Contemporary discussion of constitutional legitimacy tends to smuggle in the assumptions of one particular constitutional tradition and mistake them for universal democratic truths.
Ask almost any educated interlocutor to evaluate a system of governance and they will reach for democracy as the gold standard. Push further and they will reach for a particular kind of democracy: elected representatives, a written constitution, separation of powers, popular sovereignty as the font of all legitimate authority. Push further still and you will find, beneath the surface of apparent neutrality, something that looks remarkably like the American Republic of 1787.
Westminster operates by a different constitutional logic entirely: not separation of powers but separation of domains, each institution sovereign within its own sphere. This is often less analysis than trained reflex, and it distorts how we read constitutional arrangements that operate by different logics.
Westminster contains a theory of liberty and restraint not reducible to electoral democracy, and that the Crown is part of that theory, not an ornamental survival. Reformers who treat the Crown as a quaint anachronism awaiting replacement by an elected or appointed substitute are not arguing for more democracy. They are arguing for a different constitutional arrangement whose properties they have not examined and whose costs they have not calculated.
The tool this essay uses to make that case is a counterfactual. By imagining the Westminster settlement of 1689 running in reverse, with Parliament holding executive authority and the Crown holding legislative initiative, we can see what the actual settlement does, and why, with a clarity that straightforward description rarely achieves. Understanding what you have is the prerequisite for any honest conversation about whether to change it.
What Democracy Actually Means
Democracy means rule by the people. That is all it means. The specific mechanisms through which that rule is exercised, elections, representatives, parties, written constitutions, are historically contingent implementations of the principle, not the principle itself. Athens, on its own terms, did it differently. Switzerland does it differently. Representative institutions did not emerge by accident; they were partial answers to the practical problems of scale, complexity and mediation that direct participation cannot easily solve. But they remain answers to those problems, not the democratic principle itself.
The equation of democracy with representative electoral systems hardened into an assumption over the eighteenth and nineteenth centuries, largely through political theorists working in the shadow of the American and French revolutions. It is worth remembering what those revolutions actually produced. The American Republic of 1787 was not a democracy by any modern definition. The franchise was restricted to propertied men. Enslaved persons were counted as three-fifths of a person for the purposes of apportioning congressional seats, primarily to avoid the South outnumbering the North, whilst having no franchise themselves. James Madison, in Federalist No. 10, argued explicitly against direct democracy. The constitutional design, with its electoral college, its Senate representing states rather than population, its appointed judiciary with lifetime tenure, was deliberately insulating governance from the popular will it claimed to embody.
The Westminster settlement of 1689 operated on a narrow franchise, far from the universal suffrage we now regard as the mark of mature democracy. But its theory of representation was broader than its franchise: the member represented the constituency entire, not merely those who had voted. That is a different constitutional logic from one in which representation flows only from the act of voting, and it matters for how we read the settlement’s democratic character. Parliament in 1689 represented landowners, aristocrats, and merchant interests. Universal suffrage would not arrive until the 20th Century. What 1689 established was not democracy in the modern sense. It was a compact between two power centres negotiating a stable distribution of powers. Democracy was grafted onto that structure two and a half centuries later. But the structure itself predates democracy and was not designed with it in mind. Keeping this distinction clear is essential, because it frees us to ask a more useful question than whether Westminster is sufficiently democratic.
The question is: What are the constitutional properties of different possible arrangements, and what are the tradeoffs between them?
The Crown Is Not Ceremonial
Before arriving at the counterfactual, one further assumption requires examination: the claim that the Crown in modern Westminster is purely ceremonial, a dignified anachronism that could be replaced by an elected president, a parliamentary appointee, or anyone drawn at random from the population, without material constitutional consequence.
Consider what a lot-drawn Crown would actually be: a citizen chosen at random for a fixed term to fulfil the function. A randomly selected citizen shares some genuine properties with the hereditary monarch: they did not seek the role and cannot be removed from it by electoral calculation. They would carry no structural faction, no political debts, no incentive to court any constituency. Personal sympathies may exist but no mechanism delivered them to power on the basis of those sympathies, which is the relevant constitutional distinction. In those respects, the lot-drawn Crown approximates the unfactional character the hereditary institution tends to produce. But it falls short on the accumulated properties that the hereditary system generates over time: no institutional formation, no intergenerational stake in the institution’s survival, no dynastic interest extending beyond a single term, and no framework of obligation built across a lifetime. The coronation oath structures the motivation of someone shaped from birth for the role differently from someone selected by chance last month.
The objection to a lot-drawn Crown, typically stated as ’how would someone random be predictable, constrained by established convention, or, dare I say it, managed?’, is itself revealing. It exposes an assumption that the head of state’s primary function is to be controlled by someone else, which is precisely what the Westminster Crown is not. The Crown operates in its own sphere, constrained by convention and mutual dependency, not by subordination. A reformer who wants a controllable head of state wants something constitutionally different from what the Crown provides.
The tendency of a hereditary institution is toward continuity, unfactionalism, and intergenerational stake in constitutional preservation. The tendency of an elected or appointed head of state is toward the preferences of whoever controls the electoral or appointment mechanism. These are different constitutional instruments with different properties.
The reserve powers make this concrete. The Crown retains the authority to dissolve Parliament, to withhold Royal Assent, to appoint a Prime Minister in a hung Parliament. These powers are exercised with extreme rarity precisely because their existence disciplines the political actors who know they exist. Remove the Crown and you do not remove those constitutional functions; you redistribute them, to someone, under some mechanism, with different properties and different incentives. The Crown is less structurally incentivised toward faction than any elected actor; advisors can influence but cannot threaten tenure the way a party donor can.
Bagehot’s famous distinction between the dignified and efficient parts of the constitution has much to answer for. It trained generations to see the Crown as theatre rather than architecture.
How the Settlement Was Actually Reached
In January 1649, Parliament executed Charles I. It then proceeded to govern without a monarch. The experiment lasted 11 years and ended in a military dictatorship. Oliver Cromwell, commanding the army that Parliament itself had raised, dissolved Parliament with soldiers in 1653. The institution that had removed the King was itself removed by the force it had armed.
After Cromwell’s death and the failure of his son’s brief succession, Parliament did the only thing it could think of: it invited the King back. Charles II, who had spent years in exile, returned to the throne in 1660. His escape after the Battle of Worcester in 1651 has passed into English legend. Pursued by Parliamentary forces, he hid for a day in an oak tree at Boscobel whilst soldiers searched below. The Royal Oak became the symbol of the Crown’s survival, commemorated still in hundreds of pub names, a piece of living constitutional history that most of its celebrants no longer read.
What matters constitutionally is not whether Parliament was correct in its conclusion but that it concluded it needed the Crown. Whether a different arrangement might eventually have stabilised, whether the republican experiment failed for contingent reasons rather than structural ones, remains an open historical question. What is not seriously in doubt is that the political class which restored Charles II acted on the view that governing without a Crown had not worked, that unconstrained parliamentary power was vulnerable to capture, and that some monarchical check and denial of apex capture was necessary against recurrence. The 1689 settlement was built on that experience.
The events of 1689 must be read in this light. When Parliament negotiated the Bill of Rights with William III, it was not Parliament triumphant, imposing terms on a defeated monarchy. It was Parliament chastened, drawing on 30 years of catastrophic experimentation. William III knew the terms he was accepting. He accepted them because the alternative was not being King at all. But Parliament knew what it needed from him too. The compact was between two parties who each understood what the other provided. The Crown retains executive authority; Parliament holds law and taxation. Each has their own sphere. By convention, the Crown’s reserve powers over Parliament are rarely exercised. Parliament, in turn, legislates within the constitutional inheritance, not because it holds equivalent powers over the Crown but because that is the compact. The stability mechanism is asymmetric: the Crown can act on Parliament but chooses not to; Parliament cannot act on the Crown at all. Neither can function without the other’s cooperation.
The Parliament in Crown: A Counterfactual
Before constructing the counterfactual, it is worth pausing on a point of constitutional history that the standard account tends to suppress. Parliament did not originate as an emanation of popular sovereignty. It emerged from the Crown’s own governing machinery: the king’s council, the great assemblies of magnates and bishops, and the petitionary and taxation processes through which the Crown sought advice, consent and money from the realm. The Commons entered that structure not because England had discovered democracy, but because royal government increasingly needed broader assent, above all for taxation and grievance-redress. Parliament was not born as the people’s rival to the Crown. It began as a royal institution through which the Crown governed in consultation with the realm, and only gradually became the representative and oppositional body that later constitutional theory imagined it to have been from the start.
This matters because it grounds what follows. One of Parliament’s earliest functions was the receipt and processing of petitions submitted to the King. Constitutional traffic ran toward the Crown. Parliament was part of the machinery through which those petitions were heard, filtered and answered. The counterfactual proposed below is therefore not an invention. It is a rearrangement of elements already present in the tradition.
With that architecture understood, the counterfactual becomes thinkable. Imagine that at the moment of the 1689 negotiation, the distribution of powers had been inverted. Parliament takes executive authority, including command of the armed forces. The Crown takes legislative initiative: the power to draft law and present it for assent. The people retain the right to petition the Crown directly about what laws are needed. Parliament retains a power of assent (veto) over Crown legislation, exercised by convention rarely and never capriciously, because the Crown, having no faction and no electoral interest, does not put forward objectionable laws.
The structural symmetry with the existing settlement is exact at the level that matters for the thought experiment. In Westminster as it exists, the Crown assents to Parliament’s legislation by convention rarely withheld. In the counterfactual, Parliament assents to the Crown’s legislation by the same convention. In Westminster, the Crown’s withholding of assent is conventionally dormant but constitutionally real. In the counterfactual, Parliament’s power to withhold assent is conventionally dormant but constitutionally real. In both arrangements, mutual dependence disciplines both sides: neither can govern without the other’s cooperation, so neither has incentive to push the other to the point of refusal. The counterfactual is not a fantasy of royal absolutism. It is a structural mirror image.
In one respect, the counterfactual compact may be fairer than the actual one. The 1689 settlement was not between equals. Parliament had demonstrated it would execute a king and import a replacement. William III effectively accepted the terms drafted by Parliament. In the counterfactual, both parties bring something the other genuinely requires. Parliament brings executive capacity and military force, without which the Crown cannot govern. The Crown brings legislative function and the petition mechanism, without which Parliament risks again becoming what it became under Cromwell: a factional body using armed force to impose its will. The mutual dependency is structural rather than merely conventional, which may make it more stable.
The petition mechanism deserves particular attention. In the counterfactual, the people bring their concerns directly to the Crown, which synthesises them into legislative proposals. As noted above, this is not a novel constitutional invention. It restores a logic already present in the tradition: constitutional traffic running toward the Crown, with Parliament as part of the machinery through which the results are ratified and given force.
Whether this arrangement would produce better legislation than the Westminster we have is a genuinely open question. The existing system produces legislation shaped by party interest, donor pressure, and electoral calculation. Indeed, for some, it produces far too many laws and regulations. The counterfactual system would produce legislation shaped by what the Crown chose to synthesise from the petitions it received. A Crown with no faction, no re-election incentive, and intergenerational stake in institutional integrity cannot simply be assumed to be a less faithful synthesiser of diffuse public need than a party whip system operating on a five-year electoral horizon. The argument is about institutional tendencies, not a guarantee, and the essay does not need to resolve it. The point is that the question is open, and that dismissing it on the grounds that the Crown is hereditary is not an answer.
What the Exercise Reveals
The Parliament in Crown is not a proposal. Its value is analytical. By inverting the constitutional architecture and asking what the settlement would look like in mirror image, several things become visible that straightforward description obscures.
Firstly, the Westminster Crown is doing constitutional work that the ‘merely ceremonial’ framing conceals. In the existing settlement the Crown holds the power to withhold assent over Parliament’s legislation, exercised by convention almost never. The counterfactual simply swaps the holder: Parliament holds the power to withhold assent over the Crown’s legislation, exercised by the same convention with the same rarity. The instrument is identical; only its direction is reversed. If the Crown’s assent is genuinely meaningless, so is Parliament’s assent in the counterfactual. But we would probably not describe Parliamentary assent as meaningless. The asymmetry in how we characterise these structurally equivalent functions reveals a prior assumption: that elected bodies are constitutionally real and hereditary institutions are constitutional decoration. The counterfactual makes that assumption visible by removing it.
Secondly, democratic legitimacy does not flow exclusively through electoral mechanisms. The petition-to-Crown path in the counterfactual offers a form of popular participation that is direct, unfactional, and unconstrained by electoral cycles. It is different from representative democracy, not obviously inferior to it on every dimension. The existing system’s claim to democratic legitimacy rests heavily on the theory of representation, that elected members speak for those who elected them. That theory is severely strained in practice. The gap between popular preference and enacted legislation is, in most Westminster democracies, wide and persistent. The counterfactual forces the question: what does democratic participation actually require? The answer most people give is shaped by assumptions they have not examined.
Third, and most importantly: constitutional legitimacy is not exhausted by electoral derivation. Westminster contains a theory of restraint that depends precisely on the Crown not being reducible to electoral competition. The disciplining function of the reserve powers, the unfactional character of the role the Crown performs, the intergenerational continuity it provides, these are not embarrassing survivals. They are part of the constitution’s theory of how power is kept from collapsing into faction. You can argue that theory is wrong. You cannot argue that it does not exist, or that removing the Crown leaves the constitution otherwise unchanged.
Finally, the counterfactual clarifies that accountability does not disappear in the inverted settlement; it is sourced differently. The Crown’s legislation is accountable to Parliament through debate and the assent mechanism. Parliament is accountable to the electorate for whether it tolerates those laws. If Parliament assents to legislation the electorate finds objectionable, Parliament is voted out. The chain holds. What changes is the character of accountability at the Crown level, which is constitutional and motivational rather than electoral. The republican objection that the Crown answers to nobody misreads the architecture. It is constrained by Parliament, to the petition mechanism, and to an intergenerational stake in institutional legitimacy. That is a different kind of accountability, not an absence of it. The counterfactual is no less democratic than the settlement we have. It is simply a less imagined form of democracy.
The counterfactual is not a pre-democratic arrangement. It is a mirror of the existing settlement, which already operates under universal suffrage. Parliament in the inverted settlement is elected by the same franchise. The petition mechanism is open to all and the Parliament, which debates the laws drafted and passed by the Crown, is elected by all. The democracy is not removed; its institutional expression is rearranged. The concern that a Crown synthesising petitions in a modern mass society would inevitably become captured by advisors and bureaucracy is reasonable as far as it goes. But it applies with equal force to the existing system. Parliament synthesises competing popular interests through a party machine that is itself heavily subject to factional and donor capture. The objection proves too much.
A Parliament in Crown might prove better at preserving liberty and less structurally inclined to intrude into the lives of the people. One study found a 421 per cent growth in regulatory restrictions in Australia between 1977 and 2019, from roughly 23,000 to over 122,000 (IPA, 2019). Renovate a kitchen or bathroom today and the regulations governing your choices reflect not safety minimums, but collective social preferences imposed on private space. That is the texture of 122,000 restrictions.
The people most confident that Westminster’s non-electoral elements are redundant are frequently those least familiar with what those elements are doing. They are comparing Westminster to a template, the American Republic, that is itself a historical artifact of the late 18th Century, with its own structural pathologies: gridlock, counter-majoritarian institutions, the fusion of constitutional rigidity with factional capture. That template has not aged especially well. Westminster, which can evolve by convention without constitutional rupture, has in some respects aged better. The comparison that makes Westminster look antiquated turns out, on inspection, to be the less instructive one.
Westminster was forged through constitutional crisis rather than deliberate design, each settlement a response to failures that destroyed less adequate arrangements, teasing out through trial and considerable error the constitutional goods that restrain power and preserve liberty. Those who wish to remake the arrangements have an agenda. The possible agendas should be named. More responsiveness to transient majority opinion? More insulation from it? More capacity for social transformation? More resistance to faction and donor capture? More symbolic equality? Or even more liberty? These are not the same object. They do not point to the same design. The question worth asking of any reformer is simple: What are you actually trying to produce?
The Parliament in Crown thought experiment does not argue that the existing settlement is beyond question. It argues that before you alter an architecture, you should understand what it is doing. The 1689 settlement was reached by people who had tried the alternative and concluded that it had not worked. The Royal Oak is still there in the pub signs: not merely a curiosity, but a record of the lesson Parliament believed it had learnt when it tried to govern alone.
