If Borges had been into static typing


type sea_maiden = Mermaid | Siren

type drawing_implement = Fine_camel_hair_brush

type animal =
| Imperial_property
| Embalmed
| Trained
| Suckling_pig
| Sea_maiden of sea_maiden
| Fabulous
| Stray_dog
| Classified of animal
| Trembling_as_if_mad
| Innumerable
| Drawn_with of drawing_implement
| Et_cetera
| Just_broken_flower_vase
| Resemblent_of_flies_at_distance

Blocking Supply


The Canberra Times and the Sydney Morning Herald have just run a somewhat misguided article about blocking Supply, written by Andrew Herscovitch. He thinks the ability of the Australian Senate to block Supply is a problem, and suggests other problems he'd like to exchange for the status quo. Originally, this blogpost was devoted to trashing his article, but while writing it I have concluded that while Herscovitch is wrong, the issues are not nearly as clearcut as I thought they were.

Australia's political systems at Federal and State level all predate the UK's Parliament Act 1911; thus the Upper House has the power to block legislation, including laws authorising taxation. As in America, therefore, either chamber of the legislature can bring the operation of government to a standstill, by withholding money; this historically is the origin of Parliament's legislative power: the King conceded legislation in return for consent to collect taxes.

An aspect of the American system not adopted by Australia is that the Government does not have security of tenure: it is possible for changes in Parliamentary strength to lead to a change of government between elections, which happened in 1940, and could happen in the present hung Parliament as well. If you are fundamentally incapable of accepting that the government lacks American-style security of tenure, then you'll have a lot of trouble reconciling yourself with other, logically related aspects of the Australian constitution.

There isn't really an electoral mandate in a Parliamentary democracy, the way there can be in a Presidential system. Some people may find this difficult to cope with too. In 1941, was it the Government or the Opposition who could claim a mandate from the electorate, the way FDR could? Neither.

The Governor General has no practical choice but to go along with whoever can guarantee Supply to the Crown, which is generally the Government. If the Government loses control of Parliament, then the Governor General has much greater freedom of action, and the attendant ability to make politically extremely controversial decisions.

The effect of the constitution is that an early election may be called by the Government when it controls Parliament, but that when the Government does not control Parliament, an early election may be triggered if the Senate Opposition, the Governor General and the electorate all favour one.

This state of affairs is neither widely acknowledged, understood or accepted, as Herscovitch's piece evidences.

What is widely accepted is that losing control of the Lower House means losing Government, and that this is so regardless of the state of public opinion. However, since Oppositions will not force an election they are going to lose, public opinion de facto plays the same rôle when either chamber blocks supply: you don't do if it you're not overwhelmingly popular. Thus the two houses are equivalent when blocking supply: either can impel the Governor General to dismiss an unpopular Government. This wouldn't surprise an American unfamiliar with the details of the Australian constitution, but it surprises Australians.

Both sides of Australian politics have used the device of blocking supply in the Upper House, though at Federal level this has only been done by the conservatives to the Whitlam government (in 1975). In Victoria, it was done by Labor to compel the removal of the bias in the electoral system (in 1952).

Herscovitch's article begins by stating that

It is not beyond the realms of possibility that the Labor opposition could combine with the Greens to block supply if the Abbott government began to falter seriously.

to which we can respond, "so what?". That's just the way the system works. He doesn't actually tell us why it is supposed to be bad that the Government must win both houses of Parliament to remain in office. Instead, he basically takes it as given that this is bad:

Should the Senate keep its right to block supply? There is a strong case for saying ''no'', but a constitutional amendment to this effect is unlikely to attract bipartisan support and therefore to succeed at a referendum.

I'll speculate a bit later about what might be the underlying reasons for his position.

His points are as follows:

  • firstly, that the electoral arrangements for the Upper and Lower Houses differ such that the Upper House doesn't face so strong a sanction as the Lower House for blocking Supply (fair enough!),
  • and the procedure whereby the Governor General may resolve disputes between Parliament and Government is
    • undefined,
    • and/or divisive
    • and/or not accepted by the public

. He acknowleges that the power to block Supply commands sufficient support as to be unchangeable, but nevertheless presents his proposal for change.

This proposal he characterises as "preserv[ing] the Senate's powers over supply", but they wouldn't, so we can ignore the details; they'd be as unacceptable to the public as an honest and overt attempt to strip the Senate of the power to block Supply, and thus would not command sufficient support for adoption. He never convincingly states why there ought to be an automatic election if the Upper House blocks Supply -- after all, there wasn't one in the early 1940s when control of the Lower House changed between elections.

It might all be very entertaining at first, but the issue would quickly become serious as the money began to dwindle and the Governor-General was forced to consider her options with no clearly defined process for ending the impasse.

The notion that the Governor General has to do anything about blocked Supply is the position which was taken by conservative lawyers (e.g., Bob Ellicott and Garfiend Barwick) during the 1975 Dismissal Crisis. This is to get the relationship back-to-front. The Governor General can appoint whomever she wants as Prime Minister, but in practice has no choice other than the leader of the Parliamentary majority, who will otherwise block Supply. Indeed, a dismissed Government might even use its Senate numbers to block Supply to its successor, though the ALP lacked the numbers to do this in 1975.

If the Governor General not only can but must dismiss Governments which cannot guarantee Supply to the Crown, this would subordinate the operation of government to the vagaries of the bond market! Is it logically sensible to say that a Government which has no choice but to default on its debt must be dismissed, but that a Government, if it has the Parliamentary numbers, need not be dismissed should it choose to default on its debt, as the Irish or Greek governments might soon do?

Herscovitch continues:

A new set of constitutional provisions is needed. They must make the Senate directly accountable to voters for the exercise of its power over supply, provide a clear path to a democratic resolution of the deadlock, and reduce the scope and need for discretionary - and potentially divisive - vice-regal (or presidential) intervention.

My instinct is that there's no need for any resolution procedure: if the Supply bills can't get through, then the Government needs to back down and give concessions to the Opposition, just as the Kings of England had to of old. How might this situation be abused? An obstructionist party which did not or could not win Lower House seats might win a majority in the Upper House (it is not sufficient to win the balance of power in the Upper House, as this entails allying with the Official Opposition to block Supply, which the Opposition won't do if they'll suffer the same fate after they've won the ensuing election). I don't think it's remotely plausible that a third party could win a majority of the Senate seats and almost none in the House of Representatives.

Herscovitch supposes that the

ensuing Senate election would focus primarily on the upper house's departure from the norm rather than on more general questions of the government's performance and the opposition's fitness for office

but does not propose the severe restrictions on press freedom necessary to achieve this. The 1975 election was not "about" the Dismissal, and any Senate election occasioned by blocking Supply is similarly not going to be restricted to the constitutional issue. After all, if the Upper House is blocking Supply, it must be doing so because blocking legislation alone is ineffective, e.g., to stop a War or curtail the abuse of some executive power.

Voters would be deciding whether the circumstances were so ''reprehensible'' as to justify the Senate's major departure from normal practice.

They didn't last time, and there's no reason to think that the circumstances confected under his scheme would differ significantly.

An important function of any constitution is to provide, as far as possible, a clear and accepted framework for resolving the most difficult disputes.

No it isn't.

A constitution does not need to specify everything. Cricket doesn't need to add a rule saying that a batsman must run as quickly as possible towards the crease, because that's already the logical behaviour under the existing rules. How could such a rule operate? Must the batsman suffer a Dismissal if he doesn't run fast enough?

I expect that Herscovitch and I would come at the whole question and any proposed remedies from the basis of different assumptions, but I'd certainly not go and presume that my assumptions were widely held. Of course, the whole point of publishing his article may be to encourage readers to think that its assumptions are too solid to be questioned. I think his assumptions are roughly as follows:

  • that there shouldn't be checks on the will of the people, as expressed by electing the Government
  • that Australia's Federal Upper House is democratically illegitimate, because the Upper Houses of the United Kingdom and Australian states are or have been
  • that the Governor General should have no discretion, certainly not to do what he did in 1975
  • that the Governor General legally did not even have the power to do what he did in 1975
  • that it is a problem if the constitution leaves anything to convention
  • that we should all drive hybrid cars

Actually, I threw the last one in for a laugh, but even if he isn't really saying we should all drive hybrid cars, I'd bet good money he wouldn't like to hear it denied.

I'm not going to answer Herscovitch's point that Australia could strip the Senate of the power to block Supply without stripping it of the power to block legislation. The scheme he proposes would certainly achieve that, but he disingenuously describes it as not removing that power. This misrepresentation weakens his argument, because he can't offer anything to justify what he's pretending not to do. Maybe the Senate should be allowed to block legislation but not Supply, but Herscovitch can't tell us why, because he's pretending his proposal does something else.

Australia's constitution has a mixture of British and American influences, and a mixture of democratic and republican elements; the tension is between American-style checks on the executive by a democratically-elected Senate, and the informal indirect electoral mandate enjoyed by the majority in the Lower House.

If Australia's constitution were more republican, the Governor General might have security of tenure, but this is not logically compatible with the position being appointed by the Queen; if this had been the case in 1975, Sir John Kerr, who wanted the job for ten years, would have been able to warn Whitlam about his potential Dismissal without worrying that the latter might recommend his replacement to the Queen, and the outcome of that political crisis would have been very different.

The situation can only be changed by a constitutional amendment which would never pass referendum until it was no longer seen as one side of politics exacting revenge for 1975.

Herscovitch repeatedly mentions that existing mechanisms for resolving these conflicts are divisive. He doesn't mention that this divisiveness is politically asymmetrical: the Left is temperamentally incapable of getting over losing in 1975 (or, indeed, in 1659), and will bear this new grudge for centuries. Not a lot that can be done about human nature.

The real issue is that limited government is a widely held and accepted view in America, almost part of the civic religion, and it is reflected in the American Constitution. Some of its provisions have been copied to the Australian Constitution, where they command less support, as the country's civic religion is closer to that of Britain, whose constitution lacks such explicit potential for deadlock; in Britain, politicians are required to exercise much more self-restraint.

What really got me going about Herscovitch's article was the idea that the supposedly undefined processes whereby the Governor General might resolve problems actually needed to be codified. There's not nearly enough work on why this is a bad idea, so I'll have to find some or do it myself.