3. Reconfiguring head-of-state offices in Australia
In this section, I advance eight propositions about the reform of gubernatorial offices in Australia. Taken together, these propositions constitute an ambitious, ultimately republican reform program. While there is a logical progression from one proposition to the next, the degree of political difficulty increases as we progress. So the extent to which the program can be implemented will vary from one jurisdiction to the next. Even if only the first proposition is accepted in a single jurisdiction, that would in itself represent a considerable initial reform. It would not, however, in itself guarantee a regime of fiducial governance. So, in the concluding section, I shall be taking as given the realisation of all eight propositions. In this way, I hope to demonstrate the ways in which the achievement of a fully republican regime could contribute to the remedying of the several regime defects identified in the preceding section.
a) Accountability of governors
Each governor should be more accountable than at present for what s/he does. Current arrangements continue to be excessively monarchical, in that governors are expected to regulate their own behaviour in accordance with the norms of what Peter Hennessy (1995) has dubbed the ‘good chaps’ culture characteristic of monarchical regimes. Under the current arrangements, as long as the governor can keep happy the head of government (who is after all usually responsible for his/her appointment) and (hopefully) the leader of the opposition as well, public accountability is deemed to have been satisfied. This bare-bones approach to the attainment of accountability is, however, dated and no longer applies to most public offices. Even in those areas where confidentiality must be maintained, experience in some of these other offices suggests ways in which gubernatorial accountability could be enhanced, without confidentiality being impaired.
Indeed, a useful distinction that can be drawn between the monarchist and the republican modes of governance is a cultural one. Cultural differences do affect the ways in which heads of state relate to heads of government, but the influence is an indirect one. The differences between republican and monarchical regimes are largely matters of style: the former are more open and place more clear and explicit institutional limitations on their heads of state than do the latter. These differences do of course affect the nature of the relationships between the head of state and the head of government, but they do so in often delicate ways. Even republican regimes sometimes depend on good chaps when it comes to some affairs of state, but they do so reluctantly, as a last resort.
b) Gubernatorial position statements
Each governor should have a public ‘job description’ negotiated with his/her head of government. Currently, the public has no way of knowing the full conception of the gubernatorial role held by an incumbent. Often, the incumbent him/herself must test the public acceptability of his/her conception through the floating of ‘trial balloons’, which inevitably attract adverse comments from some parliamentarians. Such a state of affairs is quite unfair for incumbents. A couple of incumbents (McGarvie, Sir Guy Green) have made behind-the-scenes attempts to assess the extent to which governors around Australia have respected the constitutional counselling function pioneered 40 years ago by Hasluck, but these informal inquiries have fallen well short of the development of an explicit code of conduct.
Our comprehension of what our heads of state should do is shaped—usually tacitly—by our understanding of what States should do. The State is responsible for making exercises of differing kinds of authority—public and private—comprehensible to its citizens. When the head of state can do this, s/he is performing an expressive role: making sense of the ways in which authority is being exercised. Because much of this authority is exercised in conditions of confidentiality, this expressive role often has to take the form of vouching. Although most citizens are not allowed ‘behind the scenes’, the head of state can be there and can vouch for the integrity of what s/he has seen.
In a republic, an elected president would be responsible for vouching to interested publics that the arrangements for furthering institutional integrity were being properly balanced against those furthering responsiveness; an appointive governor would be responsible for the coordination, through the leadership of a governance review committee, of those myriad activities aimed at securing institutional integrity.
Accordingly, an important role is that of vouching to the citizenry that overall the integrity of our institutions is being satisfactorily balanced against the requirements of democratic responsiveness by the activities of our governors and other statutory officers (auditors- general, ombudsmen, and so on). This is not a role suitable for governors, for if they were responsible for such a role they would be in part vouching publicly for their own performance and would in any case be too close to the executive branch to be perceived by the citizenry as someone independently representing their interests in institutional integrity. A directly elected officer—a president—would be needed for the performance of this ‘vouching’ role.
As far as I am aware, ten years into the new millennium, no other contemporary republican democrat has yet begun theorising the developmental potentialities of the office of head of state. Only the monarchists have in recent years been celebrating that office, but their approach could hardly be described as developmental. It is quite the opposite; for the more intelligent monarchists, the lower the executive profile of the head of state, the better (Bogdanor 1995; Hennessy 1995).
Indeed, there is a curious bifurcation in the contemporary literature on heads of state. One would have expected that discussions of the roles of heads of state would have attended closely to theorisations of the modern state. Anyone who expected this would be seriously disappointed, for the treatments of heads of state (of which Boyce  is an outstanding example) ignore the literature on the modern state, and those that deal with the nature of the modern state (of which Thompson  can serve as an equally outstanding example) ignore the roles of heads of state. Why should this be?
The short answer is that experiences in the twentieth century with heads of totalitarian states have so frightened us all (the outstanding statement of this condition of fright remains Cassirer [1963:ch. 18]) that we have considered it safest to adopt minimalist and formalist stances when confronting the inescapable need for heads of state. To engage in wider speculations about the development of new head-of-state roles has been considered altogether too dangerous. Yet it could turn out to be much more dangerous to attempt to sweep such considerations under the rug.
As Cassirer reminds us, the non-rational and symbolic dimensions of public life come to the fore at times of crisis. If we continue to shun the positive roles that our heads of state could perform in the fiducial governance of our modern states, we leave ourselves open to new totalitarian initiatives. It is much safer to consider openly the ways in which heads of state might be brought to play active but constrained roles in the securing of constitutionalist regimes around the globe.
c) Reviews of gubernatorial performance
Sir Paul Hasluck was Australia’s first modern governor, in that he set out to establish an ambitious gubernatorial counselling role:
I tried to satisfy myself first that the [Executive] Council had the power under the Constitution or a statute to make the decision recommended, that the recommendation was made by competent authority and that any preliminary enquiry or other steps required by law had taken place…
On matters which might be more controversial I would seek to satisfy myself that there was no conflict between the action recommended and any agreements, commitments or decisions of the government, and that respect had been paid to the conventions of the Constitution and the established procedures…
I was also concerned with ensuring that there was no conflict among my advisers…If the subject matter obviously was of interest to several Ministers and departments I required an assurance that there had been interdepartmental consultation and that the recommendation was supported by all those directly concerned…if I saw a possible conflict of policy, I would ask whether or not a recommendation had been considered by Cabinet and, if not, would suggest that the Prime Minister should be asked for his direction whether it should go before Cabinet. (Hasluck 1979:38–9)
Of course, this activist conception has proven controversial, with some governors (such as McGarvie) being strong supporters and others (such as Green) equally strong critics. Part of the reason for these differences of opinion might rest on variations in the performance of such bodies as cabinet offices. Even when such offices are performing well, it must, however, remain a gubernatorial responsibility to be satisfied that this has been happening.
So Hasluck’s former political adversary, Gough Whitlam (1998:5), subsequently paid tribute to the value of the counsel that his government received from him. And a later governor-general, the formidable legal theorist Zelman Cowan (1985:142), went so far as to agree with Hasluck that the work ‘demanded the highest intellectual and personal resources available to me’. It is hard to see what role other than that of constitutional counselling could have required the full intellectual resources of a mind such as Cowan’s. Similarly, the equally formidable jurist Sir Henry Winneke ‘found the constitutional side of the office interesting, satisfying and rewarding’ (Coleman 1988:330). And McGarvie himself (1999:65, 68 ff.) provided an account of counselling—‘a vital part of my role’—that has been very much in the Hasluckian tradition—one that seemed to be widely performed: ‘From discussions, particularly at the annual Governors’ Conferences, with those holding office in recent times, it is clear that the discreet but influential role personified by Sir Paul Hasluck is now widely followed in Australia’ (McGarvie 1999:26).
Despite McGarvie’s impressions, by no means all Australian governors have trod the Hasluck path. An early modifier was Governor-General Sir Ninian Stephen. Galligan (1991:69–70) reports that ‘[b]y Sir Ninian Stephen’s time this [Hasluckian] watchdog function had been largely replaced by procedures designed to ensure that matters coming to the Executive Council have been properly dealt with by the appropriate government Ministers and law offices’.
Nevertheless, even after much of the quality of the work had been enhanced by cabinet offices and the like, a gubernatorial responsibility should have remained—that of satisfying him/herself that the quality of the work done was at an acceptable level.
Another former governor, Sir Guy Green, has subsequently produced an elegant elaboration of the position of his fellow former judge Stephen. Green identifies three models of the gubernatorial role: (Hasluckian) interventionism; the ‘benign mentor’ (unsurprisingly, Green’s favoured model); and the ‘mechanical idiot’ (which Green rejects ‘fairly summarily’, although it should be noted that at least one recent governor appears to have conformed to this model). Green does, however, give qualified support to McGarvie’s impressions. Reporting on an informal survey of governors that he conducted in the late 1990s, Green (2006b) noted that ‘[g]overnors or the Secretariats of Executive Councils in Australia do in practice exercise an effective monitoring function by raising significant queries with the Ministers or departments presenting items to the Council. In some jurisdictions this occurs “very frequently”. It remains unclear, however, just how the majority of governors distributed themselves between the interventionist and the benign mentor models. Indeed, one well-placed observer has gone so far as to claim that the distinction is now ‘a distinction without a difference’ (Smith 2005:161). Whether or not this is an accurate criticism, an important issue remains, for it is yet to be determined just what should be the boundaries of gubernatorial concern with the activities of the executive branch.
If a Hasluckian constitutional counsellor role is deemed appropriate, a governance review council/committee should regularly assess performance in this role. On Hasluck’s own account of this role presented above, it nicely complements that of the ombudsman, with the governor protecting institutional integrity at the political level and the ombudsman doing so at the administrative level. A governor who played this role could thus be accorded membership of the governance review council/committee the creation of which in each of our nine jurisdictions was recommended by the NISA project.
d) The need for a council of state
If a governance review council/committee were to be responsible for work of such political sensitivity, it would need to be protected by a body such as the Irish Council of State. While the NISA report recognised the need for political champions to take up the reforms it was recommending, none has yet appeared, and such champions are highly likely to remain absent while our style of partisan politics remains unreformed. The introduction of a council of state (with a membership along Irish lines) could begin to discipline our parties in new ways, for such bodies would offer new pathways of advancement for the more consensually minded of our political leaders.
Councils of state are curious chameleon-like bodies. All states that differentiate heads of state from heads of government—and they are a considerable majority in the modern world—possess some such body to handle relations between these two centres of authority. None, however, has yet emerged to take a leading role in securing the integrity of all the major institutions of governance. Some, such as the British Privy Council, have fallen into desuetude; others, such as the French Conseil d’Etat, have become specialised in one area of governance (in the French case, constitutional and administrative law). And some, such as the Irish Council of State, have been designed to exercise very little authority. Nevertheless, the Irish Council of State possesses a membership that is well suited to the securing of institutional integrity at the highest levels, for it brings together the heads of the three recognised branches, together with a wide range of community representatives.
In the Australian case, it might be objected that we already possess our own variations on the privy council theme, in the executive councils that exist in each jurisdiction. These councils have, however, in modern times become so captured by our partisan systems that it is impossible to envisage them playing a broad fiducial governance role. Since the Hasluck innovations, however, they all possess the potentiality of supporting a governor with strong interests in the integrity of government programs.
In approaching these tasks of fiduciality assurance, councils of state will benefit from their composition as collegial bodies. As Baylis (1989) suggests, such bodies are particularly well suited to the furtherance of two of the ideals that Ackerman proposes for a regime of constrained parliamentarianism: democratic governance and integrity of public institutions. Indeed, the widening of the scope of our inquiry from the ‘-cracy’ on which Baylis focuses (his entire analysis rests on a fundamental distinction between collegiality and monocratic leadership) to one that encompasses ‘-archy’ (as in the above distinction between monarchical and polyarchical regimes) allows us to move beyond the findings of Baylis. Take, for example, his characterisation of monocratic leadership as being one mainly of symbolic reassurance (not unlike the role that used to be claimed for heads of state). When we consider the function of symbolic reassurance within the wider framework of a polyarchical regime, we can see that the function breaks down into two quite distinct parts: direct and indirect. When a head of government provides direct reassurance, s/he is referring to the performance of institutions under his/her direct control. When a head of state provides indirect assurance, s/he is vouching for the integrity of the institutions involved, many of which are not under his/her direct control.
e) Public ‘vouching’ for the integrity of council of state processes
The task of publicly vouching for the integrity of the work of these collegial bodies should not be the responsibility of any of the ‘working members’. If the proposed new bodies were to earn reputations for trustworthiness, they would need to have spokespeople who could regularly vouch for the integrity of their proceedings. And, if allegations of conflicts of interest were to be avoided, such spokespeople should be independent of those whose work was being vouched for. An independent chair of a council of state would be well suited to this task.
f) Directly elected ‘voucher’
If public trust in these new institutional arrangements were to be optimised, the ‘voucher’ would need to be directly elected. A publicly elected ‘voucher’ could appropriately be dubbed ‘president’. If the new institutional arrangements came to be widely seen as helping to increase public trust in our institutions of governance, the spokesperson would probably have to be elected. In the Australian federal system, it could well turn out to be the case that the national president could come to chair several (or even all) of the state/territory councils of state as well. If this were to happen, a more effective form of federalism could emerge. There would certainly be sufficient work to justify the continuation of a strengthened bicephalous configuration of head-of-state roles.
g) Two forms of politics
Although many of the more perceptive writers on democratic governance have differentiated two forms of politics—a ‘lower’ and a ‘higher’—the fundamental distinction made by Thompson (2001) best suits the purposes of this monograph. Her characterisation of the ‘masculine principle’ of politics closely resembles that of partisan contestation in Australia—one that ‘employs power, with reason, to meet immediate objectives’. The ‘feminine principle’, on the other hand, ‘is attached to traditions that maintain the strength of social affection across communities…It is linked with creativity, the unconscious, interiorization and mystery’ (Thompson 2001:103). This differentiation is valuable in the present context, for it asserts that both ‘principles’—in the terms used throughout this monograph: the partisan and the fiducial—are essential to good governance, for they function ‘in a distinctive equilibrium and [are] dependent upon maintaining a fine balance’ (Thompson 2001:102).
Such a ‘feminine’ president would not be, as is frequently asserted, apolitical or ‘above’ politics, but would be involved in a form of politics different from the partisan form that decides the composition of the government of the day. The boundaries between the two forms of government would have to be negotiated—and continually renegotiated—in the council of state. The mainstream view that the head of state should be ‘above’ politics stems from fears that a directly elected head might challenge the head of government on major policies (Power 2008a). While this is undoubtedly an important issue, it is one that now confronts every one of the ‘semi-presidential’ regimes—about one-quarter of all the national regimes in the contemporary world (Elgie 2004; Elgie and Moestrup 2007, 2008). Some of these regimes—such as the half-dozen that have developed ‘corrective’ head-of-state roles not unlike those introduced into Australia by Hasluck (Siaroff 2003)—should offer some relevant lessons for us.
h) A pathway to the republic
The development of the two separate roles of constitutional counsellor and of ‘voucher’ would continue the current bicephalous practices, but would do so in ways that offered a politically feasible pathway to a republic. As long as the republican movement remains divided between the selectionists (who value most highly the constitutional counselling work that has been performed by at least some of our governors) and the direct electionists (who favour a presidential conception of the office of head of state), the attainment of the republic will probably continue to elude us. It is only when each side is given what they most value that the passage of a referendum on the republic could begin to appear feasible.
 Hennessy (1995) has pointed out that this culture is normally secretive. To the extent that we know anything about this culture, it comes from ‘insiders’ discussing what they have been cleared (or, as good chaps, have cleared themselves) to divulge. According to Hennessy (1995:56–7), at the core of the British arrangements is a ‘golden triangle’ of good chaps, consisting of the Queen’s and the Prime Minister’s Private Secretaries and the Cabinet Secretary. This triangle is not formally accountable to anyone else for much of what they do together, although they no doubt consult closely with their principals. They provide the best example, however, of the ‘good chaps’ culture especially typical of the monarchist regime. On the assumption that only good chaps make it to these high positions and are loyal to the service of ‘the Crown’, nothing more in the way of accountability is deemed to be required. And in Australia, the leading good chaps have been the governors themselves, as they have been chosen by their heads of government on criteria that are never made public. The activities of the ‘good chaps’ are of course regulated by conventions: ‘the general agreements of public men about the “rules of the game” to be borne in mind in the conduct of public affairs’ (Hennessy 1995:36–7). There is, however, usually a large measure of uncertainty about the current bindingness of any specific convention. Hennessy reports having been persuaded that any putative convention that could be abandoned by an incoming prime minister is not worthy of classification as a convention, no matter how many previous prime ministers have respected it. Yet some prime ministers are themselves uncertain about some of the conventions that should govern their own behaviour; Hennessy (1995:23) reports Baldwin’s view that ‘there may be one practice called “constitutional” which is falling into desuetude and there may be another practice which is creeping into use but is not yet constitutional’. Hence the preference for the judgments of ‘good chaps’ over the inflexibilities of formally stated constitutional rules.
 Bagehot (1872:94) put the difference this way: ‘Royalty will be strong because it appeals to diffused feeling; and Republics weak because they appeal to the understanding.’
 I know of at least one recent governor who totally ignored the function of constitutional counselling, the importance of which is stressed here.
 Although both Boyce and Thompson are political scientists, in the works cited they are working within very different traditions: formal constitutional arrangements and the pragmatics of statecraft, respectively.
 Winterton (2004:43) supports the assessment of McGarvie that Hasluck was ‘the founding architect of modern governorship in Australia’.
 What is missing from this activist gubernatorial agenda is any explicit mention of the appropriateness of Executive actions under the Prerogatives power. Review of principles that should inform such actions would be an important concern of each Council of State.
 This is an excellent example of the ‘good chaps’ mystique still enveloping the gubernatorial office. Nothing as vulgar as an empirical study of the practices of governors appears to be possible; the public must be satisfied with the retrospective impressions of one former governor!
 As each governor still remains free to define the role as s/he pleases, it would be unfair to identify the one who on my information conforms to the ‘mechanical idiot’ model.
 Over many years, the Privy Council served as a model for many constitutional designers. For example, it provided a model for the Australian executive councils that are discussed below. More spectacularly, the American founding fathers vested the famous ‘advise and consent’ powers in the Senate in the hope that this might encourage it to emulate the House of Lords as the seat of some of the powers of the Privy Council (Sundquist 1992:37, 61, 313).
 The membership of the Irish Council of State is as follows: Prime Minister; Deputy Prime Minister; Chief Justice; President of the High Court; Presiding Officers of the two Houses of Parliament; Attorney-General; any former president, prime minister or chief justice willing to serve; up to seven presidential nominees. For reasons stated elsewhere in this monograph, it would be highly desirable to grant leaders of the opposition seats on each council of state. Republican reformers have been slow to perceive the great potential of the Irish arrangements—which were after all fashioned in the 1930s by a great statesman confronting a situation not unlike our own—because they have assumed that the current bicephalous arrangement that we possess must disappear, so that we would have a single officer, a president, as head of state. As the Irish President does not have the authority to perform the functions of Australian governors, her office and the Council of State that supports it have been incorrectly deemed irrelevant to Australia (see Power 2005, 2006).
 Baylis’s (1989:9–10) important work is explicitly concerned only with executive branches. Nevertheless, the work throws much light on the strengths and weaknesses of collegial ways of proceeding, and this light can assist us to deepen our understanding of how a collegial body that headed up an integrity branch should proceed. The most important lesson is one that carries a strong caution. Collegial bodies are particularly well suited to the tackling of technically complex problems that require heavy support from specialised bureaucracies, especially those embedded in influential policy communities. This is all very well, but the synergies that link collegial deliberative bodies and their bureaucrat officials strengthen a tendency that both usually exhibit strongly: a tendency to secrecy and insensitivity to outside criticism.
 From a quite different feminist direction, Eisler (1987: 105) has introduced the term ‘gylany’, which bears strong resemblances to Thompson’s conception of balance.
 On one perceptive reading of Ackerman (Choudhry and Mount 2006), a similar distinction—between ‘normal’ and ‘constitutive’ politics—is made by him, and is, as is argued in this monograph, central to his rationale for serial referenda.
 There is one important lesson for comparative governance studies that should be noted here. Writing two decades ago, just before the explosion in the number of semi-presidential regimes, even so capable a scholar as Baylis was dismissive of the importance of such regimes. Admittedly, he devoted considerable attention to the French Fifth Republic, which he was content to classify as presidential. The only other regime of this nature to receive even the slightest attention was Finland, which Baylis (1989:128) passed over as an uninteresting ‘hybrid’.
 On 17 April 2008, I published an article in the Melbourne Age, in which I identified these regimes: Ireland, Finland, Lithuania, Poland, Bulgaria and Macedonia. The article attracted no interest. The surprising lack of Australian interest in semi-presidential regimes continues. In the just published volume The Australian Study of Politics (Rhodes 2009), I am the only one of 37 contributors to mention such regimes. This paucity of citing of course reflects the lack of interest in the wider political science community, so that the eminent comparatist LeslIe Holmes (2009) does not mention these regimes in his characteristically thorough survey of Australian contributions to his specialty.