By Mark Elliott
Professor of Public Law, University of Cambridge

When the United Kingdom Parliament wishes to make law by enacting a piece of legislation, three things must normally happen. First, the ‘bill’ (as legislation is known until it is enacted) must be approved by a majority of MPs in the House of Commons. Second, unless a special procedure under the Parliament Acts 1911 and 1949 is followed, it must be approved by a majority of members of the House of Lords. And, third, it must receive ‘Royal Assent’. As the UK Parliament’s website explains, ‘This is when the Queen formally agrees to make the bill into an Act of Parliament (law).’ But this implies, somewhat surprisingly, that if the Queen does not formally agree to make a bill into an Act of Parliament, it will not become a law — suggesting that the Queen is legally capable of  refusing to approve it.

Law and convention

All of that is true. But it is also highly misleading, because it presents only one part — the legal part — of the constitutional picture. What it leaves out is an equally — perhaps even more — important part of the picture, concerning what are called ‘constitutional conventions’. In one sense, constitutional conventions can be thought of as traditions, but they are much more than that: they tell us what, in fact, usually happens, what should happen, and what is likely to happen.

Take, for instance, the granting of Royal Assent. Although, as a matter of law, the monarch can veto (reject) a bill, royal assent has not in fact been refused since 1708. Constitutional convention helps us to understand this discrepancy between what is possible in legal terms and what actually happens in practice:

  • First, constitutional convention describesmore than three hundred years of history: it tells us that no monarch has sought to exploit their legal power to veto legislation in over three centuries. In this sense, convention tells us what has happened.
  • Second, convention also has something to say about what should happen. It isn’t merely a historical accident that monarchs happen not to have blocked legislation for over three hundred years. Rather, monarchs have not done so because of an understanding that they should not do so. Viewed from a modern perspective, the reason for this sense of what should — and should not — happen is plain, for it would be fundamentally undemocratic if a hereditary monarch attempted to block a law that had been adopted by the legislature.
  • Third, convention tells us what is likely to happen. The sense of constitutional morality that has caused successive kings and queens over hundreds of years to assent to legislation is very likely to continue to exert a powerful influence, causing the monarch to continue to observe this practice by assenting to legislation that has been passed by the House of Commons and the House of Lords.

The upshot, then, is that while monarchs can, as a matter of law, block legislation, the position is very different once matters are viewed through the lens of constitutional convention. Looked at it from that perspective, monarchs have not, should not and will not veto legislation, thanks to the weight of historical precedent and the sense of moral obligation that arises not only from that precedent but also from the powerful constitutional reason — namely, democracy — that tells against monarchical veto of laws approved by Parliament.

Why not leave it all to law?

Of course, all of this invites a fairly obvious question. If constitutional reality is so very different — at least in this instance — from the position described by constitutional law, why not just change the law so that it matches up with reality? Why say, in one breath, that the Queen has the power to veto legislation and then, in the next breath, that she never does this, mustn’t do it and won’t do it? And, stepping back from this specific question, a broader one arises: why rely on constitutional conventions at all — why not make sure that there is constitutional law in place to deal with important constitutional matters? Nor, it’s important to bear in mind, are constitutional conventions limited to this context. They apply to a host of other matters too: it is, for instance, convention, not law, that says that Government Ministers are responsible to Parliament and collectively responsible for Government policy; that the House of Lords should not oppose legislation that implements Government manifesto commitments; and that governs how decisions are made about the formation of new Government following a general election. Why leave such matters to convention?

In trying to arrive at answers to those questions, four overlapping factors are worth considering:

  • First, to some extent, it is inevitable that law will end up being supplemented by informal practices or non-legal rules that develop over time. It is impossible (and probably unwise) to legislate for every eventuality, and even in countries with written constitutions, practice, tradition and convention still have their parts to play.
  • Second, ‘if it ain’t broke, don’t fix it’. Given that no monarch has sought to frustrate Parliament by vetoing legislation in over three centuries, it might be thought that there is little point in legislating to take the power away. (On a related note, we might observe, in passing, that there must be very few, if any, laws that are obeyed without exception over such an extraordinarily long period of time!)
  • Third, conventions enable a form of constitutional ‘doublethink’ — that is, they allow us (in a way) to reconcile two constitutional ideas or values, each of which might be appealing and neither of which we might want to deny. For example, the Royal Assent convention allows us (to an extent, at least) to say both that the monarch has meaningful powers (thus reassuring monarchists that the monarch is not legally irrelevant) and that the UK is a democracy (because in practice an hereditary monarch cannot veto legislation adopted by Parliament). In this way, views of the constitution that are in tension with one another can be reconciled — or at least allowed to peacefully co-exist. 
  • Fourth, and finally, conventions reflect what we might call the art of the constitutionally possible. Sometimes, convention is the best that can be offered if we wish to avoid disturbing a legal arrangement that is considered sacrosanct. Again, the Royal Assent convention provides an example. So too does the ‘Sewel convention’, which says that the UK Parliament will not normally legislate on devolved matters (ie things that the Scottish Parliament, the Northern Ireland Assembly and the Senedd, or Welsh Parliament, can legislate on) without the consent of the devolved legislature in question. Here, convention once again attempts to reconcile two apparently irreconcilable constitutional ideas: that, on the one hand, the UK Parliament is sovereign, and can therefore make whatever laws it wishes (even if, in doing so, it rides roughshod over the devolved legislatures) and, on the other hand, that if devolution is to work, and if the devolved institutions are to be accorded proper respect, the UK Parliament cannot simply interfere whenever those institutions do things that it doesn’t agree with.

Is the UK constitution too reliant on conventions?

While some reliance on practice, tradition and convention is, as already noted, inevitable, there is an important question about whether the UK constitution is too reliant in this regard. As the examples given above demonstrate, convention in the UK does not merely fill in the gaps where legal rules are incomplete in some minor respect. Rather, they reflect and give effect to fundamental constitutional principles and arrangements. This means, in turn, that the UK is heavily reliant upon those who are governed by conventions respecting them, bearing in mind that conventions, because they are not laws, cannot (at least in any straightforward way) be enforced in the way that legal rules can be.

If politicians and others who are involved in the operation of the constitution are willing to ‘do the right thing’ by respecting conventions and the principles that conventions embody, little difficulty arises. But it is important to recognise that the system assumes our willingness to place a high degree of trust in such people — and it cannot, and should not, be taken for granted that that trust will always be repaid. Another way of putting this is to say that how relaxed we are about leaving important constitutional matters to convention is closely linked to how much trust we think it is sensible to place in politicians and others to behave properly — a matter on which readers will no doubt have their own views.

The image accompanying this article is reproduced under a Creative Commons Attribution-ShareAlike 2.0 Generic (CC BY-SA 2.0) licence. A link to the original version of the image can be found here.