1st Prize in VSP 2023: The Rule of Law. Its Meaning and Purpose.

Rule of law: The universal unwritten constitution

Martin van Staden


Constitutionalism – one of classical liberalism’s significant legacies in jurisprudence – is the idea that the state must be subject to law; in particular, that state power must be subject to legal limitation. Every written constitutional instrument in the world differs on the scope and extent of this limitation – granting some commonalities – but there is a universal core of rules and limits to which the state is regarded as subject. This core is derived from the very nature and logic of the law, and has been referred to as the ‘Rule of Law.’ This brief contribution explores what the Rule of Law (alternatively, the ‘constitutional state’) means, and what ends it is ultimately directed toward. It is concluded that the Rule of Law is that legal institution that, as far as possible, attempts to minimise arbitrariness in the content, making, and application of public (state) law.


The term ‘Rule of Law’ reveals the essence of its meaning: it is the law that rules.1 This excludes mere political opinion masquerading as law, for to classify this as ‘law’ would be to recognise the ‘rule of man.’ Only real law can rule, and ‘law’ has a definite existence outside of the agenda of whatever party controls the legislature.

Law and constitutions2 have always existed, but prior to the entrenchment of constitutionalism, the understanding was that the law did not apply to the supposed
sovereign: princeps legibus solutus est. The dawn of constitutionalism amounted to a revolution in jurisprudence that is inherently entwined with classical liberalism, and
brought with it a discernible notion of ‘public law’ – the law as it relates to the state, the state’s relations with legal subjects, and relations between state entities inter se.
Public (that is, state) law is necessarily accompanied by the notion of the Rule of Law.

This liberal constitutional thinking, which was arguably instigated by John Locke – although it would only come to fruition long after his death – and under which we still live today (Grayling 2007, 127-128, 260),3 largely stands on two interrelated legs. The first is that the core directive of the state is to protect the inherent liberty of legal subjects. The second leg – wherein the seeds of the Rule of Law lie – is that the state may not act capriciously, whimsically, or arbitrarily, but reasonably, and only through law.

Constitutionalism, then, is the more general term that encompasses both the Rule of Law (Mathews 1971, 274), and also the written constitution, if any.

The Rule of Law exists only in the public or state law paradigm, and only in circumstances where law has been ‘made’: legislation, regulations, or judicial pronouncements. Indeed, the continental equivalent of the Rule of Law, the Rechtsstaat or ‘constitutional state,’ implies this in the term itself. The Rule of Law therefore has limited application in private law or the ius commune:4 The notion plays no role in a contractual agreement between parties, although it does come into play if a dispute arises out of the contract and the courts – bodies within public law – become involved.5 The court’s judgment, as binding law that is ‘made,’ must satisfy the requirements of the Rule of Law, but the court need not – indeed it ought not – require that the contract or the parties thereto satisfy those same requirements.6

It is evident that the Rule of Law is a recognition of the inherent nature and logic of public law and law-making, and for this reason can be thought of as universal. A law
anywhere is a rule that is binding, and to not adhere to it is regarded everywhere as worthy of punishment. Here enters the Rule of Law: For that legal rule to be binding,
people must know about it, and more than that, they must be capable of understanding it fully, and must be capable of complying with it.7 If these requirements are not met, that rule is not law, as it defies logic and causality (not merely fairness or justice) to expect obedience under such circumstances. Indeed, there is nothing ‘Western’ about the expectation that a binding law must be known to those who are expected to comply with it. There is no ‘Rule of Law’ in a dictatorship, which is self-consciously the rule of man.

There is also no ‘Rule of Law’ in an unqualified democracy: such a democracy is self- consciously the rule of many men. There is only ‘Rule of Law’ where the law is recognised as an extra-political phenomenon that – while given items of ‘legislation’ can be ‘made’ by legislatures – ‘the law’ itself has an independent content and
existence that must also govern the state itself. The Rule of Law can only exist in the same breath as constitutionalism and limited government (Mathews 1971, 6), and for
this reason, the Rule of Law is a phenomenon that coheres with liberal political thinking (Sartori 1976, 15-17).

First leg of constitutionalism: the civil-libertarian purpose of government

Civil liberty is freedom under the law (Brookes and MacAulay 1958, 1). Lawless freedom might be peaceful and might create prosperity, but it could also be violent and
unstable. Law creates order, in this case, by recognising and defining each person’s sphere of free action. Where the liberty of one ends and that of another begins is the
question law answers both in principle and in discrete circumstances. Constitutionalism is often said to mean that the state may not do that which it is not
explicitly allowed to do by law, and the people may not be prevented from doing that which they are not explicitly prohibited from doing by law. To borrow a term from British constitutional law, this is known as the general power of competence (Cockell 2013, 8).

In other words, that which is not legally allowed for the state is forbidden, and that which is not legally forbidden for the people is allowed. Inherent in this formula is the
recognition of the distinction between the public (or the governmental) and the private (or the civil). Thus, when the state does what is not allowed, it almost always infringes on someone’s liberty, and when the people do what is forbidden, they usually commit crimes. The law – inherently – therefore protects the private sphere.

The private sphere is where a legal subject, or group of associating subjects, are secure from coercion by others or by the state. Hayek (1960, 206) wrote that only ‘in
a society that has already attempted to prevent coercion by some demarcation of a protected sphere can a concept like “arbitrary interference”’ in the liberty of another have meaning.

All endeavours of constitutionalism therefore seem inseparable from the limitation of government power in favorem libertatis.

Second leg of constitutionalism: the legal prohibition on arbitrariness

The Rule of Law is premised largely on the aforementioned second leg of constitutionalism: the abhorrence of arbitrariness. This is why the eminent scholars on the Rule of Law have been and remain primarily concerned with the phenomenon of the unconstrained exercise of discretion by the state in society.

While a cogent argument can be made that the Rule of Law is itself concerned directly with protecting liberty, it does so indirectly. Nonetheless, as Mathews (1986, xxvii, xxix) explains, there exists ‘an intimate and permanent relationship between freedom and the rule of law’ precisely because ‘the legal control of the government is in the interests of freedom and justice.’ As an aspect of constitutionalism, the Rule of Law is about regulating how precisely the state may exercise its limited power. It comprises various requirements about the content of law, the ‘making’ of law, and the application of law in practice.

The Rule of Law, as far as is possible, seeks to ensure that legal subjects are governed by law, not by the whims or caprice of state officials, hence the contrast with the ‘rule of man’ and why a defining characteristic of the Rule of Law is that both governor and governed must be subject to the same law. This necessarily follows from the notion that it is ‘the law that rules,’ because only a ‘rule of man’ could produce legal privileges for a political elite over ordinary subjects. Such a privileging is ultimately arbitrary.

It is therefore submitted that an apt definition for the Rule of Law is that it is the legalinstitution that regulates or attempts to eliminate state arbitrariness. The opposite of
arbitrariness is reasonableness, meaning the Rule of Law is necessarily an exercise in ensuring the law is made, interpreted, and applied reasonably (that is, there is a
reason for the law, it can be justified, it is not incoherent, and it is practically realistic). For the Rule of Law to exist – where arbitrariness is regulated or eliminated – certain things are necessarily required, including:

  • The law, or official conduct, whatever the case, must be knowable. This means that every rule one is expected to obey must be both published and publicised. Retroactive law is not knowable and therefore disallowed.
  • The law must be accessible. A layperson must be able to easily find, read, and understand the rule themselves.
  • The law must be unambiguous. There must be only one reasonable meaning that is ascribable to that rule, and that meaning must on its face be evident to those
    who have to comply.
  • The law must be certain. If one is expected to obey a rule or face jeopardy, the rule may not be implied or unstated.8 The rule must also not be subject to regular change.
  • The law must be reasonable. This means that there must be a good reason for the rule, and this reason must be backed by evidence and lucid argument – justification to an impartial body – rather than mere preference or expedience. It also means that the reasonable person in society, understood to be Joe Public, must not regard the rule as unfair.
  • The law must be proportional. As a leg of reasonableness, this essentially means that the rule must do no more than necessary to solve the problem that it is addressed to. It must interfere as little as possible in the spontaneous free development and course of society.

Koos Malan (2012, 276-280) has also theorised on some core features of the Rule of Law, noting that there must be a corpus of law, distinguishable from non-legal considerations such as personal morality or popular preferences. The application of law therefore is forward-looking, relying on the authority of the past. In other words,
there must be a temporal separation between the rule’s creation and the rule’s practical application. Law must have a written foundation if the corpus is to be clearly
recorded and distinguished from non-legal considerations. The application of law must also be characterised by impersonality and abstractness.

Without all these requirements of the Rule of Law being adhered to, the purpose of (public) law is itself – to a limited or fatal degree – defeated.9

This highlights that the Rule of Law is nothing more than a recognition of the internal consistency and essence of law, if law is regarded as that system of rules that govern human behaviour to the end of delimiting the spheres of free action. After all, how can the law effectively fulfil this function if the law is not known, accessible, certain, reasonable, and based on evidence? Discretionary power that is unconstrained in substance also undermines this function of law, because whatever a state official might whimsically decide is completely unclear and unknowable.

Realities of modern politics

While the Rule of Law is routinely invoked as a primary principle of modern government, exceptions to compliance with its strict standards are common:

  •  ‘Yes, the Rule of Law does require clear and unambiguous legal provisions, but if we insist on this too rigidly the legislature would have been unable to adopt many
    of the statutes that we take for granted today.’
  •  ‘Yes, the Rule of Law does require legal provisions to be understandable and accessible, but modern society is complex and the legal regulations applying to it
    will naturally be equally complex.’
  •  ‘Yes, the Rule of Law does require that the same law applicable to legal subjects must apply to the political class, but in pursuit of social transformation certain
    positive discriminations may be justified.’

Political invocations of the ‘Rule of Law’ often amount to little more than virtue- signalling, used to benefit from the Rule of Law’s formal appeal but with scant respect or intention to adhere to its substance.

Indeed, the requirements of the Rule of Law might well appear unduly strict. One might wonder how the state could get much, if anything, done, if all state rules and conduct were rigidly measured against these standards (Ocran 1971, 40-41). This is particularly the case when one realises that much of the legislation and regulations
one might approve of – such as zoning, building codes, and maximum speed thresholds on the roads – contain some of the very arbitrariness the Rule of Law is aimed at averting.

This restraint of state activity, however, is exactly what the Rule of Law, and more particularly constitutionalism, exists for. The purpose of law, to simply define legal subjects’ spheres of free action, has long been subverted (Bastiat 2007, 1) primarily by legislatures, and this would not have been the case had the spirit of constitutionalism and the Rule of Law been observed. Adherence to the Rule of Law would produce only basic justice, which is considered below.

Before this subversion, there was a time when representative assemblies (that is, legislatures) were not ‘law-making’ bodies per se (Leoni 1972, 9-11). There was already law that existed – the ius commune – meaning there was no reason to have a conception of ‘law-making.’ Summarising the founding idea behind the federal legislature in the United States, for example, Randy E Barnett (2014, 38) writes:

‘The danger of viewing legislatures as rulers rather than as checks on rulers had been warned of even as the English Parliament was asserting its authority against
the king. Writing in 1651, Isaac Penington, Jr., observed: “The proper use of Parliaments is to be a curb to the extravagancy of Power, of the highest standing Power: But if they themselves become the standing Power, how can they be a fit curb for it?” […] In sum, after their initial experimentation in “republican” or democratic rule, the founders devised a new scheme in which the electorate of “the People” by voting in elections would exercise, not a lawmaking power, but the power to check the lawmakers. “We the People” would not rule directly, but an electorate reflecting the rights and interests of the people would have effective power to check those who would issue commands to the people.’

This is likely why early liberals, like Locke (1823, 114), were generally comfortable with the notion of democracy. The contemporary classical liberal reader would find it
jarring that Locke could, in one breath, argue that governments must be democratic, and at the same time have a strictly limited scope: that is, the protection of liberty and property only. This is because we view Locke from a modern perspective, where democracy has become synonymous with the notion that the electorate goes on a
‘law-shopping spree’ by determining the composition of the legislature. In Locke’s days, the notion of what ‘law’ is, was significantly less fluid. Democracy was about
choosing the identity of governors, holding them accountable, and ensuring the voices of the electorate could be heard by said governors.

Legislation did exist, to be sure, but these did not as a general rule dictate preferred behaviour to legal subjects. The ius commune ably did that. Legislation was what Hayek (1978, 78) regarded as thesis, laws ‘of the superstructure of government originally erected only to ensure the enforcement of private law.’ These instruments
organised the structures of, and set the objectives for, the state.

This does not mean there was complete individual freedom prior to what Guido Calabresi (1982) calls ‘the age of statutes’: the ius commune still existed side-by-side
with the notion that the sovereign may tax, seize property, and enforce public order and morality – but this was not a matter of law as understood today. There was no
Sales Tax Act in 10,000 BC, nor did there (or does there) exist a ‘common law duty to pay taxes,’ even though taxes certainly existed.10 Today we take for granted that every aspect of government has an enabling legal provision – a legal text that clearly empowers the state to do something – but this was not always the case. It is a relatively recent phenomenon – constitutionalism – that state behaviour came to be defined within the ambit of law.

But even then, the arbitrary whims of man can and very often do manifest themselves in law, be it in legislation or in the case law of judicial pronouncements. Judged from the perspective of the Rule of Law, these manifestations are of the ‘rule of man.’ This is why it is important to distinguish between the notion of ‘the law’ on the one hand, and mere ‘laws’ on the other. The rule of man, even when manifested in laws, is not the Rule of Law. Hayek (1960, 206) said it best when he wrote:

‘The rule of law is therefore not a rule of the law, but a rule concerning what the law ought to be, a meta-legal doctrine or a political ideal.’

The Rule of Law, then, is not, as is often errantly assumed, the so-called duty to obey political authority. As a doctrine, the Rule of Law does not focus on ordinary legal
subjects and how they obey or disobey legal rules. The Rule of Law applies only to state authorities (Van Schalkwyk 2017). The Rule of Law thus exists above and
throughout the law, rather than being merely ‘a part’ of law, and addresses itself to the state.

The likely outcome of adherence to the strict standards of the Rule of Law in practice, will be the realisation of a dispensation with a significantly more coherent so- alled
‘social contract’ foundation that respects basic justice.

The Rule of Law’s universal guarantee of basic justice

Barnett (2014, 39-43) explains that the constitutional social contract does not truly exist, because in part it is impossible for everyone, today or in future generations, to
have consented to be governed in the way that they are by the modern state. Barnett nonetheless argues that laws can be assumed to have received consent when they relate to what I call ‘basic justice.’ In his words (ibid, 44), laws must be ‘necessary to protect the rights of others and proper insofar as they do not violate the preexisting rights of the persons on whom they are imposed.’

This approach – basic justice – is effectively that absolute minimal conception of justice that we can intuitively assume the average person everywhere on Earth agrees
to. These are the ‘laws’ against murder, rape, and robbery, etc. – the mala in se. They are clearly just laws, and virtually every communist, nationalist, liberal, libertine,
Christian, atheist, Jew, and Muslim would agree that they are just.

Beyond this basic justice, however, everyone has different conceptions of substantive justice, and disagree wildly about what the law should be. And at this level, it can no longer be assumed that everyone ‘consents,’ even in the abstract, to the application of laws that conflict with their respective notions of justice. These laws, which go beyond the minimum basic justice, invariably infringe upon our spheres of free action.

Imposing substantive conceptions of justice on nonconsenting individuals with different conceptions of justice is an act of arbitrariness, which the Rule of Law proscribes. To some, not working on a Sunday is a matter of the highest importance. But to make this a matter of law, to those who do not buy into this injunction, amounts to imposing something completely arbitrary upon them.

It is submitted that these laws – that give effect to substantive conceptions of justice – are not necessarily binding in conscience. If that is the case, obeying them in practice is not a duty, even though it might be wise for self-preservation. That determination, of course, can only be made on a case-by-case basis – the mainstay of legal reasoning. Were the strict standards of the Rule of Law truly observed, what society would be left with – in public law – is a framework of basic justice. It is in this manner that the Rule of Law indirectly, but unavoidably, serves liberty. Where more than basic justice is legally pursued, arbitrariness and an unconstrained notion of law is guaranteed, and the Rule of Law evaporates.


Martin van Staden is a South African jurist, author, and policy advocate currently serving as the Head of Policy at the Free Market Foundation (founded 1975).

He received a Master of Laws (LL.M.) degree cum laude from the University of Pretoria in 2020 and is presently pursuing his Doctor of Laws (LL.D.) at the same institution. Martin previously served as a Legal Fellow at the independent business community Sakeliga and Deputy Head of Policy Research at the Institute of Race Relations (Africa’s oldest classical liberal institution, founded 1929). His academic work has been featured in the Cato Journal, the Economic Freedom of the World annual report, the Potchefstroom Electronic Law Journal, the Journal for Contemporary Roman-Dutch Law, Cosmos + Taxis, and the Pretoria Student Law Review, among others. His book, The Constitution and the Rule of Law: An Introduction, was published by the Free Market Foundation in 2019.


1 Clearly distinct from ‘a rule of the law.’

2 While the notion has been shrouded in mystique for many centuries, any political community from even before the formation of the modern state had ‘a constitution’: the framework and functioning of public authority.

3 In form, if not in substance.

4 The ius commune is the ‘common law,’ but should not be confused with the distinctive English Common Law tradition.

5 Other legal rules that appear similar to Rule of Law requirements, such as the courts inquiring into whether a contractual provision is suitably clear, emanate from private law and not the ‘meta-legal doctrine’ (to borrow from Hayek) that is the Rule of Law.

6 Whether the courts ‘make’ law or merely ‘find and explain’ it is a separate debate. For our purposes it suffices to note that even if one firmly believes the courts do not make law, their ‘explanation’ of law in most contexts has highly persuasive – if not necessarily binding – effect.

7 Here it is interesting to note that uncodified common-law crimes – also legal rules requiring obedience in lieu of punishment – such as rape, murder, and robbery, necessarily satisfy these requirements because they are mala in se: obviously wrong by their very nature.

8 Mala in se, considered elsewhere in this contribution, is distinct. The rules against murder, rape, and robbery, etc., are in fact implied, but these are not examples of law that is ‘made’ – this is law that simply is.

9 In this regard, the work of Lon L Fuller (1969, 38-41) is instructive.

10 Taxation is wholly a creature of legislation. In the absence of tax legislation, all taxes would amount to the state exercising a (extra-legal) sovereign power, which the common law might countenance but which liberal constitutionalism would deem intolerable.


Barnett, Randy E. 2014. Restoring the Lost Constitution: The Presumption of Liberty. Princeton University Press.
Bastiat, Frédéric. 2007 [1850]. The Law. Ludwig von Mises Institute.
Brookes, EH, and MacAulay, JB. 1958. Civil Liberty in South Africa. Oxford University Press.
Calabresi, Guido. 1982. A Common Law for the Age of Statutes. Harvard University Press.
Cockell, Merrick. 2013. The General Power of Competence: Empowering councils to make a difference. Local Government Association.
Fuller, Lon L. 1969 [1964]. The Morality of Law. Yale University Press.
Grayling, AC. 2007. Towards the Light: The Story of the Struggles for Liberty & Rights That Made the Modern West. Bloomsbury.
Hayek, Friedrich A. 1960. The Constitution of Liberty. University of Chicago Press.
Hayek, Friedrich A. 1978. New Studies in Philosophy, Politics, Economics, and the History of Ideas. University of Chicago Press.
Leoni, Bruno. 1972 [1961]. Freedom and the Law. Nash Publishing.
Locke, John. 1823 [1689]. The Works of John Locke: A New Edition, Corrected, in Ten Volumes. Volume V. Thomas Tegg.
Malan, Koos. 2012. The rule of law versus decisionism in the South African constitutional discourse. De Jure. 272-305.
Mathews, Anthony S. 1971. Law, Order, and Liberty in South Africa. Juta.
Mathews, Anthony S. 1986. Freedom, State Security, and the Rule of Law: Dilemmas of the Apartheid Society. Juta.
Ocran, Tawia. 1971. Law, African economic development, and social engineering: A theoretical nexus. 3(1-2): Zambia Law Review. 16-42.
Sartori, Giovani. 1976. Liberty and law. 5: Studies in Law (Institute for Humane Studies). 3-50.
Van Schalkwyk, Rex. 2017. Babylonian gods, the rule of law & the threat to personal liberty. CNBC Africa.

Our Partners

Liechtenstein Academy | private, educational foundation (FL)
Altas Network | economic research foundation (USA)
Austrian Economics Center | Promoting a free, responsible and prosperous society (Austria)
Berlin Manhatten Institute | non-profit Think Tank (Germany)
Buchausgabe.de | Buecher fuer den Liberalismus (Germany)
Cato Institute | policy research foundation (USA)
Center for the New Europe | research foundation (Belgium)
Forum Ordnungspolitik
Friedrich Naumann Stiftung
George Mason University
Heartland Institute
Hayek Institut
Hoover Institution
Istituto Bruno Leoni
Institut Václava Klause
Instytut Misesa
IREF | Institute of Economical and Fiscal Research
Johns Hopkins Institute for Applied Economics, Global Health, and the Study of Business Enterprise | an interdivisional Institute between the Krieger School of Arts and Sciences, and the Whiting School of Engineering
Liberales Institut
Liberty Fund
Ludwig von Mises Institute
New York University | Dept. of Economics (USA)
Stockholm Network
Students for Liberty
Swiss Mises Institute
Universidad Francisco Marroquin