TT Arvind and Lindsay Stirton
Legal scholars have long reached for architectural metaphors when thinking about the constitution. Blackstone described English law as a ‘noble and ancient castle,’ built up over generations, its parts Gothic and irregular but still coherent. Dicey spoke of the ‘foundations’ of constitutional principle. And we today continue to talk of ‘pillars’ of the rule of law or the ‘scaffolding’ of democracy.
Architecture also provides a suggestive metaphor for thinking about reforms to fundamental constitutional structures, in the form of Robert Venturi, Denise Scott Brown and Steven Izenour’s distinction between the ‘duck’ and the ‘decorated shed.’ A duck is a building whose form is its message—where function and symbolism are fused into a single, expressive structure. Venturi et al. named the concept after the Big Duck Poultry Barn in Long Island, New York, which was literally built in the shape of a duck. A decorated shed, by contrast, is—like the buildings of Las Vegas—little more than a plain box with a sign on it. Its structure is practical and flexible. Symbolic meaning is communicated via elements that can be changed or retained independently of the structure. You can change the sign, repaint the walls, even rewire the interior without touching the structure.
This year marks 20 years since the Constitution Reform Act 2005 controversially stripped the Lord Chancellor of his judicial and legislative roles and separated the highest court from the House of Lords. To an external observer, it might seem peculiar that disentangling the executive from the judiciary and the judiciary from the legislature was at all controversial. But we can better understand what made the issue so heavily contested if we view it as reproducing in legal form the distinction between ducks and decorated sheds. Proponents of the reforms saw the constitution as being at its best when it acted like a duck. Opponents saw the constitution’s strengths residing in features that brought it closer to being a decorated shed. The tension between the two visions is worth reflecting on as we mark the 20th anniversary of the 2005 reforms.
*****
In duck-mode, constitutional structures are designed from first principles to fit perfectly their function. An office created for one purpose cannot easily be repurposed to meet a different need. Principles and their logic, rather than the accumulated experience embedded in existing structures, are the foundations of constitutional change. Decorated-shed-mode constitutional theorizing, on the other hand, sees the constitution as performing a series of more-or-less generic tasks: allocating power, protecting rights, mediating between different visions of the constitution, and providing ‘how-to’ instructions on everything from appointing a prime minister to financing local public goods. Any structure can be repurposed to perform these tasks, as needs be. Effectiveness matters more than elegant design, and messy structures that do their job are constitutionally unobjectionable despite their messiness. Radical ‘duckish’ reform, in contrast, carries the hidden danger that it might accidentally sweep away structures whose importance the reformer failed to see, or lock the constitution into patterns that serve the polity poorly.
Venturi et al. favoured decorated sheds. Las Vegas-style architecture, they observed, could be easily adapted to changing circumstances—internals could be reconfigured while preserving the external symbolism and vice versa. The arguments they made bore more than superficial resemblance to the claims of those who praise the famed flexibility of the United Kingdom’s constitution, and to Blackstone’s image of English Law as a gothic castle, whose rooms had nonetheless adapted over time to meet arising challenges.
Take the office of Lord Chancellor, effectively ‘abolished’ by the 2005 reforms. This was an office that had evolved beyond recognition since its creation. The keeper of the king’s seal and head of the royal chapel became the head of the royal secretariat (the ‘chancery’), which only later exercised a recognizably judicial function. We can recognize the symbolic continuity of the office of Lord High Chancellor held by Sir Thomas More through to Lord Irvine in the 2000s. But the function had changed almost beyond recognition. This is exactly the flexibility we can expect of decorated sheds: an institution that is effective and enduringly built will almost inevitably be adapted to serve new ends.
It is, therefore, not surprising to see that opponents of the 2005 reforms adopted decorated-shed reasoning. During a debate in the House of Lords in 2004, Lord Hoffmann lamented the fact that the reforms broke a mechanism that had served the country so well in the past. The genius of the UK’s constitution, he argued, lay in the fact that institutions such as the Lord Chancellor had adapted themselves over centuries to new circumstances, through the steady growth of accepted constitutional conventions. The reforms unwisely abandoned this tried and tested approach in favour of what he termed a ‘fundamentalist interpretation’ of the separation of powers.
Proponents of reform, in contrast, took an unabashedly duckish approach. In the foreword to the 2003 consultation paper introducing the reforms, Constitutional reform: reforming the office of the Lord Chancellor (CP 13/03), Lord Falconer said that the Lord Chancellor’s office had ‘obscured, even confused’ what should logically be distinct functions: that of a minister, of a judge, and of the Speaker of the House of Lords. If the UK was to continue to persuade developing countries to ‘adopt clearer constitutional mechanisms’ and insulate the judiciary from political pressures, it would have to follow its own advice and abolish the Lord Chancellor. Similar considerations of principled design were also highlighted by other supporters of the reforms, who pointed to the Council of Europe’s view (Resolution 1342/2003) that the Lord Chancellor’s multiple roles were incompatible with the Council’s legal and constitutional principles; or to the decision of the ECtHR in McGonnell v UK [2000] 30 EHRR 289 holding that the dual role of Guernsey’s Bailiff could vitiate the impartiality of its Royal Court (although the court stressed both in the majority judgment and in a separate concurring note by Sir John Laws, that its finding was a response to the specific facts of the case and not an general indictment of the constitutional structure of the office of Bailiff).
*****
The clash between these two dispositions to constitutional reform long preceded the 2005 Act. In 1836, a proposal to lighten the Lord Chancellor’s judicial workload by assigning some of it to other judges foundered in the face of two diametrically opposed arguments: the duckish argument, most powerfully expressed by Lord Langdale, that nothing short of completely separating his political and judicial functions would suffice; and the decorated shed counter-argument, advanced in particular by Lord Lyndhurst, that any change to the Lord Chancellor’s functions would degrade an office of utmost constitutional importance. As discussed below, similar arguments resurfaced in 1918 when the Report of the Machinery of Government Committee (the Haldane Report) suggested changes to the Lord Chancellor’s office.
By way of provocation, we suggest that this clash of dispositions reflects a deeper truth: duckish structures retain an enduring appeal in constitutional reform. Sometimes this reflects the genuine need to keep form and function entwined, but as Jenny Steele warns, conceptual allure carries with it the risk of ‘neglect[ing] the underlying interests and outcomes that are inevitably involved.’ The familiar imagery of ‘separation of powers’ illustrates how such concepts can foster the conviction that institutions must embody them, regardless of their actual effects. In constitutional settings where many European polities boast carefully designed institutions guided by overarching principles, a patchwork of decorated sheds can appear anomalous—an eccentricity that invites rationalisation.
More often, however, the persistent appeal of the duck reflects the difficulty a reforming government faces when it encounters an area filled with decorated sheds whose successive adaptations bring them to the point where form and function resist disentangling. Whilst all three played a role in the 2005 reforms, the last was arguably the most significant. The government of Tony Blair was on its face a government that liked modernization to be modular, rational, and not especially theatrical. It introduced devolution, the Human Rights Act, reform of the House of Lords, the Freedom of Information Act in a rhetoric of managerial adjustment more than grand reimagining. But what reforming governments often encounter, we want to suggest, are decorated sheds with which they do not know how to deal. The British constitution is full of them—institutions where form and function have accumulated in layers that are difficult to disentangle, with key functions often obscured by the baffling forms behind which they live. Institutions of this type can certainly evolve organically, but altering them through deliberate redesign is harder. The office of Lord Chancellor exemplified this point: simultaneously head of the judiciary, a member of Cabinet, and a presiding officer in the legislature. That is not a role that can easily be remodelled behind the façade. A government which starts tinkering soon finds that it is not just redecorating. It is remaking the building.
On one level, the Constitutional Reform Bill was presented in the language of rational design: separating powers, clarifying responsibilities, modernising. But in practice, it dismantled one of the constitution’s oldest symbolic structures and replaced it with something quite new. A Supreme Court, with a new building, a new name, and a new crest. A Lord Chancellor stripped of robes, judicial office, and ceremonial weight in a short time became little more than an alternative name for the Secretary of State for Justice when exercising functions in relation to the judiciary. A Lord Speaker now presided over proceedings of the House of Lords. If anything, this was the construction of a new set of ducks.
So perhaps the real story of the 2005 reforms is this: New Labour set out to work with decorated sheds. But the constitutional structures they touched had more complexity in them than they wished for or liked to admit. In trying to rationalise the system, they found themselves drawn into duck-style rebuilding and not just internal reconfiguration. This was also true of the failed 1836 reform. A government that set out to reconfigure an institution to remove acknowledged impediments to its efficiency had to abandon its proposals because it was unable to resist the conclusion that achieving its goals in relation to an institution as complex as the Lord Chancellor would require the institution to be entirely rebuilt on first principles.
*****
Historically, the best constitutional reforms reached workable compromises between duck and decorated-shed perspectives by striking a balance between functions, symbols, and structures. The reforms to the judicial House of Lords in the 1870s are an example. The Judicature Act 1873 sought to abolish the final appeal to the House of Lords, and to constitute the Court of Appeal as the apex court of England and Wales. George Edmund Street’s magnificent cathedral-like design gave concrete expression to Bagehot’s sentiment that, ‘[t]he supreme court of the English people ought to be a great conspicuous tribunal… not hidden beneath the robes of a legislative assembly.’ But while in purely architectural terms, the Royal Courts of Justice was very much a duck, the approach to constitutional reform was much more decorated shed. Despite the provision to abolish the House of Lords’ judicial role in s20 of the 1873 Act, there was no great symbolic attachment or unshakeable commitment to doing so and these aspects of the reforms were quickly reversed by Disraeli’s second government. The removal of the final appeal to the House of Lords was first postponed (by the Judicature Act 1875) and then reversed (by the Appellate Jurisdiction Act 1876). Appeals would be heard by Lords of Appeal in Ordinary—the ‘Law Lords’—who were to be legally qualified and to have outstanding experience. This epitomizes the decorated-shed dimension to the reform of the system of final appeal, which modernized function while preserving form. Under the direction of Lord Cairns, then Lord Chancellor, the move sought to professionalise the judicial role without disturbing the symbolic shell, which remained unaltered until the Constitutional Reform Act 2005.
Despite its lack of success, this was also true of the Haldane Report’s proposals. The Report made two substantial criticisms of the Lord Chancellor’s role as it then was. First, it said that there was a need for a single minister of justice who would carry responsibility for the civil and criminal justice systems, rather than bifurcating them between two departments. Second, echoing the debate in 1836, it said that the combination of judicial, administrative and parliamentary functions meant that the Lord Chancellor was over-burdened ‘without really being allowed either the time or the machinery requisite for their performance.’ Nevertheless, it recognised both the symbolic significance of the office of Lord Chancellor and the constitutional importance of the functions that the office had accumulated and that Lord Chancellors could perform more effectively if ‘freed from the duty of daily or even of frequent judicial sitting.’ The practical and symbolic authority of the office of Lord Chancellor gave its occupant a real and useful weight within Cabinet, enabling him to occupy a prominent position as a ‘legal and constitutional adviser of the highest standing.’ Its importance extended beyond cabinet. Haldane envisaged that the Lord Chancellor would continue to sit in important constitutional cases, such as Appeals from the Dominions, cases which required ‘familiarity with constitutional usage which can never be acquired so well as in the school of long experience in Parliament.’
This was not, however, true of the 2005 reforms. Unlike the Haldane Report, the government in 2005 does not appear to have considered whether any positive contribution was made to government or the legal system by the breadth of the Lord Chancellor’s traditional functions, or the presence of judges in the legislative House of Lords. The consultation paper notoriously only solicited views on four specifically selected functions of the Lord Chancellor: ecclesiastical patronage, visitorship, royal peculiars, and charities and schools. Responses were not sought on any other issues. Given that the reforms’ goal was to give concrete form to values and ideas already settled on by government, this may well have been inevitable. But it nevertheless leaves open the question of whether anything of constitutional value was lost in the process of doing so.
*****
It is arguably in the area of less obvious functions occupying the interstices that the impact of the reforms has been most significant. In December 2014, the House of Lords Select Committee on the Constitution published a report on how the office of the Lord Chancellor had evolved since 2005. It found that the ability of Lord Chancellors to ‘act as a guardian of the rule of law’ had diminished, and the duty had become more difficult for them to discharge. This was in part because their office had become more political, in part because they had acquired substantial new responsibilities for policy in contentious areas, and in part because they no longer sat on the Parliamentary Business and Legislation Committee. However, it also pointed out that the Lord Chancellor’s new status as simply one more Secretary of State amongst others meant that influence no longer inhered in the office, but depended on the personality of the person holding the office.
It is difficult to see how this could be otherwise, once the Lord Chancellor’s functions in Cabinet were detached from his functions as a judge. Leaving aside the question of whether it is appropriate for a minister to sit as a judge, it is nevertheless self-evidently true that being seen to have the qualifications to sit as a judge on the highest court gave pre-reform Lord Chancellors added authority in Cabinet. So, too, did their other functions, such as being the speaker of the House of Lords. Post-reform Lord Chancellors, in contrast, have little to set them apart from other Cabinet members. As the Select Committee pointed out in a subsequent report in 2023, post-reform Lord Chancellors tend to be mid-career politicians who typically go on to hold other roles, and who are likely to have political ambitions that may make them less inclined to stand up to the Cabinet or the Prime Minister. The Select Committee in 2014 had urged Prime Ministers to appoint persons with ‘sufficient authority and seniority’ and to above all consider the importance of the Lord Chancellor’s duty ‘to uphold the rule of law across Government’. The profile of Lord Chancellors in the decade since does not suggest that Prime Ministers have paid much heed to this exhortation.
Governments have not, of course, been blind to or complacent about these trends. But their response has been to create formal institutional norms to replace the constraints of culture, convention, and practice that came with the Lord Chancellor being a judicial heavyweight. As the work of Christopher Hood et al. on regulating Whitehall has shown, codifying norms and expectations is a typical response when an institution is opened up to admit persons who would previously be considered ‘outsiders’. Expanding regulation, as exemplified by the principles enacted by Section 2 of the Constitutional Reform Act, however, never wholly substitutes for the dismantling of an institutional culture. An awareness of this is likely to be at least a partial factor behind the growing concern that one sees in legal commentary about the extent to which modern governments embed a real commitment to the rule of law.
*****
Our purpose in making these points has not been to decry the 2005 reforms, or to suggest that their impact has been baneful. They are now firmly entrenched as part of the constitutional landscape and are unlikely to be substantially changed in the foreseeable future. It does, however, suggest the need to temper principled design with a pragmatic approach to practical and symbolic function when planning further constitutional change. To return to the analogy with which we began, constitutional reform will need to deal with the fact that the UK’s constitution contains institutions and structures that are best understood as decorated sheds, to the reform of which principled design by itself will always be inadequate.
TT Arvind is a Professor of Law at the University of York.
Lindsay Stirton is Professor of Public Law at the University of Sussex.
