Pulling the Trigger: The 1914 Double Dissolution Election and Its Legacy
It is not a small thing to exercise a constitutional power for
the first time. Yet, in June 1914, when Prime Minister Joseph Cook advised the
Governor-General, Sir Ronald Munro Ferguson, to simultaneously dissolve both of
the Commonwealth houses of parliament, following the procedure set out in
section 57 of the Constitution, he did so with a surprising confidence, even
bravado, which seemed to belie the fact of novelty and experiment.
In 1914, the Constitution was still very young, not quite in
its infancy, but not yet fully grown, and far from fully tested. Many sections
had been tried out for the first time in the decade since the inauguration of
the Commonwealth, on New Year’s Day, 1901. Constitutional novelty was an
unavoidable feature of the era. But section 57 was not just an untried
provision. It was unprecedented, indeed unique, in the history of bicameral
legislatures. The problem it was intended to resolve—deadlocks between two
legislative houses—was not unfamiliar; indeed, it was well-known in Australian
colonial history and easy to characterise. But there was nothing familiar about
the backdrop to the deadlocks that were anticipated to arise between the House
of Representatives and the Senate, and little to guide the specific design of
There was also nothing straightforward about the evolution of
the section. Section 57 was adopted after one of the lengthiest and most
tortuous debates in the Federal Convention of 1897–98 at which the bill that
became the Constitution of the Commonwealth of Australia was drafted. Numerous
complicated formulae were tried out, multiple amendments were proposed, and the
vote was repeatedly taken, but just as soon as it appeared that something had
been settled, another iteration was suggested and the debate began again.
Indeed, at its peak, the debate on what emerged as section 57
occupies four hundred pages of the official record of the Convention debates.
Just one week before the Convention finally wound up, in March 1898, the
delegates finally agreed on its wording. Even then, as we shall see, it wasn’t
The Australian Commonwealth is built around two great
institutional principles: representation of the people of the nation in a
system of British parliamentary government, and equal representation of the
states in a system of American-style federalism. These two principles find
their expression respectively in the House of Representatives and the Senate.
They work together, in a lopsided, power-sharing arrangement, complementary but
constantly in tension: an asymmetrical symmetry, we might say.
What we see in section 57 is a reflection of this arrangement.
It captures the multiple interests that were at stake in the project of
federating Australia’s self-governing colonies into one ‘indissoluble federal
Commonwealth’. It displays, in particular, the tensions. The section provides
for the resolution of otherwise unresolvable disagreements—deadlocks—between
the houses of parliament via an election in which all seats of the House of
Representatives and all Senate places are simultaneously re-contested. In the
normal course of constitutional business, only half the Senate is elected every
three years, and a half-Senate election may, but does not have to, occur at the
same time as an election for the House of Representatives.
Section 57, in contrast, contemplates the extraordinary:
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.
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 both Houses of
the Parliament, and shall be presented to the Governor-General for the Queen’s
The first paragraph describes the circumstances in which a
deadlock arises. It sets out what have come to be known as the ‘triggers’.
First, the Senate rejects or fails to pass a bill (a ‘proposed law’) that has
been passed in the House, or the Senate passes a bill with amendments that the
House will not accept. Then, after an interval of three months, if the Senate
again rejects the same bill or fails to pass it or passes it with unacceptable
amendments, the Governor-General may dissolve both houses simultaneously, but
not less than six months before a regular House of Representatives election is
The second paragraph defines what may follow if the deadlock
on the same bill continues after a double dissolution election has taken place:
namely, the convening of a joint sitting of both houses of parliament.
The third paragraph describes how the joint sitting will
proceed, and how the vote on the deadlocked bill is to be counted—that is, the
steps required for the previously deadlocked bill to become law.
We note several salient features. The deadlocked bill must
originate in the House of Representatives. Section 57 cannot apply to bills
that originate in the Senate. The reference to passing a bill with unacceptable
amendments cannot apply to money bills, since, under section 53 of the
Constitution, the Senate cannot amend ‘proposed laws imposing taxation, or ...
appropriating revenue or moneys for the ordinary annual services of the
The House of Representatives does not have to take any action;
it does not have to treat the Senate’s rejection of, or failure to pass, a bill
as a step towards a double dissolution. The House is not obliged to reintroduce
the deadlocked bill after a dissolution, and a subsequent joint sitting is not
mandated. The Governor-General’s role is central. He or she dissolves both
houses and convenes a joint sitting, if this takes place.
The stipulations in the section appear precise, but this
precision is deceptive. Notwithstanding the length of the provision—it is one
of the wordiest sections of a mostly very economical Constitution—numerous
essential matters are missing from its text. Although its framers laboured long
and hard on its wording, many questions were left unanswered when the provision
reached its final form. These included:
What constituted a ‘failure to pass’?
How was the interval of three months to be calculated? From when
did it start to run?
Could the rejection of any proposed law, on any subject matter,
qualify as a double dissolution trigger—or did the rejected bill have to be on
a matter essential to government, or central to the government’s program?
Did the provision allow for only one bill to be the subject of a
double dissolution, or could multiple bills serve that purpose? Could rejected
bills be ‘stockpiled’, stored up by a frustrated government for potential
resolution at a joint sitting?
Could a government deliberately ‘manufacture’ the conditions for
a double dissolution, effectively enticing the Senate to twice reject a bill?
Does the Governor-General have to follow the advice of the prime
minister to dissolve both houses?
Or does the Governor-General have to satisfy him or herself
personally that the preconditions of section 57 have been met, or on the
general workability of government?
Whom could the Governor-General consult in determining whether to
grant a double dissolution?
Little by little, over the decades, and in the course of
Australia’s six double dissolutions,
answers to these questions became known, although even now, not all are clear.
In 1914, notwithstanding how recently the Constitution had been completed,
notwithstanding the fact that many who sat in parliament, and fully four of the
seven sitting Justices of the High Court, had been delegates at the Federal
Conventions of the 1890s, the scope of the provision remained uncertain.
On 2 June 1914, meeting with the Governor-General, Prime
Minister Cook explained the legislative history of one particular bill—the Government
Preference Prohibition Bill (a bill to prohibit union preference in
Commonwealth employment)—which had been rejected twice by the Senate, with an
unambiguous three months interval between the two rejections. Cook explained
that obstruction in the Senate had made government unworkable. He advised the
Governor-General to dissolve both houses, as provided for in section 57.
What, if any, guidance did the Governor-General have on how to
proceed? It was well known that there had been constitutional crises in the
colonial parliaments, associated with bills deadlocked between the houses. In
few of these, however, was the crisis over ordinary legislative bills. Most involved
deadlocks over money bills and were therefore potential roadblocks for
government. But Cook’s Government Preference Prohibition Bill in no way
resembled a money bill or threatened the viability of government.
In any case, most colonial precedents involved upper houses
that were appointed, not elected. At that time, in South Australia alone both
houses were elected. In 1881, a provision had been inserted in the South
Australian Constitution specifically to deal with the settlement of deadlocks
between its houses; one of the measures it contemplated was a double
dissolution. But such had never taken place.
Even had it done, however, the deadlock issue was different; neither the South
Australian Legislative Council, nor any of the other colonial upper houses had
been designed, as the Senate was, to represent different regional interests or
There were no historical lessons either from other countries.
Australia’s parliament was unique in having both houses directly elected. Double
dissolution procedures could have no application in bicameral systems with only
one elected house. There were no indirect analogies to be made, either. The
British House of Lords was, of course, unelected, and Britain’s then very
recent constitutional crisis, 1909–11, had been resolved with an Act limiting
the House of Lords’ power to obstruct bills from the Commons. But such an Act
was simply not constitutionally available to the Australian Parliament. The
Canadian Senate was not an elected house either. The United States Senate was
unelected until the ratification of the 17th amendment in 1913 which
transformed it into a directly elected house, but the first Senate election did
not take place until November 1914, and no deadlock issue arose.
The request for a double dissolution
Lack of precedents notwithstanding, Prime Minister Cook had
firm views about the application of section 57 to the circumstances facing his
government. Cook had not been a delegate at either of the Federal Conventions,
but he had worked closely with colonial politicians who were, including in the
1890s in the cabinet of NSW Premier George Reid (who was to take the lead in
the final shaping of section 57). In 1901, in common with Reid (later
Australia’s fourth prime minister) and many other members of the Federal
Conventions, Joseph Cook entered the new Commonwealth Parliament, representing
the NSW seat of Parramatta.
In colonial politics and for the first few years in the life
of the Commonwealth, federal politics had been organised around two major
parties—their names and purpose now lost to popular memory—the Protectionists
and the Free Traders. But the Labor Party was already a significant
alternative. By 1903, the three parties stood, effectively, shoulder to
shoulder. Alfred Deakin famously called them the ‘three elevens’. By 1909, it
had become clear to the non-Labor side that they had more in common as
opponents of Labor than against each other; the Free Traders and the
Protectionists amalgamated, or ‘fused’. Cook became deputy leader of the new
Fusion Party (soon after known as the Federal Liberal Party) led by Alfred
In 1913, Deakin stepped down and Cook took his place; the
Liberals defeated the Fisher Labor government that same year. It was a fragile
victory. The new government’s margin in the House of Representatives was the
narrowest possible: one seat alone. Cook, now prime minister, tried to entice
the outgoing Labor Speaker to stay on, but was unsuccessful. Thereafter, the
Liberals had to rely on the casting vote of their own Speaker for the passage
of bills. Even more frustrating was the lack of balance in the Senate. The
Opposition was firmly in command. Of the 36 Senate places, 29 were Labor; seven
alone were Liberal.
From the start, both parties had their eye on the strategic
possibilities in this arrangement. The Senate repeatedly obstructed where it
could, including refusing to grant pairs. Less than a year after forming
government, Cook sought, and was granted, a double dissolution. The parliament
was dissolved on 30 July 1914, and the election set for 5 September.
In opening his election campaign, Cook dwelled on his
government’s frustration at the hands of the Senate, describing the calling of
a double dissolution as ‘the only honourable course to take’:
Our opponents never once, ‘played the game’. From the first
day we had to meet a hurricane of virulent abuse and bitter recrimination ...
[I]n the Senate, the one place of all where one would expect calm and
reasonable consideration of the country’s affairs, [there has been] insolence
and over-bearing arrogance ... Altogether, the performances of these gentlemen
have made a perfect parallel to one of Gilbert and Sullivan’s operas.
Cook assured his audience that:
it was only after exhausting all the possibilities of the
situation that we decided to clear the decks, alter the course, and steer the
ship of State back to the port of public opinion from which it had set out on
its tempestuous voyage a year ago.
‘Any action’, he concluded, ‘which does not finally and firmly
uphold the right of the democracy to sovereign power, and compel the Senate
ultimately to bow to that power, would be a betrayal of the people’.
Cook made no attempt to disguise the fact that, almost
immediately after the 1913 election, he had devised a plan to activate section
57. Barely months into government, he had two bills drafted that he knew to be
contrary to Labor policy, and that he openly and calculatingly planned to offer
as double dissolution ‘triggers’: a bill to restore postal voting, and the
Government Protection Prohibition Bill. The second, as intended, was a red rag
to the Senate. Like clockwork, the bill was passed in the House, rejected in
the Senate, passed again in the House three months later, and rejected a second
time. The last federal election had been only a year earlier; there was no
danger of running into the six months cut-off before a regular election was due
‘by effluxion of time’.
All appeared to be in place, and Cook was, it seems, entirely
confident that the Governor-General would act as advised and dissolve both
houses. Indeed, Cook was of the view that the Governor-General was obliged
to act on the advice of the prime minister, that he had no discretion in the
The role of the Governor-General
Between the first Senate rejection and the request for a
double dissolution, a new Governor-General had arrived in Australia. Sir Ronald
Munro Ferguson was no stranger to politics. He had behind him thirty years’
service as a Liberal member of the House of Commons, including as parliamentary
private secretary to the one-time British Prime Minister, Lord Rosebery. He
was, in addition, a man whose vigorous, independent temperament had made the
British Government more than happy to offer him the role of representative of
the Crown in a far-distant dominion. His favourite form of exercise, we are
told, was chopping down trees.
Before his arrival, Munro Ferguson had been following the
Australian political scene, receiving information from the outgoing Governor-General,
Lord Denman, and consulting with the former prime minister, Sir George Reid,
now High Commissioner in London, on how to proceed in the likely event of a
request for a double dissolution. It was clear that the technical preconditions
surrounding section 57 had been satisfied; the major unanswered question about
the section’s operation was whether he, as Governor-General, could exercise any
discretion in the matter.
Cook’s view, expressed in a memo prepared by his
Attorney-General, was that the representative of the Crown was obliged to
follow ministerial advice. This opinion was fortified by the recent writings of
the leading British constitutional authority, Arthur Berriedale Keith.
But, there were alternative perspectives. Keith’s views were based on British
conventions, and formed in relation to the dissolution of lower houses. The
powers of Australia’s Governors-General were to be exercised in a different
context, and were complicated by the difference between the constitutional
status of the Governor-General and the monarch.
Until 1926 at least, when constitutional equality between the
Dominions and the United Kingdom was declared, Governors-General were,
effectively, dual representatives: both the constitutional representative of
the Queen or King, and agent of the British Government. The conventional
restraints upon the exercise of discretion by the monarch regarding a prime
minister’s advice (so vividly illustrated in the British constitutional crisis)
did not necessarily apply in Australia. Indeed, in the decade since federation,
as Munro Ferguson knew, Governors-General had refused a prime minister’s
request to dissolve the House of Representatives three times (1904, 1905,
1909), following the government’s loss of the confidence of the House.
But Cook’s was a request for a double dissolution. The
government had not lost the confidence of the House. And there were textual,
not merely conventional, guidelines that appeared at least to offer guidance.
Section 57 stated that, following certain preconditions, ‘the Governor-General
may dissolve the Senate and the House of Representative simultaneously’. Just
what did that ‘may’ signify?
Munro Ferguson had determined to satisfy himself personally
that the conditions for a double dissolution had been met. In addition, he
wanted to be satisfied that government had become unworkable, and that no
alternative government could be formed. To satisfy the last, he proposed to
consult with the Leader of the Opposition, Andrew Fisher. In London, George
Reid had advised him that he could, and Munro Ferguson raised the idea with the
prime minister. Cook was not happy, and the Governor-General gave way.
The proposal to consult with members of the High Court
received a warmer response. Sir Samuel Griffith, Chief Justice of Australia
(and leading member of the 1891 National Australasian Convention), was more
than happy to be consulted. Griffith joined the Governor-General for lunch the
day after the meeting with the prime minister and, at the request of the
former, prepared a memo setting out his own opinion of the scope of the section
57 power. He later wrote a second version of the memo, anticipating the tabling
in parliament of the Governor-General’s correspondence on this matter.
The Chief Justice wrote: ‘An occasion for the exercise of the
power of double dissolution under Section 57 formally exists ... whenever the
event specified in the Section has occurred, but it does not follow that the
power can be regarded as an ordinary one which may properly be exercised
whenever the occasion formally exists’: there was, that is, an apparent
operational power, but in reality, a discretionary power. The power, the Chief
Justice continued, was only to be exercised when the Governor-General was
after independent consideration of the case, either that the
proposed law as to which of 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 a state of practical deadlock in legislation as
can only be ended in that way.
On these matters, the Governor-General could not act without
the advice of his ministers, but he was not bound to follow their advice. He
was, Griffith concluded, an ‘independent arbiter’.
Cook did not deviate from his view that the opposite was the
case. He considered the origins of section 57, and claimed that it was ‘not
reasonable to suppose that the framers of the Constitution intended to place
the responsibility of granting or refusing the double dissolution upon the
Governor-General personally—to place him in the invidious position of appearing
to take sides with one House or the other, or with one political party or the
When Munro Ferguson granted the dissolution, Cook chose to
interpret this as a vindication of his own views. Munro Ferguson, on the other
hand, considered his actions to be an exercise of his vice-regal discretion.
Andrew Fisher, who held the ‘independent arbiter’ view (the Governor-General
should ‘read the Constitution ... and act accordingly’, he said) had chafed at
the Governor-General’s back-down on his proposal to consult the Opposition
leader. However, although concluding that Munro Ferguson had improperly
followed Cook’s advice, he agreed with the decision. Everyone, it seems,
thought they had got the result they wanted.
The 1914 election
The outcome was another matter. Cook’s confident strategy for
gaining control of the Senate failed abysmally. With a 73.5 per cent voter
turnout (in the days of voluntary voting), Labor attracted a 2.42 per cent
swing and gained a convincing majority in the House of Representatives, winning
42 of the 75 seats. While Labor lost no seats, the Liberals lost six: five to
Labor and one to the sole independent MHR. Not only did Labor retain control of
the Senate, it increased its control, winning 31 of the 36 Senate places. Cook,
back as Leader of the Opposition, but ever cheerful, would describe the result
as a ‘jolly good licking’.
The 1914 election returns strikingly illustrate the shift in
Australian party politics that had occurred since federation: out of 75 House
of Representatives electorates, only three offered any alternative candidate to
either Labor or Liberal. The alternatives were all independents (including, in
Victoria’s electorate of Kooyong, where the Commonwealth’s first female
candidate, Vida Goldstein, stood as an independent).
It is also striking that, in 13 of the 75 electorates, candidates (seven Labor
and six Liberal) were elected unopposed. The two political parties, both
relatively new, were now absolutely dominant.
The election had made something else clear: the deadlocked
bill was dead in the water. There would be no prospect of its being
reintroduced and no need for a joint sitting of the houses to resolve any
further deadlock. So, had section 57 served its purpose?
The framing of section 57
What was the purpose of section 57? We can say that it
was to provide a means of breaking deadlocks in the passage of legislation, but
this describes only the problem that the section was intended to address. The
other sense of purpose—what it represented in the minds of the framers—was
never entirely settled. For one thing, a higher or ulterior purpose is difficult
to put into words with practical application; and secondly, most significantly,
the framers held many different objectives and understandings about the
Was the section intended to ‘punish’ the Senate for repeated
obstruction to an elected government (as Joseph Cook appeared to think)? Or was
it intended to hold the House of Representatives or the elected government to
account, forcing it to test public approval for proposed legislation,
specifically legislation that was contrary to the interests of the states? Was
it, in short, an affirmation of the ultimate sovereignty of the House of
Representatives, or of the federal equality of the states? There was simply no
consensus among the Constitution’s framers.
The need for such a provision was hinted at in the first
Federal Convention, in Sydney in 1891. It arose in debate on the fiscal powers
of the respective houses of parliament—an issue that was to be the most
contentious and controversial of all. The equal representation of each state in
the Senate, regardless of population (as in the United States), had raised some
objections among the delegates from the more populous colonies, but it had been
conceded relatively early. Indeed, without this concession, there would have
been little prospect of accomplishing federation. But, legislative equality
between the houses was not so easily conceded.
The Convention finally agreed that the Senate should have
co-equal powers regarding all non-fiscal legislation, but would be prohibited
from originating or amending money bills. However, it was permitted to
recommend amendments, or to fail to pass such bills. (We see these arrangements
now in section 53 of the Constitution.)
In 1891, debate on the Senate’s power to suggest amendments to
money bills prompted Victorian delegate, Sir Henry Wrixon, to raise the
prospect of deadlocks between the houses. The Senate, at that time, was
conceptualised as an appointed chamber, and the option of a double dissolution
was therefore not available, but the recognition of tensions between the
different principles embodied in the two houses was growing and already pushing
delegates to offer procedural solutions. Wrixon suggested a provision: ‘If the
house of representatives decline to make [an] ... omission or amendment [as
suggested by the Senate], the senate may request a joint meeting of the members
of the two houses, which shall thereupon be held, and the question shall be
determined by a majority of the members present at such meeting’.
This procedure, he explained, would provide the ‘means of securing finality,
and ... settling a difference if it arises ... on such a critical measure as an
appropriation bill’. The
proposal was quickly negatived. Sir Samuel Griffith (Queensland delegate)
objected that Wrixon’s scheme would mean that ‘The senate need not ask for a
joint meeting unless it likes, and it would not ask for it unless it counted
heads and saw that it would have a majority; so that by his proposal the senate
would be able to coerce the house of representatives’. He objected, he said, to
any ‘artificial means of settling differences between the two houses’.
Victorian delegate, Nicholas Fitzgerald, also objected to what
he called ‘mechanical’ means, but he conjured up more honourable motives and
Can we have the great national life which we all say we shall
call into existence by federation without an enhanced sense of national honor?
Must not the two go together; and, if we have both, cannot we rely upon the
proper spirit and motives which will actuate the members of both houses, and
believe that questions of difference will not lead to confusion, and that the
members of the federal parliament will not be governed by the consideration of
party or personal politics, but by the interests of the country at large?
By the second Federal Convention, however, with the Senate now
conceptualised as an elected house, recognition of the need to provide for
potential deadlocks had grown. Still, the debate got off to an unpromising
start. At the Adelaide session of the Convention, NSW delegate, Bernhard Wise,
moved the first new deadlock proposal. It involved consecutive dissolutions;
first of the House of Representatives and then, if the Senate again rejected
the disputed bill, dissolution of the Senate. This arrangement, Wise said,
would preserve the independence of the Senate on matters affecting state
interests and at the same time secure dominance of the popular vote on national
matters. At this stage, it was clear that the primary concern was the
likelihood of deadlocks arising over money bills (as had happened in colonial
parliaments). It was a view that persisted for many years in some quarters at
least, and lent an interpretation to section 57, as intended specifically for
In this vein, responding negatively to Wise’s proposal, Edmund
Barton pointed out that the mixing of taxation measures with ordinary subjects
in legislation was to be prohibited by the Constitution (now in section 55),
and that a deadlock provision was therefore unnecessary. But Victorian
delegate, William Trenwith (the one labour representative at the Convention),
responded that deadlocks might also arise on non-fiscal bills. And this,
increasingly, was to be the tone of the Convention.
Indeed, late in the final session of the Convention, in
discussion of the appropriate majority for a bill to pass at a joint sitting,
Victorian delegate (and future High Court Justice), Henry Higgins, felt the
need to remind fellow delegates to ‘[k]eep an Appropriation Bill in your mind’.
This was, he said, ‘the most awkward case [of deadlock] which can arise, and it
is quite possible that it may arise’. But,
although appropriation and taxation bills were mentioned in debate with some
frequency, this reminder did not prove influential.
Wise’s proposal for a consecutive dissolution was lost.
Higgins then proposed a simultaneous dissolution. After this was defeated,
Victorian delegate (also a future High Court Justice and first Australian
Governor-General), Isaac Isaacs, joined by Victorian Premier, Sir George
Turner, moved a new provision: that there should be a popular referendum on a
deadlocked bill. If the bill were supported by the electors in a majority of
states containing a majority of the population, then—the proposal went—it was
to be submitted to the Governor-General for assent. This, too, was defeated.
Nothing on deadlocks was agreed at the Adelaide session of the
Convention and the draft Constitution bill that was completed there did not
include a deadlock provision. The various defeated resolutions, however, had
effectively set out the broad alternatives: dissolution of both houses, either
consecutive or simultaneous, following a deadlock, or a referendum on the
disputed bill, to be submitted to a ‘mass’ or ‘national’ vote, or alternatively
to a ‘dual’ vote, factoring both national and state majorities into the count.
The issue picked up steam in the break between Convention
sessions. In the long debate that followed, in the Sydney session, in September
1897, and in the final session, in Melbourne, from January to March 1898, many
different iterations appeared, some of which included a combination of several
or all alternatives. The break between the Adelaide and Sydney sessions had
been designed to allow the colonial legislatures to discuss the draft
Constitution bill, and to make suggestions for amendment. Among the 300 or so
suggestions, a good number of deadlock-breakers featured. The larger colonies
tended to support resolving deadlocks by referendum, but the smaller were
concerned about the prospect that a referendum would favour the more populous
states, and they tended to support consecutive, conditional dissolutions.
At the Sydney session of the Convention, Sir John Forrest,
Premier of Western Australia, vividly illustrated the preference for the
consecutive dissolution. Everyone knows, he said, that:
no case has arisen in Australia, nor anywhere else, in which
the upper house has been the aggressor, forcing upon the lower house some
measure of which it disapproved ... If any conflict occurs in the Commonwealth in
the future it will be caused by the house of representatives trying to coerce
The house that causes the trouble, he continued, should be the
one to go to the country. When ‘people ... have had time to work off their angry
passions’ and ‘cool down’, if the government has been returned at an election
of the House of Representatives, and the Senate still refused to give way, then
it could be sent to the country. Consecutive dissolution, Forrest concluded,
would make the House ‘much more careful, [and] much less eager to enter into
conflict’ with the Senate.
Higgins objected that Forrest’s proposal ‘simply makes the
house of representatives a cats paw in order to pull the nuts out of the fire’.
He went on to explain his allusion. The proposal, he said, ‘simply allows the
members of the senate to see by the voting in the different states which way
the feeling ... is going, and they will know exactly then as to whether it is or
is not worth while for them to face a dissolution’.
Gradually, however, support for the referendum proposal
weakened. Consecutive dissolution also gave way, albeit narrowly, to
simultaneous dissolution. Support for a joint sitting to follow a dissolution
now inched forward. Although Isaacs objected that a joint sitting would
effectively create a unicameral parliament as an arbiter of disputes, this
reasoning was not followed through, and the size of the majority that would
count for the passage of a previously deadlocked bill at a joint sitting became
the focus of debate. Finally, the provision was settled; the joint sitting was
adopted, and the majority for the passage of bills was set at three-fifths of
the members present. The survivor from an exhausting tussle, even this
provision did not endure.
The Constitution Bill, as adopted at the conclusion of the
Melbourne session of the Convention in March 1898, was sent for the voters’
approval in referendums in four colonies that year. In New South Wales the
referendum failed. While technically possible for federation to proceed with
the agreement of three colonies alone, it was politically inconceivable for the
others to go ahead without the ‘mother colony’. A special Premiers’ conference
met in January 1899, and George Reid was now in a position to extract
concessions from the other premiers, making the Constitution Bill more
attractive to the voters of his colony. The primary concession—what the people
of NSW really cared about—was that the Constitution should specify that the
federal capital site, when chosen, must be in NSW. But several other provisions
in the Constitution had concerned Reid, an advocate of a stronger House of Representatives.
One was the three-fifths majority in section 57. Reid now persuaded the others
to substitute ‘an absolute majority of the members of the Senate and House of
Representatives’, giving the House the decisive numerical advantage. Section 57
was finally finished. This was the version that went to the people in the new
round of colonial referendums, this time successfully, in 1899 and in Western
Australia in 1900.
Notwithstanding the multiple conflicting preferences expressed
in the deadlocks debate, the Convention delegates had all agreed on one thing.
They expected section 57 to be very rarely used. In this, they were proved
correct. Although the threat of a double dissolution has been invoked with
predictable regularity whenever the houses have clashed, the double dissolution
experiment of 1914 would not be repeated for another 37 years. In particular
since the introduction of proportional representation for Senate elections and
the rise of minor parties, governments have often enough faced a hostile
Senate, but only five more double dissolutions have occurred (in 1951, 1974,
1975, 1983, 1987). This total of six represents less than two per cent of
Australia’s 44 federal elections.
Of the six double dissolutions, half resulted in return of the
government; half in defeat. In one case alone (1951) did the returning
government gain control of the Senate. If mastering the Senate is the primary
motivation for contemplating a double dissolution, history suggests caution.
The year 1914
The 1914 election was scarcely the major event of that year. A
much larger, much more dramatic and tragic event was unfolding. The grant of
the double dissolution occurred in peacetime; the election itself occurred one
month after the declaration of war. Looking at the newspapers of that year, it
is surprising—a little disconcerting even—to see how long it took before the
realisation dawned that government would be a different matter for the
foreseeable future, and how little the imminence of war featured in the debates
surrounding the election.
It did not go unnoticed, however. There were disagreements
about preparation for war, reflected in the campaign speeches of the Prime
Minister and the Leader of the Opposition concerning funding for Australia’s
defence. Cook proposed borrowing, to finance the expansion of Australia’s
defence capacity; Fisher described this approach as ‘shameless’, as a
reflection of the Cook Government’s having squandered the surplus built up
previously by the Fisher Government. But these issues were swallowed up in
lists of many other promises and policy commitments: from the government,
expanded immigration, Commonwealth and state cooperation in managing the River
Murray, a program to build transcontinental railways and achieve a uniform rail
gauge, development of the Northern Territory and Papua, pension reform,
construction of lighthouses, anti-trust measures and more (Cook also,
interestingly, advocated proportional representation for Senate elections); and
from the Opposition: tariff reform, anti-trust measures, the expansion of
social welfare to provide for orphans, the establishment of a uniform rail
gauge, and a reminder of the Fisher Government’s previous achievements,
including the establishment of compulsory national military training.
It is unclear how much the declaration of war affected the
election outcome. Andrew Fisher famously vowed during the campaign that
Australia would ‘stand beside our own to help and defend Britain to the last
man and the last shilling’, but there was no evidence that Joseph Cook thought
otherwise. He, too, promised a commitment of all Australia’s ‘resources ... for
the preservation and the security of the Empire’.
By the eve of the election, the focus on the war had dramatically increased;
war measures, including recruitment for the Australian Imperial Force, had
already begun, notwithstanding that the parliament had been dissolved.
Recognising that the normal grant of funds for government services would now
not suffice, urgent alternatives were canvassed. The Governor-General consulted
with members of government and with the Chief Justice. The Chief Justice
proposed a mechanism for recalling parliament
or, alternatively, postponing the election. Cook, clearly untroubled by any
intimation that a delay might have offered some advantage, wanted the election
to proceed. And so it did.
On New Year’s Eve 1914, the Adelaide Advertiser
reflected on ‘The Dying Year’. Nineteen fourteen, it observed, had begun well:
‘The skies were blue and cloudless, and there were no portents to threaten that
the black shadows of war and drought would fall athwart the fortunes of the
State’. The coming of war, it said, had evaded even ‘prophets with the keenest
December, many new war measures were in place: press restrictions, bans on
trading with the enemy, special measures governing the sale of wheat and other
commodities, and financial measures, including an increase in taxation.
In the reporting of these developments, the fact that an event
of such constitutional novelty and drama—the double dissolution—had taken place
Addressing the first Federal Convention in 1891, Henry Parkes
spoke of the as-yet imagined Senate as an upper house that should ‘have within
itself the only conservatism possible in a democracy—the conservatism of
maturity of judgment, of distinction of service, of length of experience, and
weight of character’.
Almost no one foresaw the rapid rise of party interests in the Senate. One
Queensland delegate in 1891, John Macrossan, was the first to predict this
development, but his death two weeks into the Convention robbed him of a chance
to persuade the others or to enjoy his ultimate vindication. At the second
Convention, Alfred Deakin also predicted that the states’ house would be
quickly transformed into a party house: ‘The contest [between the houses],’ he
said, ‘will not be, never has been, and cannot be, between states and states ...
it is certain that once this constitution is framed, it will be followed by the
creation of two great national parties’.
The 1914 double dissolution, and the five that have followed,
have been governed by party distinctions; the anticipated tug of war between
national and state interests that shaped the complex debate at the Federal
Conventions over the framing of section 57 did not eventuate. We cannot,
however, conclude that the section would have been framed differently, had this
reality—which may, of course, alter over time—been realised. Two houses, both
directly elected, with almost co-equal powers, but with different electorates
and, since 1948, subject to different electoral systems, will always have the
potential to differ, and those differences will always have the potential to
turn into deadlocks.
Over the years, the unanswered questions in section 57 have
‘failure to pass’ may include delaying tactics as well as outright rejections;
multiple bills may be stockpiled and together serve as double dissolution
three months’ interval between the two rejections or failures to pass runs from
the date of the Senate’s action; the bill or bills in question do not need to
be on a matter of national importance, let alone confined to fiscal matters;
the government does not need to have been rendered unworkable by the deadlock;
a government can certainly ‘manufacture’ the conditions for a double
dissolution. Governors-General have an undefined quantum of discretion over
whether to grant a double dissolution (although possibly less than in the
past); they may consult whom they like in reaching their decision. But, no
Governor-General has ever refused a prime minister’s advice to dissolve both
Given the relative ease with which double dissolutions have
been granted and the relatively low threshold for a deadlock to qualify as a
section 57 ‘trigger’, and given that it is far from rare for bills to be
rejected in the Senate, it is notable that there have been so few double
dissolutions. Notable, but not surprising.
If there is one ‘lesson’ from the 1914 election, it is a
simple one. A government cannot count on getting what it wants from a double
dissolution. Like the delegates at the Federal Conventions of the 1890s who
laboured so valiantly over the wording of the section but who failed to achieve
consensus over its rationale, the Australian public has never held a common
position on its application. They have never risen up either to punish a
recalcitrant Senate or to chastise an overbearing House of Representatives.
They have never—as far as we can tell—decided how to vote in a double
dissolution election by weighing up the virtues of the particular bill or bills
that provided the trigger. They began in 1914 as they meant to go on: voting
according to their party loyalties, and to their perception of the respective
merits of the government and Opposition of the day. Governments can never be
confident that the Australian public shares their sense of the importance of
particular policy measures, let alone of the frustrations encountered in inter-cameral
politics. As Senator John Faulkner has written, ‘[v]oters are not swayed by
government complaints about Senate obstruction’.
Joseph Cook was necessarily a pioneer, and, in the event, a
casualty of constitutional novelty. Still, he acted deliberately and had no one
else to blame. At a dinner held in his honour a month after the election, he
cheerfully quipped that the ‘jolly good licking’ his party had suffered would
do them ‘no end of good’. There was ‘one consolation’ in changing sides of the
House, he said: ‘we have picked a job we all know about’.
This is not a thought that other prime ministers might find consoling.
Question — I was wondering if you see any parallels
between what happened in 1914 and the double dissolution in 1983. So what we
had there was that Malcolm Fraser wanted to recommend a double dissolution. Sir
Ninian Stephen as Governor-General wanted fuller and better particulars and
that delay then allowed the Labor Party to change leaders from Mr Hayden to Mr
Hawke. I was just wondering about that process of seeking the further details.
Did Mr Fraser wrongly assume that it would be automatically granted? And do you
have any other aspects of 1983 that you could share with us?
Helen Irving — Yes, there are parallels of course. The
gamble in both cases failed: the government lost and the prime minister was
defeated. There were and have been over the years different views as to whether
the Governor-General should satisfy him or herself in particular on the
question on workability of government. In 1951, when Prime Minister Menzies
advised Governor-General McKell to dissolve the houses, Menzies, who was
himself a very experienced former constitutional lawyer, advised that
workability of government was not an issue that Governors-General should take
into account. It seems that Governor-General McKell who, like all
Governors-General, wanted to take time to make up his own mind, probably shared
In 1983, Governor-General Sir Ninian Stephen held the opposite
view. He has made it clear that he held the view that a Governor-General should
satisfy him or herself not only that the technical preconditions have been
satisfied but also that it is a question of unworkability of government.
Malcolm Fraser had put the case that a double dissolution was needed in
particular in what he saw as a critically difficult economic situation with the
potential of unworkability. It is not necessarily clear that Malcolm Fraser
held the view or predicted that the Governor-General would simply accede to his
request, but certainly the delay, as you pointed out, allowed for the Labor
Party to change leaders in a surprise tactic.
In 1914 a war gets declared. It is not quite the same as a
change of political leader but was a big surprise in terms of the strategy of
the prime ministers. We also note that Malcolm Fraser is the one prime minister
to have been brought into government through a double dissolution election and
lost government through a double dissolution election. That is surely a first,
maybe a one and only!
Question — I was just wondering what you thought of
section 57. Do you think it is a good provision? Or do you feel that there are
ways we could improve it? And is there any parallel overseas for a model which
you think works better?
Helen Irving — That is a very big question. I do not
really know enough to answer the second part of your question on whether there
are good overseas alternatives. Each legislative system is very much shaped by
the particular national context. Look at the House of Lords in 1911. The
British constitutional crisis was ended then by the Parliament Act that very
much restricted the House of Lords’ chance to obstruct bills. Ultimately the
House of Lords lost the power to reject money bills altogether, which had been
the cause of the crisis. Focusing on the question of money bills would be
interesting in terms of evaluating whether there is anything further that could
be done with section 57, but being constitutionally realistic, anything that
involves a referendum is extremely unlikely to be successful.
There have been, over the years, many proposals—from the
constitutional convention of the 1970s, the Constitutional Commission which
reported in 1988, proposals from a discussion paper produced for Prime Minister
Howard—which involved amending section 57 to make it less likely that the
Senate would obstruct the House of Representatives or to provide an alternative
procedure such as a joint sitting, which was, as we saw initially proposed back
in 1891, to resolve deadlocks. But as I said, the reality is that anything that
required a referendum, that required constitutional alteration, would be extremely
unlikely to succeed. There have been a couple of referendums which have related
to the Senate’s relationship to the House of Representatives—not on section 57
itself—but those referendums have also failed.
There are, I am sure, procedures that could be dealt with
through standing orders or practices or conventions that might be adopted
between the two houses such as conferences of managers and so on which might
minimise the likelihood of a disagreement blowing up into a deadlock. That is
certainly something that could be contemplated, but as for the Constitution
itself, I think we are pretty well stuck on section 57.
Navigation: Previous Page | Contents | Next Page