towards the total ACT population for the purposes of calculating the ACT’s quota of House of Representatives seats. Jervis Bay Territory residents are also able to vote in elections for the two ACT Senate positions.[7]

External territories

The Commonwealth also possesses the following external territories, listed with their date of establishment: Christmas Island (1959) and Cocos (Keeling) Islands (1955), known collectively as the Indian Ocean Territories; Ashmore and Cartier Islands (1933); Coral Sea Islands (1969); Australian Antarctic Territory (1933); Heard and McDonald Islands (1953); and Norfolk Island (1914).[8] Of these external territories, Ashmore and Cartier Islands, the Coral Sea Islands, the Australian Antarctic Territory, and the Territory of Heard and McDonald Islands are considered uninhabited and are not represented in any form in the federal parliament.[9]

The Indian Ocean Territories and Norfolk Island are both afforded some measure of representation in the federal parliament via association with mainland territory electorates. The Indian Ocean Territories are included in the Northern Territory federal electorate of Lingiari and residents in those territories are able to vote in Senate elections for the Northern Territory.[10]

The relationship of Norfolk Island with the federal parliament is currently a more complex variation on that of the Indian Ocean Territories. Under current arrangements eligible residents of the island, who are not required to enrol but can do so if they wish, are able to nominate a mainland electoral division in which to vote. Their choice of electorate may be based on where they were last eligible to be enrolled, or their next of kin is currently enrolled, or they were born, or they have the closest connection. If none of these criteria apply, Norfolk Islanders can enrol in the division of Canberra in the ACT or Solomon in the Northern Territory, but are excluded from enrolling in Fraser in the ACT or Lingiari in the Northern Territory.[11] Norfolk Islanders also participate in ACT Senate elections.[12]

In 2015, however, the federal government, having reached the conclusion that the ‘current governance arrangements have been unable to deliver an adequate level of services to the community or an effective safety net for those most vulnerable in this small isolated community’, introduced a package of eight bills intended to reform the governance arrangements of Norfolk Island.[13] This package of bills passed the federal parliament on 14 May 2015.[14] These reforms include the abolition of the Norfolk Island Legislative Assembly and Executive Council and the establishment of an Advisory Council in its place, the application to the island of New South Wales law as Commonwealth law, and the extension of certain mainland health, social security and immigration arrangements to the island.[15] It is also intended that, from 1 July 2016, it will become compulsory for Norfolk Island residents to vote in federal elections and they will be required to vote in the electorate of Canberra.[16]

A central feature of these reforms to the governance of Norfolk Island is the revocation of the current self-government arrangements, which were originally granted by the Norfolk Island Act 1979. The Legislative Assembly of Norfolk Island expressed its opposition to this change in a recent remonstrance addressed to the Speaker of the House of Representatives and the President of the Senate, which requested that:

… the Commonwealth Parliament affirm the rights of the people of Norfolk Island to self government by re-examining those aspects of the Norfolk Island Legislation Amendment Bill 2015 that result in the removal of the Norfolk Island Legislative Assembly and call on the Prime Minister to confer on the people of Norfolk Island the right to freely determine their political status, their economic, social and cultural development and be consulted at referendum or plebiscite on the future model of governance for Norfolk Island before such changes are acted on by the Australian Parliament.[17]

In summary, insofar as residents of inhabited external territories and the internal Jervis Bay Territory participate in elections for House of Representatives and Senate seats, they have some small measure of influence in the composition of the federal parliament and also specified parliamentarians to whom they can take their grievances. However, the electorates into which they are subsumed are geographically distant and far larger in population, making the influence of these small territories minimal.

A further difficulty with this arrangement stems from the practice of applying state law to external territories as Commonwealth law in that the territories have no representation in the state jurisdiction under which those laws are determined. For example, Western Australian law applies to the Indian Ocean Territories, but those territories are not represented in any way in the state parliament. The same situation will occur on Norfolk Island after the implementation of New South Wales law as part of current reforms.[18]

House of Representatives representation for the Northern Territory and the ACT

Various pieces of legislation to allow representation of the Northern Territory and the ACT in the House of Representatives were passed by the parliament with relatively little controversy, despite such legislation relying on section 122 of the Constitution, which in the case of Senate representation created great controversy. The Joint Select Committee on Electoral Reform observed in its 1985 report Determining the Entitlement of Federal Territories and New States to Representation in the Commonwealth Parliament:

The full potential of section 122 was not realised, despite the fact that the Northern Territory had been represented in the House of Representatives since 1922, until the issue was brought to a head by the enactment of the Senate (Representation of Territories) Act 1973.[19]

Although the Northern Territory was granted representation by a single member in the House of Representatives in 1922, the terms of this representation were strictly limited. Section 5 of the Northern Territory Representation Act 1922 set out a series of restrictions, including that the member: not be allowed to vote on any question, not be taken into account when determining if a quorum is present, not be capable of being chosen as Speaker or Chairman of Committees, and not be counted in the determination of absolute majorities.[20] Sir Granville Ryrie, Member for North Sydney, noted at the commencement of the second reading debate in the House that the disabilities listed above had been modelled on the position of representatives for the territories of Alaska and Hawaii in the US Congress at that time, namely that they could be present and speak, but had no power to affect the outcome of votes or quorums.[21]

The Northern Territory Representation Act 1922 was amended in 1936 to allow the member for the Northern Territory to vote on a motion for the disallowance of a Northern Territory ordinance, and again in 1959 to allow a vote on any matter relating solely to the Northern Territory. A further amendment in 1968 removed all remaining restrictions on the Northern Territory representative.[22]

A similar staged introduction of representation in the House of Representatives occurred for the ACT. The ACT was not granted a representative in the House until the enactment of the Australian Capital Territory Representation Act 1948. The voting rights of the ACT representative were initially restricted to disallowance motions for ordinances of the ACT. A 1959 amendment expanded these rights to cover all matters relating solely to the ACT, and a 1966 amendment removed all remaining restrictions.[23]

The ACT gained a second House of Representatives seat in 1974 and briefly had a third seat between the 1996 and 1998 elections.[24] The Northern Territory gained a second seat in 2000. A 2003 determination by the Australian Electoral Commission removed this second seat, a decision which was set aside by the passage of the Commonwealth Electoral Amendment (Representation in the House of Representatives) Act 2004, which restored the second Northern Territory seat.[25]

Senate representation—1920 Northern Territory proposal

While the Northern Territory was first granted representation in the House of Representatives in a limited form in 1922, the first attempt to provide for territory representation under section 122 of the Constitution in fact came in 1920 and aimed to provide for Senate representation for the Northern Territory. The 1920 legislation was put forward by the Hughes Government to attempt to address local concerns regarding the lack of Northern Territory representation in the federal parliament—a complaint expressed in shorthand as ‘no taxation without representation’.[26]

The Northern Territory Representation Bill 1920 was introduced in the Senate on 9 September 1920 by Senator Edward Russell, Vice-President of the Executive Council, and debated on 15 and 16 September. The bill proposed to allow a single representative for the Northern Territory to be elected to the Senate for three-year terms aligned with those of the House of Representatives. The bill also stipulated that this representative not have the right to vote.[27]

In debate, senators raised the constitutionality of the proposal, questioned whether it would not be better to grant a greater measure of self-government to the Northern Territory, debated the effect of including territory senators on the balance of state representation in the chamber, argued that it would be better to introduce representation in the House of Representatives, and compared the size of the population of the Northern Territory with that of the various states. In this respect it was argued by some that the bill was worthy of support despite the Northern Territory possessing only ‘2,800 white people, less about seventy-six foreigners’.[28] This estimate excluded the Aboriginal population of the Northern Territory, which at the time stood at around 19,500—over 80 per cent of the total population.[29]

This early attempt to provide for Senate representation for a territory did not progress beyond the second reading stage, which concluded when the motion that the bill be read a second time was amended in the following terms:

That the Senate is of the opinion that residents of the Northern Territory who would under ordinary circumstances be entitled to a vote under the Commonwealth Electoral laws, should for that purpose be attached to the State of South Australia and accorded the right to vote at the election for Senators for that State, thus being granted representation in the Commonwealth Parliament.[30]

The proposal put forward in this motion was not taken up by the parliament.

Territory governance to 1968

Following the abortive 1920 proposal, no further legislative attempts were made to allow for the representation of territories in the Senate until 1968, when Gough Whitlam, initially as Leader of the Opposition, again made attempts to territory senators. In the intervening period territory representation, with various forms of restriction, was allowed in the House of Representatives. The populations of both the Australian Capital Territory and the Northern Territory had increased greatly over those 48 years. According to Australian Bureau of Statistics figures, at the time of the debate on the 1920 legislation, the Northern Territory included only 3,989 people (excluding the Aboriginal population), and the ACT the even smaller population of 1,972. By the time legislation on this matter was again introduced in the federal parliament in 1968, the Northern Territory had grown to include 70,223 people and the ACT had grown to 116,604, and went on to approach double this figure in the following 10 years.[31] Although the populations of both territories had increased in the intervening period, governance arrangements within the territories had not yet evolved into the forms of self-government that currently exist.

Northern Territory governance

The Northern Territory, from the time of its surrender to the Commonwealth in 1911 up until 1946, with several short exceptions, was governed by an administrator appointed by the Governor-General, who was subject to instruction by the Commonwealth minister and aided by a Council of Advice consisting of six appointed members. Legislation for the territory was made by the federal minister and the Commonwealth Parliament.[32] In 1947 a legislature was established for the Northern Territory in the form of a Legislative Council, which consisted of a mix of appointed and elected members and the Administrator, the make-up of which varied up to 1974, when a fully elected Legislative Assembly of 19 members was established. Selfgovernment, with certain restrictions on responsibilities, was conferred on the Northern Territory in 1978.[33]

These arrangements, along with a perceived lack of economic development in the Northern Territory, were the subject of considerable political agitation prior to the grant of self-government in 1978. In 1962 the Northern Territory Legislative Council presented a remonstrance to the two houses of the federal parliament detailing a list of eight grievances regarding the governance of the territory. The first three of these grievances illustrate the dissatisfaction caused by a lack of effective self-government and restricted representation in the federal parliament:

1. The political rights of the citizens of the Northern Territory are inferior to those of other citizens of Australia.
2. The Commonwealth Government has failed to develop the Northern Territory to the reasonable limits of the capacity of the Commonwealth and the Territory itself.
3. The Legislative Council for the Northern Territory, although responsible for the making of laws for the peace, order and good government, has no voice in the allocation or expenditure of government moneys in the Territory.[34]

ACT governance

The Australian Capital Territory, established in 1909, was administered by the federal Minister for Territories with the assistance of advisory bodies from 1920 onwards. These advisory bodies included both elected and appointed members, in varying combinations; however the first fully elected body, known initially as the Legislative Assembly and subsequently as the House of Assembly, began in 1974. The minister responsible for the administration of the ACT was under no obligation to follow the advice provided by any of these bodies. Legislation to establish self-government for the ACT was passed by the federal parliament in 1988, with the new ACT Legislative Assembly commencing in 1989.[35]

The prospect of self-government was not universally embraced by ACT residents. The dominant Commonwealth department in the administration of the ACT in the period prior to self-government was the National Capital Development Commission, which was established in 1958 to oversee the rapid development of Canberra. It was initially admired but came to be a source of dissatisfaction due to the ‘pervasive intrusiveness’ of its planning powers.[36] Schemes for self-government had been discussed throughout the history of the ACT; however, it was the Hawke Government that pressed ahead with its implementation in the face of considerable local opposition. Philip Grundy et al summarise the attitude of ACT residents towards self-government as follows:

At a time when people in other parts of the world were fighting for the right to govern themselves, the majority of the people of Canberra had, by their opposition to self-government, effectively perverted the cry of the American revolution into: “No representation if it involves taxation”. While democracy in Europe emerged because the governed imposed their wishes on the governors, in Canberra’s case the governors “imposed” democracy on a largely reluctant populace whose only apparent concern was to avoid paying any more for the administration of their city.[37]

Senate representation—Whitlam era reforms, 1968-75

The process of legislating for Senate representation for the Northern Territory and the ACT occurred over a period of six years from 1968 to 1974. Pursued by the Whitlamled ALP, first in opposition and then in government, this reform took place in the context of rising populations and governance regimes that had not yet delegated powers to local democratic institutions.

Whitlam brought forward legislation to provide for territory representation in the Senate in line with the then Labor Party platform, which otherwise expressed the party’s critical attitude towards the very existence of the Senate. The 1967 platform listed among the methods by which the party hoped to achieve ‘democratic socialisation’ several constitutional amendments including ‘to abolish the Senate’ and, pending the achievement of this end, ‘to remedy defects as they appear and to keep the Constitution abreast of changing conditions’.[38] The intention to abolish the Senate remained in the platform until 1979.39 In line with the intention to ‘remedy defects as they appear’, a further element of the platform was to provide:

Full voting rights for the representative of the Northern Territory in the House of Representatives and full voting rights for representatives of the Northern Territory and the Australian Capital Territory in the Senate.[40]

On 7 November 1968 Whitlam, then the Leader of the Opposition, introduced the Territory Senators Bill 1968. He explained the democratic motivation behind it:

The Bill represents the culmination and consummation of the process of representation of the two mainland Territories in the Australian Parliament. For as long as the Australian Parliament is bicameral, it is in accordance with all our political tenets and instincts that all portions of Australia and all people of Australia should be represented in both Houses of the Parliament. It is proper that the governed should have a share in choosing their governors and calling them to account. The Constitution requires this as regards the States; it permits it as regards the Territories. We should not permit the position to continue any longer where residents and electors of the Northern Territory and the Australian Capital Territory can choose representatives in one alone of the two chambers of this Parliament.[41]

In this bill, and in all subsequent attempts to pursue this matter, Whitlam proposed that two territory senators be elected each from the Northern Territory and the Australian Capital Territory. In contrast to the initial restricted representation granted to the territory members in the House of Representatives, these four senators would have full voting rights on all questions arising in the Senate, have all the powers privileges and immunities of senators for the states, and be included in the calculation to determine a quorum.

Whitlam noted, however, that under his proposal, ‘there are some differences between the extent and nature of representation of each Territory and those which the Constitution requires for the States’. These differences, beyond that of the number of senators elected, were that territory senators would not enjoy six-year terms commencing on 1 July following the election. Rather, they would have terms equivalent to those of members of the House of Representatives—that is, a maximum of three years without a fixed start or end date. Furthermore, and as a consequence of the shortened terms, they would not be elected according to the half-Senate rotation system set out for senators of the original states. Rather, both senators in each territory would face an election at the conclusion of each House of Representatives term.[42]

Whitlam’s stated rationale for setting the number of senators for each territory at two was that it would all but ensure, given the operation of the proportional representation system in Senate elections, that there would be representation of both major political parties from each territory:

I have suggested 2 senators because I believe it would be proper to have an even number representing the Territories. If there was one senator alone representing each Territory, almost certainly the one party would be represented for long periods. It is very probable that both senators would belong to the same Party.[43]

He further commented that this parity would probably be maintained with four senators from each territory, but that six senators may lead to ‘unequal representation as between the two major parties’. As noted in Odgers’ Australian Senate Practice, it is a strength of the Senate’s proportional representation system that it recognises both majority and minority interests in the parliament.[44] However, the proposal for two senators for each territory appears to have also been settled on in order that, barring highly unusual voting patterns, the new senators would not disturb the balance of power between the major parties in the upper house.

Odgers’ notes that this element of the proposal, once implemented, has operated largely as intended in that the two Senate positions in the ACT and the Northern Territory have generally been divided between the major parties, in contrast to their House of Representatives seats:

Given that each territory’s representation is currently limited to two senators, the practice of electing both at the one election by proportional representation preserves the Senate’s role as a House which enhances the representative capacity of the Parliament and provides a remedy for the defects in the electoral method used for the House of Representatives … [S]ince the 1980 general election all members of the House of Representatives for ACT electorates have usually been members of the Australian Labor Party. Throughout much of this period, one senator has been a member of the ALP, the other senator from the Liberal Party. Oneparty representation in the House has also been common for the Northern Territory, so that its two senators are also essential to providing that territory with balanced representation.[45]

The decision to tie the terms of territory senators to the terms of the House of Representatives was based on a policy position of the ALP at the time that elections for the House of Representatives and the Senate ought to be held simultaneously. This position was based on a recommendation of the Joint Committee on Constitutional Review made in 1958:

The Committee recommends that the Constitution be altered to omit the provision now made for senators to be chosen for terms of six years and to provide instead that senators should hold their places until the expiry or dissolution of the second House of Representatives after their election, unless the Senate should be earlier dissolved under the provisions of section 57 of the Constitution.[46]

As Whitlam explained, his intention was to implement this measure in the case of territory senators, as it was in the power of the parliament to do so, whereas constitutional change was required to align the terms of state senators with elections for the House of Representatives:

It was believed by that Committee that this would cut down the number of elections, that it would avoid the situation which has now obtained ever since the premature election for the House of Representatives in 1961, and that it would also promote concentration on the same issues for elections for both Houses. It would minimise the distraction of elections; it would minimise the differences between the Houses.

The Parliament can make this provision as regards the Territories. Accordingly I have suggested that both senators for each Territory should be elected every time there is a general election for the House of Representatives. The Acts which give representation to each Territory in this chamber provide that there shall be an election for the member for each Territory if there is a general election for members of the House of Representatives for the States. So at least in the two Territories there would be an election for both Houses of the Parliament at the same time.[47]

It should be noted, however, that Whitlam’s proposal did not in fact correspond to that proposed by the joint committee as the territory senators would serve for only one House of Representatives term, rather than two.[48]

The proposed amendment to the Constitution outlined by the joint committee to tie Senate terms to two House of Representative terms, was later defeated at three referendums, first on 18 May 1974 and again on 21 May 1977 and 1 December 1984. As noted in Whitlam’s explanation, much of the determination to alter the terms of senators arose because of the series of unsynchronised elections held in the course of the 1960s—half-Senate-only elections were held in 1964, 1967 and 1970. A half- Senate-only election was not held again until 2014, and then only in Western Australia due to the voiding of the 2013 half-Senate election result in that state.[49]

Finally, Whitlam noted an apparent conflict between sections 7 and 122 of the Constitution regarding the composition of the Senate, but made the following argument regarding how this conflict ought to be interpreted:

It might be thought, looking at section 7 in isolation, that the Senate could never be other than a States House. However section 122, being later and more specific, would override it to the extent of any representation of the Territories in the Senate. It will be noted that this Parliament determines the extent and the terms of Territory representation in either chamber. I have detailed the history of that representation in this chamber. But similarly both Houses of this Parliament can determine the extent and terms of representation of all the Territories in the Senate.[50]

When pushed by an interjection on the constitutionality of the proposal, Whitlam further noted:

I know of no decisions, but I have consulted the academics in this field and the view which they unanimously express is that section 122, being later and specific, extends the provisions of section 7.[51]

Legislation history

Whitlam’s Territory Senators Bill 1968 was introduced and debated on 7 November 1968. Time expired during the second reading debate and the bill was never returned to. However, the proposals it contained for territory representation in the Senate eventually became law after a convoluted and highly contested journey through the parliament. While still in opposition, Whitlam reintroduced the same legislation in 1970 and, once again, time expired during the second reading debate and the bill was not brought on again.

Having achieved a majority in the House of Representatives in the 1972 election and thereby forming government, the ALP introduced the Senate (Representation of Territories) Bill 1973, which again contained essentially the same proposal for territory representation in the Senate as it had put forward from opposition in 1968 and 1970. On this occasion, the bill was accompanied by a further piece of legislation, the Representation Bill 1973, which was intended to exclude the proposed territory senators from nexus calculations set out by section 24 of the Constitution, which states in part:

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.

Fred Daly, the Minister for Services and Property, when introducing this legislation, explained the necessity for this accompanying legislation:

A short time ago I introduced the Senate (Representation of Territories) Bill which provides for senatorial representation for the Australian Capital Territory and the Northern Territory. The Government’s legal advice is that section 24 of the Constitution does not have application in relation to senators who may be provided for a Territory under the provisions of section 122 of the Constitution. In other words, the requirement contained in section 24 for the number of members of the House of Representatives to be as nearly as practicable twice the number of senators does not relate to Territory members or senators provided under section 122 of the Constitution. Furthermore, ‘the people of the Commonwealth’ in the context of section 24 are the people of the States.[52]

Thus, the bill sought to amend the Representation Act, which contained the formula for determining the number of members for the House of Representatives to be chosen from each state, to clarify that the people of the Commonwealth are the people of the states in this case, and that territory senators are excluded from the formula for determining the number of members of the House of Representatives.[53]

Both the Senate (Representation of Territories) Bill 1973 and the Representation Bill 1973 passed the House of Representatives on 30 May 1973.[54] However, as the ALP did not command a majority in its own right in the Senate at this time, the two bills failed to progress beyond the second reading in the Senate. Having been introduced on 31 May 1973, taken together by leave and debated on 7 June 1973, the second reading motion was negatived.[55]

This process was repeated later that same year. The bills were reintroduced in the same form in the House of Representatives on 25 September 1973, and debated and passed on 27 September.[56] They were then reintroduced in the Senate on 9 October 1973, debated, and the second reading motion negatived on 14 November 1973.[57]

In accordance with section 57 of the Constitution, these two bills, along with four others, having been twice rejected by the Senate, were used by the Whitlam Government as the basis on which to advise the Governor-General to dissolve both houses of parliament. A general election for both houses occurred on 18 May 1974. Labor again achieved a majority in the House but again fell shy of a majority in the Senate. Thus, while the six bills were reintroduced and passed in the House of Representatives on 10 and 11 July 1974, they were each negatived at the second reading in the Senate over the course of the following two weeks.[58]

The parliament thereby moved to the final phase of the dispute-resolution process contained in section 57 of the Constitution, with a joint sitting held on 6 and 7 August 1974 to consider the six pieces of legislation. To date, this is the only joint sitting of the Senate and House of Representatives held to resolve a legislative deadlock between the two houses. The two pieces of legislation of interest in this context, the Senate (Representation of Territories) Act 1973 and the Representation Act 1973 were both affirmed on 6 August 1974.[59]

Themes of debate

The sixth edition of Australian Senate Practice provides a brief summary of the arguments raised in favour of, and in opposition to, the proposal to introduce territory representation in the Senate. It states:

Factors advanced in support of territorial Senators included the increased territory population; payment of taxes in accordance with the law; and the right of the people of the Australian Capital Territory and the Northern Territory to have presented to the Senate the views of the residents of their respective territories.[60]

As an amplification of the final point above, it is important to note that the lack of self-government in the mainland territories at this point meant that much of the law by which they were governed took the form of Commonwealth secondary legislation, more specifically ordinances. Although both territories had representation in some form in the House of Representatives, the Senate has historically been the chamber, through its committee system, where significant scrutiny of delegated legislation takes place and, as such, a strong argument could be put forward at that time that territory residents ought to have a representative who could participate in this process.[61]

Aside from accusations that the proposal would lead to a political advantage for the ALP, and that it was motivated by this possible outcome, arguments raised in opposition to the introduction of territory senators during debate in the Senate included the following:

1. Territorial Senators might undermine the constitutional concept of the Senate as a States House.
2. The Constitution was a compact between the Commonwealth and the States and a principle of the agreement was that, by equal representation of the original States in the Senate, the three numerically big States could not ride roughshod over the interests of the three numerically smaller States, which were heavily outnumbered in the House of Representatives.
3. The States, which agreed to federate on the basis that State rights would be safeguarded, had not been consulted on the effect of the proposed territory representation in the Senate.
4. Although provision is made in the Constitution for representation of territories in the Federal Parliament, the founding fathers did not envisage that such representatives would have voting rights. In fact, in the Convention debates concern was expressed that the provision did not specifically preclude such representatives having the right to vote.
5. The Bill proposed tying the terms of territory representatives to the life of the House of Representatives and this was the thin edge of the wedge to tying the terms of all Senators to the life of the Lower House, thus destroying the Senate’s independence.
6. Once territory representation in the Senate began, where would it end, having in mind other territories, including Australian Antarctic, Christmas Island, Cocos (Keeling) Islands, Coral Sea Islands, Heard Island and McDonald Islands, and Norfolk Island.
7. Doubts were expressed whether the Bill, if passed would be held to be legally valid. It was considered a nice legal question whether the power given in section 122 of the Constitution to give representation to the territories would allow the appointment of a person who actually becomes a voting Senator and one who is counted in a quorum.
   There was an inconsistency between section 7 which provided for a States House only and section 122 which related to territory representation in either House of the Parliament.
8. The Australian Senate was largely modelled on the Senate of the United States of America, where territories have never been represented in the Senate. In 1967, a Congressional Committee considered the proposal that the District of Colombia be represented by two Senators, but the Constitution was not so amended. One of the arguments advanced against the proposal was that small State influence in the Senate could be defeated as effectively by according senatorial representation to a non-State as by according more representation to a larger State.[62]

Several of these arguments touch on fundamental conceptions of the function of the Senate and interpretation of the Constitution. The conception of the Senate as a states’ house and the constitutionality of the legislation were the two most fundamental issues raised in debate and they are discussed in turn below. Both of these issues were at play in the subsequent High Court challenges to the Senate (Representation of Territories) Act 1973.

Senate as a states’ house

That the Senate is a states’ house, a status incompatible with the introduction of voting territory senators, was perhaps the view put most regularly by opponents of the Whitlam legislation. The debate in the Senate makes it clear that senators understood the term ‘states’ house’ to mean that the Senate had been conceived as a place where the interests of particular states would be expressed. This view was summarised in 2001 by the then Clerk of the Senate, Harry Evans:

Because the framers used the shorthand expression ‘States’ House’ in relation to the Senate, it is assumed that they intended that senators vote in state blocs and according to the effect of proposed measures on the interests of particular states. Because senators have never voted in this way, it is assumed that the Senate has not achieved its original purpose.[63]

Much of the debate over territory senators occurred against the background of such an assumed view. Thus opposition senators argued that, as a states’ house, the Senate should contain only senators from the states as it is exclusively the interests of the states that should be represented there. In opposition to this view, Labor senators argued that the Senate was no longer a states’ house, if it ever was, because senators do not vote as state blocs.

However, it has been argued by Evans that this understanding of the role of the Senate is mistaken. He argued that the fundamental constitutional principle underlying the Senate is that of a ‘geographically distributed majority’:

The purpose of the Senate was to ensure, by securing equal representation of the states, regardless of their population, in one House of the Commonwealth Parliament, that the legislative majority would be geographically distributed across the Commonwealth. In other words, it would be impossible to form a majority in the legislature out of the representatives of only one or two states. Without that equal representation in one House, the legislative majority could consist of the representatives of only two states, indeed, of only two cities, Sydney and Melbourne, and this would lead to neglect and alienation of the outlying parts of the country.[64]

Odgers’ also emphasises the vital role this second majority plays in preventing regional alienation from central government in a geographically large nation:

In such a nation, particularly a nation occupying a large geographical area, a central legislature elected by the people as a whole necessarily involves the danger that a majority within that legislature could be formed by the representatives of only one or two regions, leading to neglect of the interests of other regions and their consequent alienation from the central government. The solution to this problem is to have one house of the legislature elected by the people as a whole, representing regions in proportion to their population, and one house elected by the people voting in their separate regions, and representing those regions equally.[65]

Dr John Cockburn’s statement during the Australasian Federal Convention of 30 March 1897, makes the point well, and illustrates that the importance of preventing more populous states from entirely dominating the federal parliament was well understood at the time the Constitution was framed:

… the great principle which is an essential, I think, to Federation—that the two Houses should represent the people truly, and should have co-ordinate powers. They should represent the people in two groups. One should represent the people grouped as a whole, and the other should represent them as grouped in the States. Of course majorities must rule, for there would be no possible good government without majorities ruling, but I do not think the majority in South Australia should be governed by the majority in Victoria, or in New South Wales. In every case the majority should rule, but that does not mean that the majority of one colony is to coerce the majority of another. If we wish to defend and perpetuate the doctrine of the rule of majorities, we must guard against the possibility of this occurring.[66]

Thus, at the time of Federation, the six original states were the political entities entering into the federal compact and the goal of geographically distributing the majority in the second chamber was, naturally enough, identified with protecting the interests of the less populous states against those of the more populous. However, with the gradual development of the populations of the two large mainland territories, the Northern Territory and the ACT, it is possible to argue that the introduction of senators with voting rights from beyond the original six states better accords with the concept of a geographically distributed majority than does their permanent exclusion.

This case can be strengthened by noting that the two large mainland territories arose from alterations to the geographic coverage of two of the original states—South Australia and New South Wales. Thus, the geographical boundaries of states are changeable and populations can be, and have been, shifted from states to Commonwealth territories. Population centres can also grow very rapidly, as occurred in the ACT in the post-war period. An inflexible constitution that prevents any alteration in the composition of the parliament, in particular the Senate, to reflect such demographic changes presents the danger of generating precisely that regional alienation from central government discussed above.

Constitutionality of appointing voting senators

As noted above, the constitutionality of legislation admitting territory senators with full voting rights was the second major theme of debate in the Senate. Following the affirmation of the Senate (Representation of Territories) Bill 1973 and the Representation Bill 1973 at the 1974 joint sitting, this matter was tested in a series of High Court challenges between 1975 and 1977. These challenges clarified the power of the parliament to determine the extent and terms of territory representation in both the Senate and the House.

Initially, in Western Australia v Commonwealth (1975) (the First Territories Representation Case) three states with non-Labor governments, Western Australia, New South Wales and Queensland, challenged the validity of the Senate (Representation of Territories) Act 1973, the Commonwealth Electoral Act (No 2) 1973 and the Representation Act 1973 on the ground that each Act had not been duly passed by both houses of the parliament within the meaning of section 57 of the Constitution, and further that the Senate (Representation of Territories) Act 1973 was invalid as it was beyond the legislative power of the Commonwealth Parliament.[67]

The three Acts were found to have been duly passed in accordance with section 57. The court also held, by a majority of four to three, that the Commonwealth had the power to enact legislation which provided for two senators each for the ACT and the Northern Territory with full voting rights. Professor Leslie Zines summarised this aspect of the case:

The main issue was which of s 7 and s 122 of the Constitution prevailed over the other. Section 7 provides in part that ‘The Senate shall be composed of senators for each state …’ and s 122 provides in part that ‘The Parliament may make laws for the government of any territory … and may allow the representation of such territory in either House of the Parliament to the extent and on the terms it thinks fit’. It was recognised that if one had regard only to the broad language of s 122 the provision of territory senators with full voting rights would clearly come within that section. The dissenting judges, however, considered that the ‘federal’ nature of the Constitution prevented s 122 being interpreted that way.[68]

The legal reasoning by which the justices determined which of section 7 and section 122 ought prevail over the other is a complex matter,[69] but, as noted in this summary, a significant consideration for the dissenting justices was their emphasis on what they took to be the overriding federal character of the Constitution. For example, Barwick CJ argued that as section 122, on which the legislation in question relied, is in a part of the Constitution dealing with the creation of new states and the acquisition of territories, it is ‘in its nature incidental, in a sense peripheral, to the central and dominant purpose of the Constitution’. In a later passage, the Chief Justice explained:

Some lesser connotation of the word “representation” must be found to make the Constitution, basically federal in nature, consistent throughout.

To fit Pt VI into the Constitution as a whole, in my opinion, the expression “allow representation” must be construed so as to be consonant with and indeed to preserve and not to endanger or destroy an essential feature of federation, namely the maintenance of the Senate as the State House.[70]

As noted in the previous section, the concept of a ‘states’ house’ is itself a matter of interpretive difficulty, but it is clear that the Chief Justice took it to mean at least that representation in the Senate with voting rights must be forever restricted to senators for the states. In contrast Mason J, part of the majority in this case, argued that:

The apparent opposition which arises from the reference to representation of the territories in s. 122 and the absence of any such reference in ss. 7 and 24 is irreconcilable only if it be assumed that Ch. I in making provision for the composition of the Senate and the House is necessarily speaking for all time. To my mind this assumption is misconceived. Sections 7 and 24 should be regarded as making provision for the composition of each House which nevertheless, in the shape of s. 122, takes account of the prospective possibility that Parliament might deem it expedient, having regard to the stage which a Territory might reach in the course of its future development, to give it representation in either House by allowing it to elect members of that House.[71]

Jacobs J took the view that, although the original states were guaranteed equality of representation in the Senate, this equality among the original states would not be disturbed by the addition of senators representing new states or territories:

… it is said that the admission of Territories to the full franchise would upset the delicate balance of State power intended in the constitution of the Senate. This is not correct. The intention was that the original States, large and small, would have equal representation. The purpose was to ensure that the larger States did not overbear the smaller States. Neither full representation of new States nor full representation of Territories in the Senate would affect this objective unless the numbers of new senators completely submerged the original intention. And that is the suggestion which has been made to us on this hearing.[72]

It is a measure of the ambiguity of the Constitution on this issue that Professor Zines uses this case as an example to illustrate the point that:

It is increasingly being recognised by some judges that in many cases a decision either way cannot be regarded in any objective sense as “right” or “wrong” but only as preferable or undesirable having regard to a number of complex factors.[73]

The issue of territory representation in the Senate was revisited by the court in 1977. Initially, this occurred in Attorney-General (NSW) ex rel McKellar v Commonwealth (1977), in which the High Court ruled that section 24 of the Constitution did not apply to the people or the representatives of the territories. This decision held the following significant implications regarding territory representation:

… territory representatives need not be “directly chosen by the people”. The Parliament may, for example, provide for them to be appointed or to be indirectly chosen, for example, through an electoral college system. Also the number of members a territory has in the House of Representatives does not have to be in proportion to its population. On the other hand the last paragraph of s. 24 giving each original State a minimum of five members in the House also does not apply. Similarly the Commonwealth is at large in determining the number of senators for any territory. The term of office of territory senators is not governed by s. 13 of the Constitution and the casual vacancy provisions in s. 15 do not apply. As s. 24 is inapplicable, there is no constitutional “nexus” between the number of senators that may be provided for to represent the people of the territories.[74]

In addition, comments made by Barwick CJ in his judgment regarding the court’s decision in the First Territories Representation Case appeared to encourage an attempt to recontest that matter.[75] Later that same year, Queensland and Western Australia initiated Queensland v Commonwealth (1977) (Second Territories Representation Case).[76] This case explicitly raised and dealt with the question of representation of territories in the House of Representatives alongside that of representation in the Senate. Although the court upheld, by a five to two majority, the decision in the First Territories Representation Case, this was not because a majority of the court believed that the legislation was valid. In fact a majority of the court (four to three) maintained that the legislation was invalid, but Gibbs and Stephen JJ nevertheless refused to overrule the earlier decision as they felt bound by the principle of stare decisis. Under this principle a court ‘is bound to follow previous decisions, unless they are inconsistent with a higher court’s decision or wrong in law’, a restriction that promotes certainty in the law.[77] Thus, in an unusual set of circumstances, the Senate (Representation of Territories) Act 1973 survived this second challenge despite being considered unconstitutional by a majority of the court.[78]

Post 1977 legislative reforms

Although section 122 of the Constitution had been relied on to provide territory representation in the House of Representatives from 1922 onwards, it was only through this series of High Court judgments that the scope of action granted to the parliament by that section was apparent. These judgments have established that the representation of territories in the federal parliament in accordance with section 122 is independent of the provisions of chapter I of the Constitution dealing with the composition and operation of the parliament. As noted by the Joint Committee on Electoral Reform in 1985, the following possibilities emerged from this determination:

• under section 122 of the Constitution the Commonwealth may be able to provide that a Territory Senator has more than one vote
• there are no limits to the number of Representatives in either Chamber that the Parliament may “see fit” to grant the Territories
• Territory representatives need not be directly chosen by the people. The Parliament may, for example, provide for them to be appointed or indirectly chosen
• as section 24 does not apply, there is no Constitutional “nexus” between the number of Members of the House of Representatives and the number of Senators representing the Territories.[79]

While constitutionally possible, the prospect of parliament agreeing to measures that would allow more than one vote to territory senators or of the appointment of nonelected senators or indeed of the appointment of greatly increased numbers of territory senators, is practically remote.[80] In the First Territories Representation Case, Jacobs J argued that such a possibility should not influence the court’s interpretation of the Constitution:

The Parliament, it is said, might create fifty or a hundred senators for a Territory with multiple voting to boot and that could never have been intended. It is a preposterous suggestion in that it puts the cart before the horse. It is the Parliament which must make the law for representation of Territories and the framers of the Constitution trusted a system of parliamentary government in which they were mostly immersed. Those who were lawyers were mostly parliamentarians as well and if as lawyers they might scan a document for its hidden traps or loopholes, their sense as parliamentarians would tell them that the Parliament itself was the safeguard against the absurd possibility. We likewise should construe the words of the Constitution by its plain terms and not by some distorting possibility.[81]

Mason J further pointed out that section 121 of the Constitution also presents the possibility of ‘swamping’ the Senate, although with senators from new states in this case. He too argued that the parliament must be relied on to employ its powers responsibly:

The first is the grim spectre conjured up by the plaintiffs of a Parliament swamping the Senate with senators from the Territories, thereby reducing the representation of the States disproportionately to that of an ineffective minority in the chamber. This exercise in imagination assumes the willing participation of the senators representing the States in such an enterprise, notwithstanding that it would hasten their journey into political oblivion. It disregards the assumption which the framers of the Constitution made, and which we should now make, that Parliament will act responsibly in the exercise of its powers.

Furthermore, such significance as the plaintiffs’ argument may have is diminished when it is appreciated that the Constitution provides no safeguard against the pursuit by Parliament of a similar course at the expense of the original States in allowing for the representation of new States in the Senate. Although s. 7 provides that equal representation of the original States shall be maintained in that chamber, neither the section nor the remaining provisions of Pt II of Ch. I place any restriction on the number of senators which Parliament may accord to a new State as its representation in the Senate. Here, again, the assumption is that Parliament will act responsibly.[82]

Nevertheless, the new understanding of the very broad scope of section 122 that emerged from these High Court challenges was a source of concern throughout the following decade, during which time various attempts to formulate restrictions were made, either through constitutional amendments or legislation.

For example, a committee of the 1982 Constitutional Convention discussed proposed amendments to the Constitution intended to ensure that representation of territories and new states would be governed by restrictions similar to those for the original six states. It was proposed that the Constitution be amended to ensure that territory or new state representatives be chosen by the people, that such representatives in the House of Representatives be in the same proportion to population as those chosen by the original states, and that representation in the Senate beyond the current two be in a ratio of one senator for every two members of the House.[83]

The matter was taken up in 1985 by the Joint Select Committee on Electoral Reform, which conducted an inquiry into the options for establishing ‘fixed formulae for determining the number of Senators and Members of the House of Representatives to which the Australian Capital Territory, the Northern Territory and other territories are entitled’.[84]

With regard to the House of Representatives, the committee proposed that the ACT and the Northern Territory be entitled to at least one member and that, thereafter, additional representation shall be in proportion to the population of the territory, using the same quota employed to determine the number of seats for each state. The people of territories other than the ACT and the Northern Territory shall be entitled to separate representation when their population exceeds one half of a quota and that, until that time, they should be included in electoral divisions of the Northern Territory and the ACT.[85]

With regard to Senate representation, the ACT and the Northern Territory will have at least two senators, and beyond this they will have one senator for every two members of the House of Representatives they are entitled to. Other Commonwealth territories shall have one senator for every two members.[86]

The committee further concluded that:

… constitutional change is required so that representation of territories and new states in the Parliament in future occurs according to principles acceptable to the Australian community. Constitutional amendments along the lines of the formulae we have proposed for inclusion in the Electoral Act … would meet the problems and anomalies that have been disclosed to exist under the Constitution at present.[87]

Although not directly the subject of this paper, it is noteworthy that this committee recommended that ‘no new State should be admitted to the Federation on terms and conditions as to representation in the Parliament more favourable than those prescribed for representation of Territories in the Electoral Act’.[88] This aspect of the report appears to have been directed at preventing the Northern Territory from gaining greater representation should it achieve statehood and was strongly criticised in a dissent by Senator Michael Macklin.[89]

The recommendations of this report regarding the formula for territory representation were incorporated into the Electoral and Referendum Amendment Bill 1989, which provided among other matters:

… fixed formulae for the representation of Territories in the Federal Parliament. The Joint Standing Committee examined this issue, following concern that it would be possible for a government with a majority in both Houses to increase the representation of the Australian Capital Territory and the Northern Territory out of proportion to their populations. The Government has accepted the Committee’s conclusion that fixed formulae for the representation of Territories should be prescribed. Accordingly, this Bill provides for the Australian Capital Territory and the Northern Territory to be entitled to be represented by at least one member of the House of Representatives, and that representation thereafter be in proportion to its population. Other Commonwealth Territories will be entitled to separate representation when their population exceeds more than one half of a quota as determined by section 48 of the Electoral Act.

The Bill also provides that where the numbers of members of the House of Representatives to be chosen in the Australian Capital Territory or the Northern Territory is six or more, that Territory will be entitled to representation in the Senate on the basis of one senator for every two members of the House of Representatives. However, each will be entitled to a minimum of two senators. Other Commonwealth Territories will be entitled to representation in the Senate on the basis of one senator for every two members of the House of Representatives.[90]

To date no attempt has been made to amend the Constitution in line with the recommendations of the 1985 Joint Standing Committee on Electoral Reform report. As such, the passage of this amendment bill set the parameters that currently govern the representation of territories in the Commonwealth Parliament. As noted in earlier discussion, the only significant departure from these formulae came in 2004 with the passage of the Commonwealth Electoral Amendment (Representation in the House of Representatives) Bill 2004, which restored a second House of Representatives seat in the Northern Territory after it had been removed by a determination of the Electoral Commissioner that it had fallen below the necessary population quota.

A curious feature of the current arrangements contained in the Commonwealth Electoral Act for determining House and Senate representation is that they reverse the direction of the nexus provisions in section 24 of the Constitution. In that section, now replicated in section 48 of the Commonwealth Electoral Act, the number of senators is doubled and used to divide the population of the Commonwealth to attain a quota, and the population of each state is divided by that quota to attain the number of House of Representatives members to be elected in each state. In the case of the territories, however, the quota obtained in the above manner (with reference only to the ‘People of the Commonwealth’, which excludes the territories for this purpose) is first used to determine the number of House of Representatives seats for each territory, and the number of House of Representatives seats is then used to determine the number of senators—i.e. one senator for every two House of Representatives members.

Conclusion

Senators for the Northern Territory and the ACT have been a fixture of the Commonwealth Parliament for 40 years and, although they enjoy full voting rights and participate in chamber and committee activities just as other senators do, the constitutional foundation of their presence is distinct from that of senators representing the original six states. As established via a series of High Court challenges, section 122 of the Constitution provides the parliament with total discretion to determine the size and nature of territory representation in the Senate and the House of Representatives. Senators for the territories are not governed by the provisions of part II of chapter I of the Constitution.

The distinct constitutional basis provided by section 122 has enabled the parliament to respond to the increasing populations of the Northern Territory and the ACT by granting these territories two Senate seats each, but it has also allowed the parliament to establish these seats on terms distinct from those for state senators. This has led to an assimilation of territory senators’ terms to those of members of the House of Representatives, rather than a reproduction of those terms granted to state senators. Thus, the three-year terms of territory senators mirror those of members of the House of Representatives and the number of senators to be elected for each territory is based on their House of Representatives entitlement, which is the reverse of the situation with the original states. This latter arrangement ties the number of Senate seats allocated to the territories to their population, a connection that does not exist in the case of the original states. Finally, with the terms of both senators from each territory ending simultaneously, the territories are not afforded the continuous Senate representation enjoyed by the states under the half-Senate rotation system.

While the constitutionality and desirability of admitting senators for the territories now appear to be settled questions, the terms of their service remain a matter of concern. As detailed above, when first introducing legislation to enable territory representation in the Senate, Whitlam stated that his proposal tied their terms to those of the House of Representatives because he believed this reform should be enacted for the entire Senate—that is, he believed that Senate terms should be equivalent to two consecutive House of Representatives terms. Although the Whitlam Government was ultimately able to introduce a one-term version of this arrangement for territory senators as this did not require constitutional change, proposed constitutional amendments affecting the whole Senate have been defeated at referendums on three occasions. The prospect of the rest of the Senate moving to terms tied to those of the House of Representatives therefore appears remote.

Furthermore, whatever the practical prospects of achieving such a reform, there are strong reasons to resist closer ties between the two chambers. As Odgers’ argues, such a change would:

… fundamentally alter the nature of bicameralism in the Commonwealth Parliament by removing one of its essential features, the principle of fixed, periodical elections, with a fixed, autonomous electoral cycle for the Senate. To lock the Senate into an electoral cycle dependent upon general elections for the House of Representatives, which can occur at any time, would significantly weaken its position as an independent house, and dilute its capacity to embrace electoral opinion which goes unrepresented in the method used for electing members of the House of Representatives. It would also remove a significant restraint on governments holding early elections for partisan reasons.[91]

While this aspect of territory representation in the Senate may have appeared 40 years ago as a harbinger of constitutional change affecting the fixed electoral cycle of the entire Senate, it now appears as an anomaly. Given the importance of maintaining the distinct electoral cycles of the two chambers of the federal parliament, a case could be made for aligning the terms of territory senators with those of senators for the states.

---

[1] Harry Evans and Rosemary Laing (eds), Odgers’ Australian Senate Practice, 13th edn, Department of the Senate, Canberra, 2012, p. 115.

[2] ibid.

[3] As at 2011, the populations of Australia’s inhabited territories were: Christmas Island, 2,072; Norfolk Island, 1,796; Territory of Cocos (Keeling) Islands, 550; Australian Capital Territory, 356,586; Northern Territory, 211,945. See Australian Bureau of Statistics, 2011 Census of Housing and Population, www.abs.gov.au/websitedbs/censushome.nsf/home/census?opendocument &navpos=10, accessed 20 August 2015.

[4] For a summary of the acquisition processes followed in the case of each external territory see Alan Kerr, A Federation in These Seas: An Account of the Acquisition by Australia of Its External Territories, Attorney General’s Department, Canberra, 2009, p. 400, http://regional.gov.au/ territories/publications/. For an account of the legal regimes applying to each external territory see also Australian Law Reform Commission, ‘External territories’ in Legal Risk in International Transactions, ALRC report 80, 1996, www.alrc.gov.au/publications/alrc-80-legal-riskinternational- transactions/10-external-territories.

[5] The high cost of administering and developing the Northern Territory made its surrender to the Commonwealth an attractive prospect for South Australian politicians, while federal politicians considered that the development of the Northern Territory was vital to the defence of the new federation and would also be of economic advantage. Alistair Heatley comments that the transfer of control to the Commonwealth was: ‘a lengthy process, extending for almost a decade (1901 to 1910); it was characterised by hard bargaining, political vacillation on both sides, and a splendid amount of posturing.’ (Alistair Heatley, Almost Australians: The Politics of Northern Territory Self- Government, ANU North Australia Research Unit, Darwin, 1990, p. 4.)

[6] Margaret Healy, ‘Territory representation in the Commonwealth Parliament’, Research Note (Department of the Parliamentary Library), no. 8, 2000–01, p. [1], http://parlinfo.aph.gov.au/parlInfo/download/library/prspub/GRC26/upload_binary/grc265.pdf.

[7] Australian Electoral Commission, ‘Calculating representation entitlements of states and territories’, 13 November 2014, www.aec.gov.au/electorates/Redistributions/calculating-entitlements.htm, accessed 17 August 2015.

[8] Brian Opeskin, ‘Constitutions and populations: how well has the Australian Constitution accommodated a century of demographic change?’, Public Law Review, vol. 21, no. 2, June 2010, p. 131. The Commonwealth also possessed the following external territories for various periods during the twentieth century: Papua (1906–42), New Guinea (1920–42), Papua and New Guinea (1942–71), Papua New Guinea (1971–75), Nauru (1920–68). See Kerr, op. cit., pp. 12–121, 160– 95.

[9] Australian Law Reform Commission, Legal Risk in International Transactions, ALRC report 80, 1996, paragraphs 10.18–10.20, www.alrc.gov.au/sites/default/files/pdfs/publications/ ALRC80.pdf, accessed 17 August 2015.

[10] Australian Electoral Commission, ‘Profile of the electoral division of Lingiari’, 23 September 2013, www.aec.gov.au/profiles/nt/lingiari.htm, accessed 17 August 2015; Australian Electoral Commission, ‘NT Division—Lingiari’, 10 October 2013, http://results.aec.gov.au/17496/ website/SenateDivisionPollingPlaces-17496-306.htm, accessed 17 August 2015.

[11] Australian Electoral Commission, ‘Norfolk Island electors’, 23 April 2013, www.aec.gov.au/ Enrolling_to_vote/special_category/Norfolk_Island_electors.htm, accessed 17 August 2015.

[12] Australian Electoral Commission, ‘Polling place—Norfolk Island’, 10 October 2013, http://results.aec.gov.au/17496/website/SenatePollingPlaceFirstPrefs-17496-65739.htm, accessed 17 August 2015.

[13] Explanatory Memorandum to Norfolk Island Legislation Amendment Bill 2015 (Cth) and associated bills, p. 2.

[14] Journals of the Senate, 14 May 2015, p. 2606.

[15] Rather than drafting a large volume of Commonwealth legislation from scratch to address responsibilities ordinarily covered by state legislation, the Commonwealth has in the past adopted state laws and applied them as its own to external territories. See Explanatory Memorandum to Norfolk Island Legislation Amendment Bill 2015 and associated bills, pp. 2–3.

[16] Department of Infrastructure and Regional Development, ‘FAQs—governance’, 20 March 2015, www.regional.gov.au/territories/norfolk_island/reforms/faq/governance.aspx; for discussion of the relationship between Norfolk Island electors and the mainland divisions of Canberra and Lingiari, see also Ms Gai Brodtmann, House of Representatives Hansard, 12 May 2015, pp. 3720–2, and the Hon. Mr Warren Snowdon, House of Representatives Hansard, 12 May 2015, pp. 3714–16.

[17] Legislative Assembly of Norfolk Island, Minutes of Proceedings, 20 May 2015, p. 126, http://norfolkisland.gov.nf/la/Minutes%20of%20Proceedings/14th%20Legislative%20Assembly/.

[18] This point was noted in the second reading debate on the Norfolk Island Legislation Amendment Bill 2015 by the Hon. Mr Warren Snowdon, House of Representatives Hansard, 12 May 2015, p. 3716.

[19] Joint Select Committee on Electoral Reform, Determining the Entitlement of Federal Territories and New States to Representation in the Commonwealth Parliament, November 1985, p. 18.

[20] Northern Territory Representation Act 1922, section 5, www.comlaw.gov.au/Details/ C1922A00018.

[21] House of Representatives Hansard, 13 September 1922, p. 2199 (Sir Granville Ryrie).

[22] See amended versions of the Northern Territory Representation Act 1922 at: www.comlaw.gov.au/Series/C1922A00018/Amendments.

[23] See amended versions of Australian Capital Territory Representation Act 1948 at: www.comlaw.gov.au/Series/C1948A00057/Amendments.

[24] Healy, op. cit.

[25] Australian Electoral Commission, ‘Electoral newsfile 115: two electoral divisions in the Northern Territory’, 7 February 2011, www.aec.gov.au/About_AEC/Publications/Newsfiles/2004/ No_115.htm, accessed 17 August 2015.

[26] Heatley, op. cit., p. 9.

[27] Senate Hansard, 9 September 1920, p. 4336, 15 September 1920, pp. 4527–30, 16 September 1920, pp. 4587–615, 4617–25.

[28] Senate Hansard, 15 September 1920, p. 4529 (Senator the Hon. Edward John Russell).

[29] Heatley, op. cit., p. 2.

[30] Journals of the Senate, 16 September 1920, p. 156.

[31] These figures include estimates of the Indigenous population of the two territories at the time. See note (a), Australian Bureau of Statistics, Australian Historical Population Statistics 2014, Population size and growth, 1.1 Population by sex, states and territories, cat. No. 3105.0.65.001 2014, www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/3105.0.65.0012014?OpenDocument, accessed 25 August 2015.

[32] Between 1926 and 1931 the Northern Territory ceased to exist and was replaced by two territories, Northern Australia and Central Australia. Following the bombing of Darwin, the Northern Territory was under military administration from 1942 to 1946. See Nicholas Horne, ‘A chronology of Northern Territory constitutional and statehood milestones 1825–2007’, Parliamentary Library, 31 May 2007, www.aph.gov.au/About_Parliament/Parliamentary_Departments/Parliamentary_Library /Publications_Archive/online/NTConstitutionalMilestones, accessed 17 August 2015; also see Heatley, op. cit., pp. 9–11.

[33] Legislative Assembly of the Northern Territory, ‘History of the Northern Territory Parliament’, www.nt.gov.au/lant/about-parliament/history-of-nt-parliament.shtml, accessed 17 August 2015; Horne, op. cit.

[34] Legislative Council for the Northern Territory, ‘The Remonstrance’, 26 February 1962, www.territorystories.nt.gov.au/bitstream/10070/213543/3/Remonstrance_1962.pdf, accessed 18 August 2015.

[35] Legislative Assembly for the Australian Capital Territory, ‘The establishment of self-government in the ACT’, www.parliament.act.gov.au/__data/assets/pdf_file/0007/367018/Assembly-Factsheet-3- June-2013.pdf, accessed 20 August 2015.

[36] Philip Grundy, Bill Oakes, Lynne Reeder, Roger Wettenhall, Reluctant Democrats: The Transition to Self-Government in the Australian Capital Territory, Federal Capital Press, Canberra, 1996, p. 2.

[37] ibid, p. 3.

[38] Australian Labor Party, Platform, Constitution and Rules, 1967, pp. 9–10, http://parlinfo.aph.gov.au/parlInfo/download/library/partypol/1040068/upload_binary/1040068.pdf, accessed 18 August 2015.

[39] Paul Kelly, ‘Labor goes in to bat for its old nemesis’, Australian, 20 August 2003; Alan Ramsey, ‘Abolish the Senate? The no’s have it’, Sydney Morning Herald, 30 October 2004, pp. 42–3.

[40] Australian Labor Party, op. cit., p. 31.

[41] House of Representatives Hansard, 7 November 1968, pp. 2581–2 (Mr Gough Whitlam).

[42] ibid., p. 2583.

[43] ibid.

[44] Evans and Laing, op. cit., p. 11.

[45] ibid., pp. 137–8.

[46] Joint Committee on Constitutional Review, Report, October 1958, p. 10.

[47] House of Representatives Hansard, 7 November 1968, p. 2584 (Mr Gough Whitlam).

[48] On the topic of three-year terms for territory senators, the sixth edition of Odgers’ contains the following commentary: ‘it is a pity that the terms of service of territory Senators (the life of the House of Representatives) are out of step with State Senators. The purpose of six year terms for members of the Upper House is as important for territory Senators as for State Senators, the question of rotation aside. It is suggested for consideration that the Senate (Representation of Territories) Act 1973 be amended to provide six year terms for territory Senators.’ (J.R. Odgers, Australian Senate Practice, 6th edn, Royal Australian Institute of Public Administration, Canberra, 1991, p. 123.)

[49] Parliamentary Library, ‘Federal election results 1901–2014’, www.aph.gov.au/About_Parliament/ Parliamentary_Departments/Parliamentary_Library/pubs/rp/rp1415/FedElect, accessed 18 August 2015.

[50] House of Representatives Hansard, 7 November 1968, p. 2583 (Mr Gough Whitlam).

[51] ibid.

[52] House of Representatives Hansard, 22 May 1973, pp. 2430–1 (the Hon. Mr Fred Daly).

[53] ibid. The provisions for these calculations are now located in the Commonwealth Electoral Act 1918, section 48.

[54] House of Representatives, Votes and Proceedings, 30 May 1973, p. 240.

[55] Journals of the Senate, 7 June 1973, pp. 273–5.

[56] House of Representatives, Votes and Proceedings, 25 September 1973, p. 379, 27 September 1973, pp. 397–8.

[57] Journals of the Senate, 14 November 1973, pp. 480–1.

[58] House of Representatives, Votes and Proceedings, 10 and 11 July 1974; Journals of the Senate, 10– 11 July 1974, 16–18 July 1974, 23–24 July 1974.

[59] Joint Sitting Hansard, 6 August 1974, pp. 87–8; Minutes of Proceedings of Joint Sitting, 6 August 1974, pp. 3–5. The bills in question were ‘affirmed’ at the joint sitting rather than ‘agreed to’ in accordance with the terminology of section 57 of the Constitution, which states in part: ‘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 both Houses of the Parliament, and shall be presented to the Governor-General for the Queen’s assent.’

[60] Odgers, op. cit., p. 120.

[61] This argument was put forward during debate on the Territory Senators Bill 1970 by the Member for the Australian Capital Territory, Mr Kep Enderby. See House of Representatives Hansard, 20 August 1970, p. 289. The Senate has taken the lead role in scrutinising delegated legislation since establishing the Standing Committee on Regulations and Ordinances in 1932. See discussion at Evans and Laing, op. cit., pp. 416–18.

[62] Odgers, op. cit., pp. 120–1.

[63] Harry Evans, ‘The role of the Senate’, Reform, issue 78, 2001, pp. 16–20, www.austlii.edu.au/au/ journals/ALRCRefJl/2001/4.html, accessed 18 August 2015.

[64] ibid.

[65] Evans and Laing, op. cit., p. 1.

[66] Dr John Cockburn, Debates of the Australasian Federal Convention, 30 March 1897, p. 340.

[67] Western Australia v Commonwealth (1975) 134 CLR 201; see also ‘High Court decision: senators for ACT upheld’, Canberra Times, 11 October 1975, p. 1.

[68] Leslie Zines, The High Court and the Constitution, 4th edn, Butterworths, Sydney, 1997, p. 440.

[69] For a summary and discussion of the various arguments deployed in this case see Leslie Zines, The High Court and the Constitution, op. cit., pp. 467–71.

[70] Western Australia v Commonwealth (1975) 134 CLR 201, 227, 232.

[71] Western Australia v Commonwealth (1975) 134 CLR 201, 270.

[72] Western Australia v Commonwealth (1975) 134 CLR 201, 275.

[73] Leslie Zines, The High Court and the Constitution, op. cit., p. 467.

[74] Leslie Zines, ‘Representation of territories and new states in the Commonwealth Parliament’, a paper prepared for Standing Committee D for the Australian Constitutional Convention, published as Appendix H in volume 2 of the 4th Report of Standing Committee D of the Australian Constitutional Convention, August 1982, pp. 5–6.

[75] ‘Before indicating my opinion as to the correct answers to these questions, it should be noted that two States during the argument of these proceedings questioned the propriety of the court’s decision in Western Australia v. the Commonwealth. However, unfortunately as I think, neither State proffered any argument in support of this questioning. I say unfortunately because, if the decision is to be reconsidered, that reconsideration should take place before what, with due respect to the opinion of others, appears to me to be a serious departure from the federal nature of the Constitution, becomes entrenched in constitutional practice by the mere passage of time.’ Attorney- General (NSW) ex rel McKellar v Commonwealth (1977) 139 CLR 527, 533 [4].

[76] Queensland v Commonwealth (1977) 139 CLR 585.

[77] Encyclopaedic Australian Legal Dictionary, ‘stare decisis’, LexisNexis Australia, January 2011.

[78] Leslie Zines, ‘Representation of territories and new states in the Commonwealth Parliament’, op. cit., p. 3; For a discussion of the reasons given, both for and against, relying on the principle of stare decisis in this case, see Leslie Zines, High Court and the Constitution, op. cit., pp. 440–1. Zines notes the following reasons in favour of not overturning the decision in the First Territories Senators Case: it was a recent decision, it had been fully argued, there were no new arguments presented in the Second Territories Senators Case, and senators had in fact been elected. Justifications for overturning the decision included the following: it was a recent decision, it had not been followed in other cases, it was by a narrow majority, and it was of great constitutional importance.

[79] Joint Select Committee on Electoral Reform, op. cit., p. 18.

[80] The fear that territory senators may ‘swamp’ the parliament was a matter raised by some justices in High Court judgments, but others countered with the view that this was equally possible via section 121 in the case of new states, and that this possibility was not something the court should concern itself with. Rather, it must be assumed the parliament will act responsibly. See Mason J argument at Western Australia v Commonwealth, 134 CLR 201, 271.

[81] Western Australia v Commonwealth, 134 CLR 201, 275.

[82] Western Australia v Commonwealth, 134 CLR 201, 271.

[83] Australian Constitutional Convention, Standing Committee D, 4th report, 1982, p. 46.

[84] Joint Select Committee on Electoral Reform, op. cit., p. 2.

[85] ibid., pp. 46–7.

[86] ibid., pp. 47–8.

[87] ibid., p. 49.

[88] ibid., p. 45.

[89] ibid., p. 55.

[90] House of Representatives Hansard, 22 December 1989, p. 3536 (Mr Stewart West).

[91] Evans and Laing, op. cit., p. 33.