On the liberties of the ancients: licentiousness, equal rights, and the rule of law

ABSTRACT

In this article, we discuss Greek and Roman conceptions of liberty. The supposedly ‘neo-Roman’ view of liberty as non-domination is really derived from negative Greek models, we argue, while Roman authors devised an alternative understanding of liberty that rested on the equality of legal rights. In this ‘paleo-Roman’ model, as long as the law was the same for all, you were free; whether or not you participated in making the law was not a constitutive feature of liberty. In essence, this Roman theory was a theory of freedom as the rule of law and the guarantee of equal rights, especially due process rights. For this Roman concept of ‘legal liberty,’ as we call it, political participation was neither necessary nor sufficient. Theorized by Cicero and historicized by Livy, the Roman understanding of freedom flourished in early-modern times, proving important to paradigmatic republican authors such as Machiavelli and Rousseau as well as to Hobbes, whose work we discuss as a helpful point of comparison.

KEYWORDS:

• liberty of the ancients
• republicanism
• neo-Roman liberty
• Machiavelli
• Rousseau
• Hobbes
• Quentin Skinner
• rule of law
• rights
• due process
• liberalism
• legal liberty

Il est vrai que dans les démocraties le peuple paroît faire ce qu’il veut; mais la liberté politique ne consiste point à faire ce que l’on veut. (Montesquieu, Spirit of Laws, 11.3)

For much of modern times, the liberty of the ancients has had a bad rap. The political idealism that led the citizens of Lucca to inscribe LIBERTAS on their town wall, inviting Hobbes's scorn, lost much of its appeal after the French revolutionaries, seemingly drunk on both liberty and antiquity, baptized their new republic with the blood of their own citizens. Such, at least, was the influential verdict of Benjamin Constant, who decried how the Jacobins had followed the Greeks and Romans in accepting ‘the complete subjection of the individual to the authority of the community.’¹ His defense of the ‘liberty of the moderns,’ which granted greater protections to individuals, would be recast 130 years later by Isaiah Berlin, this time in opposition to the threat of Soviet totalitarianism. Berlin's account of ‘negative freedom’ (freedom from political interference), as opposed to the ancient and proto-totalitarian ‘positive freedom’ (freedom to participate in a political community), long framed analyses of a concept which, as Montesquieu noted, may have had more meanings than any other.²

Since Berlin's time, classicists have provided us with much more fine-grained understandings of the many ways that the ancients conceived of liberty.³ The end of the Cold War also led political theorists to question the sufficiency of freedom from interference as a political ideal. Taken together, these developments encouraged scholars to cast a more charitable gaze on ancient theories. Perhaps the most successful of these efforts has been the endeavour to retrieve a ‘neo-Roman’ definition of liberty, understood as freedom from domination. Elaborated chiefly by Quentin Skinner and Philip Pettit, this theory has been well received by historians working on a broad variety of periods, from ancient Rome to the Enlightenment.⁴

In this article, we highlight a problem with the neo-Roman theory and sketch out an alternative history. The problem is that the theory of freedom as non-domination was in fact well known to ancient Greeks (whom Skinner and Pettit leave out of their story), but widely disparaged by Greek political theorists as licentious and democratic. As these attacks were very familiar to European readers long before Constant, it is difficult to see how such a tainted idea would be so warmly received.⁵ Already in the eighteenth century, Montesquieu had summarized a common criticism of democratic freedom: ‘it is true that in democracies the people seem to do what they want (ce qu’il veut), but political liberty in no way consists in doing what one wants (ce que l’on veut).’⁶

In opposition to this view of liberty imputed to democrats, Roman authors built on Greek ethical and judicial arguments to devise an alternative understanding of liberty that rested on the equality of legal rights. In this ‘liberal’ model, so long as the law was the same for all, you were free; whether or not you participated in making the law was not a constitutive feature of liberty. Theorized by Cicero and historicized by Livy, this Roman understanding of freedom flourished in early-modern times, proving of particular importance for ‘republican’ authors such as Machiavelli and Rousseau. In essence, it was a theory of freedom as the rule of law and the guarantee of equal rights, especially due process rights, as later outlined by nineteenth-century liberals such as A.V. Dicey.

The key difference with Skinner's neo-Roman view is that for Cicero and Livy, having the rule of law and equal rights – what we call ‘legal liberty’ – is sufficient for having liberty: if you have the rule of law and equal rights, you necessarily have a free state.⁷ As Cicero puts it, the essence of the free city–state (proprium liberae civitatis) is the right of appeal (provocatio),⁸ the right of due process every citizen is guaranteed under the law. This kind of equal right cannot be taken away even by the popular assemblies, which is a limiting case for the republican view, for it shows the potential for domination and arbitrariness in popular sovereignty.⁹ For Skinner and the neo-Romans, by contrast, it is participation in ruling and law-making that is sufficient for liberty: if you have participation, you necessarily have a free state. So, for the neo-Romans, a free state means participation and this implies liberty in the sense of non-domination; while for the paleo-Romans such as Cicero and Livy, rule of law and equal rights imply liberty and the free state:

Neo-Romans/republicans:

If free/participatory state ⇒ liberty as non-domination

Paleo-Romans (Cicero, Livy):

If rule of law and equal rights ⇒ free state

For the ancient Romans, then, participation is neither sufficient nor necessary for liberty. This is not to say, of course, that they deny the importance of political rights. Cicero, e.g., insists that everyone have some modicum of participation rights – but this is because justice, not liberty, demands these political rights.¹⁰ For some of the neo-Roman or republican thinkers, especially Philip Pettit, the rule of law and rights have increasingly gained in importance; but the problem is that for the neo-Roman view to retain anything distinctively neo-Roman, it has to maintain the centrality of participation; otherwise it simply collapses into a paleo-Roman or proto-liberal position.¹¹ What this also shows is that the paleo-Roman view, with the priority it accords to law and rights, tends to be rationalist at bottom, while the neo-Romans tend to have a voluntarist, will-based outlook. This might account for the ambiguity we find in a writer such as Rousseau: to the extent that he accords law independent weight, he shows a more rationalist outlook, while he tends to appear more as a republican voluntarist whenever he puts participation front and centre. His identification of popular sovereignty with the rule of law glosses over this ambiguity.

In retracing this history, our ambition is different than Skinner's or Pettit's, as we do not wish to suggest that this Roman concept is sufficient for our times. We would argue, however, that it still has much to offer; and, if it is not sufficient, it is a necessary component of freedom. Designed to resist both plutocratic and populist abuses, it is a theory that speaks to many of the ongoing political challenges of our day. By highlighting the classical prehistory of modern liberalism, we also hope to challenge the standard view of liberalism itself as ‘freedom from interference.’ This definition misses the key constraint that classical and early-modern ‘liberals’ insisted on, namely that laws and rights be equal for all.

1. Neo-Romans and ancient Greeks

For both Skinner and Pettit, the theory of freedom as non-domination rests on an analogy between civil and political concepts of freedom. Drawing on the Roman legal definition of a free person (liber homo), as codified in the imperial Corpus juris civilis, Skinner and Pettit argue that to be free, politically, was structurally homologous to being without a master, or dominus, in the civil sphere. A slave (servus) might enjoy negative freedom, in Berlin's sense: he could perhaps come and go as he pleased. At a moment's notice, however, his master could command him to perform a certain action. In this respect, he was not genuinely free, as he remained in a literal state of domination (regardless of whether the dominus chose to impose his will). Political freedom, they argue, is much the same: it is not enough that we enjoy the freedom to come and go as we please, and act as we will (‘freedom from interference’). If there is a potential for domination – a benevolent dictator, an emperor, or an unelected bureaucracy – then we are not free.¹²

As a theory of freedom, this neo-Roman account has much to commend it. It is mainly from a historical perspective, we argue, that there is a problem. This problem stems from the fact that the Greeks had a very similar definition, but it was widely condemned. In democracies, Aristotle wrote in the Politics, freedom (eleuthería) is defined as ‘doing just what one likes’ (boúletai): ‘everyone lives as he likes.’ To live as one likes, Aristotle adds, is to live the life of a free person, since ‘to live not as one likes is the life of a man who is enslaved’ (douleúontos).¹³ The analogy between civil and political freedom was thus explicit, unlike in the Digest. Democrats favoured this definition, Aristotle observed, since they privileged two values above all: majority rule and liberty.¹⁴

This last point might seem like a case of circular reasoning, but Aristotle is almost certainly referring back to a key historical moment in Athenian history: the famous seisachtheia, or ‘shaking-off of burdens,’ ordered by Solon. In the Athenian Constitution, probably written by Aristotle or one of his students, we learn how Solon cancelled all debts and prohibited the enslavement of Athenian citizens for debt. These laws also liberated formerly enslaved (douleuótōn) Athenians. Henceforth, civil freedom really did become a defining feature of democratic citizenship.¹⁵

The democratic freedom to ‘live as you like’ was thus explicitly related to the civil status of not having a master, or despotes, who could thwart your will. Aristotle himself emphasized the double meaning of ‘despot,’ using the term in both its civil and political senses. Constitutions that only serve the interests of their rulers ‘have an element of despotism’ (despotikaì), he argued, which made them incompatible with a just city, as ‘a city is a partnership of free men’ (koinōnía tōn eleuthérōn estín).¹⁶

Discovering a precedent for Skinner and Pettit's theory of freedom in Aristotle could be seen as further evidence of its wide-ranging application. In some respects, this interpretation may be warranted. But it also faces a serious challenge. In no uncertain terms, Aristotle characterized this democratic definition of liberty as misguided: ‘they [the democrats] define liberty wrongly’ (kakōs); ‘this is bad’ (phaulon).¹⁷ Any reader of Aristotle would have come away with a clear sense that this democratic conception of freedom was profoundly dangerous, and certainly not what Aristotle himself advocated.

There were deep historical reasons for disparaging the idea of freedom as ‘doing what one likes.’ Herodotus repeatedly described Cambyses, the mad son of the Persian king Cyrus the Great, as ‘doing as he likes’ (poiéein ta boúletai), behaviour that Otanes, in his famous contribution to the Constitutional Debate, identifies as the hallmark of the tyrant (tyrannos): ‘How can monarchy be a fit thing, when the ruler can do what he wants with impunity?’ (poiéein tà boúletai).¹⁸ Sophocles similarly associated tyranny with ‘doing and saying what one likes’ (dran légein th’a boúletai) in Antigone.¹⁹

But this criticism also reflects Aristotle's own understanding of the rule of law. The democrats are wrong, he insisted, because ‘to live in conformity with the constitution ought not to be considered slavery, but safety (douleían … allà soterían).’²⁰ The implication here is that, left to its own devices, democratic freedom will destroy a democratic constitution. Aristotle's discussion about the democratic conception of liberty, in fact, occurs in book 5, where he considers the different ways in which ‘revolutions in constitutions’ (metabállousin ai politeiai) take place.²¹

The most likely candidate for the change Aristotle had in mind here was the ‘demagogical’ scenario, typical of fifth-century Athens. The danger with demagogues is that they ‘cause the resolutions (tà psephísmata) of the assembly to be supreme and not the laws (toùs nómous), by referring all things to the people.’ In this way, they destroy the constitution, since ‘where the laws do not govern there is no constitution.’ The problem with letting the people ‘do as they like,’ then, is that they no longer accept any restraints, including constitutional laws. The result is a kind of tyranny, as ‘the decrees voted by [such an] assembly are like the commands issued in a tyranny.’²²

A classic example of such a ‘tyranny’ occurred in 406 BCE when the Athenian assembly, egged on by demagogues, sentenced six victorious Athenian generals to death for failing to collect shipwrecked sailors after the Battle of Arginusae. As recounted by Xenophon, the assembly's proceedings were highly irregular, even unconstitutional (paranómos). The generals were not allowed to speak in their own defense, nor were they tried according to an established law, nor even offered individual trials. But the majority insisted that the people (demos) should not be prevented from doing as they wished (boulétai) – precisely the argument for democratic freedom.²³ Among the presiding officials, Socrates alone insisted that he would only approve what was allowed by law (kata nómon). Indeed, throughout this incident, and particularly in the speech by Euryptolemus (Alcibiades's cousin), the contrast between what is lawful (kata ton nomon) and what is unlawful (para ton nomon) is repeatedly emphasized.²⁴ Here we catch a glimpse of an alternative ideal of freedom, based on the rule of law, albeit not (yet) framed in terms of rights, or even labelled as freedom. This episode also sheds light on Socrates's own insistence, in Plato's Crito, that ‘the laws speak the truth’ (álethē légein tous nómous), even when they condemned him to death, and that he could never be a nómon diaphthoreús, or ‘destroyer of laws.’²⁵ Instead, Socrates went to his death proclaiming his freedom (eleuthería), using the term in a sense clearly and strongly opposed to ‘doing as one likes.’²⁶

Aristotle's own preferred idea of freedom was similarly dependent on certain qualities of the law. Without any guarantees that the legal and political order will not be summarily overturned, or selectively applied, there can be no freedom. Aristotle of course recognized that there are times when the laws must be changed. But if ‘it is proper for some laws sometimes to be altered,’ he cautioned, ‘it is a bad thing to accustom men to repeal the laws lightly.’ The reason is one that political theorists often overlook, and concerns the normativity of the law.²⁷ How do laws receive their force? Why do we obey them? For Aristotle, the answer had to do with time: ‘the law has no power to compel obedience beside the force of custom, and custom only grows up in long lapse of time.’ This was the reason why demagogues, by encouraging the people to overturn existing laws at will, replaced the rule of law with a type of tyranny: ‘lightly to change from the existing laws to other new laws is to weaken the power of the law (to nómou dúnamin).’²⁸ As we will see, later authors, writing in a Roman republican tradition, would equate this rule of law with freedom.

2. Greek democracy and its critics

The specific problems with the democratic idea of freedom are laid out in Thucydides's history of the Peloponnesian War, where we also find one of the rarer, more favourable accounts. In his funeral oration, Pericles celebrated ‘the freedom (eleuthéros) which we enjoy in our government,’ which he identified as democratic. Unlike in regimented Sparta, he continued, ‘far from exercising a jealous surveillance over each other, we do not feel called upon to be angry with our neighbor for doing what he likes (hedonén).’²⁹

But this positive assessment was swiftly overturned. While ‘doing what one likes’ in the civil sphere, as an individual, could indeed be assessed favourably, as it was by Thucydides's Pericles, ‘doing what one likes’ in the political sphere, or in the aggregate, posed difficult problems in the view of almost all Greek writers on politics. The ‘hedonism’ of free Athenians acting in concert, in Thucydides's reckoning, turned out to be a root cause of the city's downfall. After evaluating Pericles's career, Thucydides assessed those who followed him: ‘With his successors it was different. More on a level with one another, and each grasping at supremacy (tou prōtos), they ended by committing even the conduct of state affairs to the whims of the multitude (hedonàs tō démō).’³⁰ Here loomed the spectre of Aristotle's demagogues, egging the multitude on to act ‘hedonistically,’ or to do as they like.

One of the demagogues Thucydides had in mind was the young and impetuous Alcibiades, who played a key role in convincing the Athenians to embark on the fateful Sicilian Expedition.³¹ Alcibiades's openly oligarchic (and possibly tyrannical) ways ultimately proved distasteful to democratic Athens, and he fled to rival Sparta to avoid likely execution. There he denounced his countrymen in both moral and political terms. Referring to his Alcmaeonidae family, he observed how ‘we endeavored to be more moderate (metriōteroi) than the licentious (akolasías) temper of the times.’³² Where Aristotle described the democratic idea of freedom fairly neutrally as ‘doing as one likes,’ Alcibiades employed a much more loaded term. Akolasía, or acting without restraint, was a distinctly oligarchic slur for disparaging hoi polloi. In one of his orations, Isocrates even opposed akolosía and eleuthería – as Cicero would after him – while another oligarchic writer associated akolosía with the lower classes.³³

Alcibiades may in fact have derived his analysis from one the most famous Greek critics of democratic freedom. In book 8 of Plato's Republic, Alcibiades's friend and mentor Socrates forcefully disparaged the democratic man who believes he has ‘licence (exousía) to do as he likes (tis boúletai).’ Such a man ‘calls this life of his the life of pleasure (hedún) and freedom (eleuthérion) and happiness.’³⁴ But for Socrates it is really an instance of ‘licentiousness (akolasía) and disease multiply[ing] in a city.’³⁵ True philosophical freedom, or what Fred Miller called ‘aristocratic freedom,’ cannot be understood in this democratic fashion, but requires domination of the best part of the soul over the others.³⁶ The ‘licentiousness’ that democrats identified as freedom – doing as we like – is equated, in the Socratic discourse, with injustice, violence, cowardice, and madness.³⁷

3. Political inequality and legal equality

One of the principles driving elite opposition to democratic freedom was an acceptance of political inequality. Political inequality was the inevitable corollary of economic inequality, which most commentators regarded as an invariable feature of human existence. ‘The state consists of unlike persons,’ Aristotle matter-of-factly observed.³⁸ Give the rich the same political rights as the poor, and they will resent the lack of consideration due to their rank. But deprive the many of any say in political decisions, they will rise up in protest. ‘Party strife (stásis) is everywhere due to inequality,’ concluded Aristotle.³⁹ The best solution, he characteristically surmised, was to split the difference: extend political rights to the many, but restrict some political offices to the few.

The inequality that Aristotle, among others, believed necessary in a well-ordered state was in fact due to a tension between two kinds of equality. The many insisted on numerical or ‘arithmetic’ equality. This was the democratic kind of equality, where every vote has the same value. Democracy itself was often understood in terms of equality: Herodotus uses isonomia as a synonym for popular government.⁴⁰

But the wealthy privileged ‘equality of worth’ (axíā).⁴¹ Just as some athletes or soldiers were superior to others, so, too, were some citizens. This was the oligarchic principle of equality, and the basis of many elitist positions. The problem with democracy, Plato has Socrates declare in the Republic, is that it assigns ‘a kind of equality indiscriminately to equals and unequals alike!’⁴² Again, some states combined both forms: even in democratic Athens, for instance, military leaders were elected, rather than drawn by lot.⁴³

Where certain forms of justice were concerned, by contrast, only one kind of equality seemed fitting. This was the case with corrective or commutative justice (diorthotikós), which Aristotle discussed in the Nicomachean Ethics. In cases involving theft, murder, adultery, or other crimes, he writes,

it makes no difference whether a good man has defrauded a bad man or a bad one a good one, nor whether it is a good or a bad man that has committed adultery; the law looks only at the nature of damage, treating the parties as equal.⁴⁴

In criminal or civil justice, the assumption was that all are equal before the law, an idea that we also find expressed in Pericles's funeral oration: ‘If we look to the laws, they afford equal justice to all in their private differences.’⁴⁵ It was this ideal of justice that was denied to the Athenian generals, at least in Xenophon's account.

Interestingly, while for Pericles equality before the law is a feature of Athenian freedom, corrective justice does not play an important role in the Politics. When Aristotle imagines the lawless, unfree state, it is not the criminal justice system he has in mind, but the constitutional order itself. Accordingly, the proper administration of ‘corrective’ justice is not one of the defining features of a free state. There is a hint in the Ethics that a flawed justice system can be experienced as a loss of freedom.⁴⁶ And Plato drew a contrast between a well-functioning justice system and freedom, understood in his more aristocratic sense.⁴⁷ So while we can find in Greek political thought both the concepts of legal equality, and of freedom as a result of the rule of law, the two are not united. Fusing them would be the work of Roman philosophers and historians.

4. Roman rights and libertas

The Greek critique of democracy was well known to the Romans. The historian (and Roman hostage) Polybius reproduced its main tenets in his description of the Roman constitution. Like many of his predecessors, he rejected the claim that democracy was the regime that best preserved freedom:

And when that comes to pass the constitution will receive a new name, which sounds better than any other in the world, liberty or democracy (tìn eleutherían kaì demokratían); but, in fact, it will become that worst of all governments, mob-rule (ochlokratían).⁴⁸

Democracy was like a ship without a captain, Polybius argued; sometimes in the face of danger everyone acted with one mind, but in normal times, dissension ruled, and the ship of state could capsize in the harbour.⁴⁹

Polybius's celebration of the Roman republic over Athenian democracy, due not least to his downgrading of the importance of virtue as a criterion of successful political orders and an upgrading of institutions, became something of a commonplace in Roman political thought.⁵⁰ Cicero gave Polybius pride of place in his own work of political thought, On the Republic, where he also challenged the attribution of ‘liberty’ to democratic government: ‘if the people hold the supreme power and everything is administered according to their desires (arbitrio), that is called liberty (libertas), but is really licence (licentia).’⁵¹ Once again, Athens served as the prime example of how democratic liberty devolved into unbridled licentiousness: ‘the absolute power (potestatem omnium) of the Athenian people … changed into the fury and licence of a mob (ad furorem multitudinis licentiamque).’⁵² This commonplace outlived the republic itself, and continued to feature in political writing of the imperial age.⁵³

It is true that Roman writers also defined political freedom in opposition to slavery. Cicero insisted that freedom was not living under a ‘just master’ (iustus dominus), but under no master whatsoever.⁵⁴ And in Livy's History, life under a king was compared to slavery.⁵⁵ Comments such as these have led classical historians to extend Skinner's neo-Roman definition of liberty back to Roman sources. Valentina Arena proposed that republican Roman politicians shared an ‘idea of liberty as a state characterised by the absence of a condition of domination.’⁵⁶ Jed Atkins similarly claimed that the ‘Romans fundamentally conceived of liberty as the absence of slavery,’ thus confirming that what Skinner and Pettit's understood as ‘neo-Roman liberty’ was also (one might say) paleo-Roman.⁵⁷

But there was an important difference with these Roman definitions. Cicero was perhaps the first to express their theoretical foundation, drawing in part from Greek precedents. First, freedom was the political condition enjoyed by citizens of a well-balanced state. As long as the monarchy was law-governed, there was provocatio and thus liberty,⁵⁸ but – due to the inherent instability of monarchy as a simple constitutional form – it is really in a well-balanced, law-governed republic that the people are ‘free from the domination of kings or senators’ (et a regum et a patrum dominatione solere in libertatem rem populi vindicari).⁵⁹ Yet as the Athenian example demonstrated, secondly, the mere absence of elite domination was an insufficient condition for liberty. What made republican citizens free was not participation in lawmaking, but rather the fact that they enjoyed equal legal rights: ‘the legal rights […] of those who are citizens of the same commonwealth ought to be equal (iura … paria … inter se).’⁶⁰

Indeed, Cicero paired libertas with aequabilitas, a term he possibly coined.⁶¹ Equality was the most essential property of justice, he wrote in On Duties (aequitatem … est iustitiae maxime propria).⁶² It defined political life for a ‘free people,’ since only a free people could ‘enjoy equal rights before the law.’⁶³ In On the Republic, Cicero's mouthpiece Scipio distinguished republican legal equality from its democratic variant, or ‘equality [that is] itself is inequitable’ (ipsa aequabilitas est iniqua).⁶⁴ Only republican freedom was genuinely equitable because it concerned ‘that equality of legal rights of which free peoples are so fond.’⁶⁵

Where Cicero formulated the Roman theory of republican freedom, Livy projected it back onto Roman history. Jed Atkins has already highlighted this feature of Livy's account, pointing to the famous passage at the beginning of book II of Livy's History, after the expulsion of the Tarquins and the of the republic. Then ‘some young men of high birth,’ who were displeased with the new regime, conspired to bring back the monarchy. What displeased them in particular was that in a republic, law had acquired a new quality:

Now that all were equal before the law (aequato iure omnium), they missed their former licence and complained that the liberty which others enjoyed had become slavery for them [… following the expulsion of the Tarquins, ] the law was a thing, deaf and inexorable (leges rem surdam), more favourable to the weak than to the powerful, showing no indulgence or forgiveness to those who transgressed.⁶⁶

Law under the new republic created a formal equality, Livy tells us, as well as a specific kind of liberty. This liberty is one that crucially cannot be had without the law, because it relies on the law's refusal to show special favours – that is, on its impartiality and generality. The underlying idea is that having law in this sense is tantamount to having liberty. But participation in lawmaking was not a constitutive factor of republican liberty. Quite apart from the more than dubious historicity of his account, Livy, interpreted as a source of Roman political thought, conveys that a majority of Roman citizens likely had no hand in making the laws, and some of them – including the gilded youth who conspired with the Tarquins – may even have found them repellent.⁶⁷ But to borrow Rousseau's later phrase, they should still be ‘forced to be free,’ or forced to accept the same laws as others. Indeed, the political problem that this theory of liberty sought to solve was preferential treatment for the powerful, or what one might call the oligarchic threat.⁶⁸

Livy repeatedly stressed the importance of legal equality for Roman liberty throughout his History. During the second secession of the plebs, for instance, the consuls Valerius and Horatius sought to convince the plebeians not to seek retribution against the decemvirs, and instead simply to accept that ‘he is humble enough who liv[es] in the State under equal laws/rights (iure aequo), neither inflicting nor suffering injury.’⁶⁹ One of the most hated of the decemvirs, Appius Claudius Crassus, when confronted in the forum, stated in his defense that he had ‘resign[ed] his consulship in order to enact equal laws for all’ (aequandarum legum), and that this action had mostly hurt the patricians.⁷⁰ When the Latins consider requesting Roman citizenship, they, too, express this desire in terms of gaining equal rights.⁷¹ Finally, much later in the History, Livy described the public's response to the impeachment of Scipio Africanus as reaffirming this principle: many Romans argued that ‘nothing guaranteed equality of freedom (aequandae libertatis) as much as that all the most powerful people should defend themselves in court.’⁷² In almost all these examples, liberty is understood as forcing the elites to obey the same laws as the people. The spectre of the decemvir Agrippa Claudius, who wantonly claimed Virginia for himself, haunted Roman and early-modern theorists of liberty just as much as did the memory and fear of tyrants.⁷³ As Atkins concludes, ‘in Livy, aequa libertas [generally] means equality before the law rather than equal political participation.’⁷⁴

One might even say that Livy as well as Cicero are in fact giving up on anything distinctly ‘neo-Roman’ in favour of a liberal view, shared by writers of various stripes such as Hobbes or Constant, according to which it is not the ‘source of law but its extent’ that matters.⁷⁵ But we should immediately add that in Livy's and especially Cicero's view, it is not only the extent of the law that matters, but also its quality. If law has a certain quality it appears as the ultimate guarantee of liberty, and as constitutive of it. Cicero writes that only if we are all slaves to the law can we be free, and with this seeming paradox seeks to tell us something about both law and his conception of liberty.⁷⁶ We can be slaves to the law only metaphorically – the law, after all, cannot own anything. But it can command, and so if we adhere to its commands, this will set us free, Cicero claims. But what kind of law do we need to be free; and what kind of liberty is secured by law? Let us try and answer the second question first.

5. Cicero: what kind of liberty?

It is quite clear that for Cicero, liberty or freedom (libertas) has independent weight and assumes an important role in his overall political theory. In an important recent book, Michael Hawley convincingly argues that ‘Cicero is the earliest extant political philosopher to defend [an] understanding of liberty’ understood as negative liberty, as a constraint on interference or the imposition of the arbitrary will of others.⁷⁷ This is in clear contrast to Cicero's Greek predecessors, whose teleological perfectionism positively requires that negative liberty be curtailed in the interest of the perfection of virtue.⁷⁸

In On Duties, Cicero states quite clearly that it is the very essence of liberty to live as one pleases (proprium [libertatis] est sic vivere, ut velis), seemingly agreeing with the Greek democratic ideal of freedom as rendered by Aristotle.⁷⁹ What Cicero has in mind, however, is not Aristotle's polemical target, namely a democratic assembly suiting their every whim and gratifying their appetites, but rather a kind of individual, personal liberty that is an inevitable result of Cicero's academic scepticism regarding the viability of any one conception of the good life.

This scepticism emerges quite clearly from Cicero's On Ends, where he presents us with a survey of the various Hellenistic schools of philosophy and their diverging views of the highest good (summum bonum), and where at the end of the treatise we are left with an aporetic sense of being unable to decide in favour of any of these diverging views. The aporia, or lack of knowledge when it comes to the summum bonum, makes it plausible that liberty, according to Cicero, may not be curtailed in the interest of the perfection of virtue, since perfectionism cannot be given any definitive content. To be able to live as one pleases, then, appears as a moral requirement resulting from Cicero's scepticism vis-à-vis current knowledge of the highest good. In Cicero's political theory this moral requirement finds expression in the independent weight that liberty is given, a weight it had neither for Plato nor for Aristotle.

In the Republic, written in the late 50s BCE, Cicero has Scipio compare the relative advantages of monarchy vis-à-vis the other simple constitutional forms and the well-balanced, ‘tempered’ Roman republican constitution. Although Scipio expresses a preference for monarchy over the other simple forms, he identifies a crucial problem with it: it is highly unstable. This instability is due to the following defect, already briefly mentioned above: ‘The people that's ruled by a king lacks a great deal, and above all it lacks liberty, which does not consist in having a just master, but in having none.’⁸⁰

This is a remarkable passage. Cicero seems, again, to be expressing a paradox: justice requires that there be liberty for the people, so that even rule by a just king will not be sufficient for a just political order. The paradox dissolves once one realizes that it is justice understood as a personal virtue – the ‘just master’ (iustus dominus) – that won't do, since it will constrain liberty unduly. A iustus dominus can mean a just owner, or just master, but the term dominus carries the potential of arbitrary or unlawful behaviour with it, so what is meant is a person whose position in the state is constrained only by virtue, yet unconstrained by law.

However, justice, understood this time not as a personal virtue, but as a property of the political order as a whole, requires that there be liberty for the people, and there cannot be any liberty as long as there is ‘one person with permanent power, especially royal power.’ The reason for this is that Cicero is suspicious of virtue as a constraint on power, and convinced of the value of liberty understood as the absence of arbitrary rule: monarchy ‘is the most unstable because through a single person's flaw it can easily be sent headlong in the most destructive direction.’⁸¹

The same problem holds for the other simple constitutional forms.⁸² Rule by the few, even if carried out with the ‘greatest justice’ (summa iustitia, understood as personal virtue), deprives the many of liberty (libertas), which turns their condition into something similar to servitude. Rule by the people (populus) at large, however just the people may be (understood as personal virtue), will be unjust and lead to a loss of the political order – as at Athens, where the many ruled everything by plebiscite and decree and not by law, which turned the populus into a mad and arbitrary multitude.⁸³

Liberty is therefore a condition that for Cicero is constituted by law. Before we turn to the crucial issue of law's quality and extent, we should remind ourselves why it is that Cicero is willing to give liberty such a prominent role in the architecture of his political theory. As Hawley asks pointedly, ‘How is it that Cicero's apparently teleological natural law doctrine could yield a politics in which citizens are largely free to pursue whatever form of life they prefer? How is it that natural law would tolerate such liberty?’ Hawley argues that although Cicero, no less than Plato or Aristotle, is a moral realist and especially a realist about natural law, Cicero ‘does not interpret natural law as foreclosing this negative version of liberty as they do; his natural law, in fact, seems to demand it.’⁸⁴

For an answer, Hawley turns to Cicero's late work of moral philosophy On Duties, where Cicero develops an extremely interesting and influential anthropology. By nature, he says, humans are given two roles (personae): first, the whole species is united by participation in reason, which is why we have epistemic access to the natural law and the rights and duties it bestows. Second, we are given an individual nature, which explains why there are vast differences in bodies and even greater variety in people's characters. This leads to a picture where moral rights and obligations have universal validity, but where at the same time, as Hawley convincingly argues, ‘the development of each individual's nature and choice of life path is up to the discernment of that individual alone and cannot be directed by someone else.’⁸⁵

This account of the independent value libertas acquired in Cicero's political thought is persuasive and consistent with other passages in Cicero's work. The idea that ‘nothing can be sweeter’ than liberty, although uttered by Scipio in the Republic as part of a report of the democratic view, was also held by Cicero himself.⁸⁶ Cicero's proto-Kantian anthropology as put forward in On Duties helps explain how a moral and political universalism of right can go hand in hand with scepticism about a universal account of the good. As we have noticed above, the same kind of scepticism emerges from Cicero's On Ends, where we were left with an aporetic sense of being unable to decide in favour of any of the Hellenistic accounts of the highest good.

This forms a stark contrast with Cicero's dogmatic account of law, especially natural law, to which we turn next. Indeed, the contrast between Cicero's scepticism regarding competing accounts of the highest good, on the one hand, and his affirmative theory of law, on the other, is perhaps best captured in Cicero's own attempt to build a value theory that is largely independent of virtue but now accords law itself the status of a good, something completely outside the inherited Hellenistic views and pointing toward an interesting value theory. Law, Cicero claims in the Laws, must not only be numbered among the goods (numerandum in bonis), it must even be considered one of the greatest goods, and it is of intrinsic, not merely instrumental, value (per se igitur ius est expetendum).⁸⁷

6. Cicero: what kind of law?

Hawley's explanation of the prominence of libertas for Cicero is sound, but it might not pay sufficient attention to a key feature of Cicero's account of liberty: that liberty, properly understood, is constituted by law. We have seen that for Cicero, even having a master who is virtuous and just deprives the people of liberty is by virtue of this deprivation unjust. How is it, then, that we are free if and only if we are ‘all slaves to the law’ (legum … omnes servi)?⁸⁸ It is because the law is necessarily, qua law, universal and therefore not arbitrary in the way a master, even a virtuous one, is. It is of the very essence of the law (vis legis), Cicero thinks, that it is perfectly general (scitum et iussum in omnis) and binds all alike.⁸⁹

This is part of his conceptual claim that law cannot be directed at individuals and that bills of attainder (privilegia) therefore cannot be law – indeed, nothing could be more unjust than a bill of attainder.⁹⁰ The generality of the law conditions the law's substance: in his civil-law speech For Caecina, Cicero reasons that there are things ‘contrary to law which the Roman People is unable to command or to prohibit.’⁹¹ This places limits on the content of the law. Cicero goes on to ask the judges:

But I ask of you whether you think, if the people ordered me to be your slave, or, on the other hand, you to be mine, that that order would be authoritative and valid? You see and admit that such an order is worthless. Hereby you first allow this, that it does not follow that whatever the people orders ought to be ratified.⁹²

The very concept of law, Cicero thinks, contains a normative claim to justice and right and could be said in proto-Fullerian fashion to contain an internal morality.⁹³ Magistrates are but mouthpieces of the law, Cicero writes, and the laws are prior to them and in charge of them, which means that there has to be a right of appeal (provocatio) against their powers of office.⁹⁴ This right of appeal, alongside law's generality and the prohibition of bills of attainder, is considered an essential mark of liberty: ‘this is the essential trait of a free state (proprium liberae civitatis): that nothing can be taken away from the status (de capite) or the property (de bonis) of a citizen without trial (sine iudicio).’⁹⁵

Law-making, to result in proper law, needs to be conditioned by the internal normativity of law, by the formal features just mentioned: generality, i.e. equality of law, and due process. If law (lex) is to bind the citizens – if it is to be the bond of the state and create liberty – then the underlying conditions of legality (ius) have to guarantee the equality of the law.⁹⁶ This equality is built into the very concept of legality, and forms the inner morality of law: otherwise, it would not be law.⁹⁷ Historically, Cicero claims, this kind of fair even-handedness or impartiality (aequabilitas) was meted out by just rulers,⁹⁸ but when this no longer worked, laws (leges) were invented to speak to everyone (cum omnibus) always with one and the same voice (una atque eadem voce). It is this specific kind of law, lex conditioned by ius, that is needed to create legal order and thus liberty.⁹⁹ ‘Liberty,’ Cicero writes in a speech, giving his idea the most succinct expression, ‘consists in the laws’ (consistit libertas in legibus).¹⁰⁰ Liberty is created by the laws, then, and both are required by justice.¹⁰¹ Who makes these laws, on the other hand, and the extent of the franchise, are not strictly speaking issues to do with liberty, but matters of justice.

This results in a picture where liberty cannot exist without the law; non-legal liberty, or liberty from the law, is liberty of an entirely unattractive and precarious character, for it is only in a legal order that we can enjoy equal liberty: without general law, which is the same for everyone, there is no liberty and no yardstick of equality to be had.¹⁰² Hobbes was to seize on Cicero's idea in Leviathan, where he wrote: ‘[E]ven Cicero, (a passionate defender of Liberty,) in a publique pleading, attributeth all Propriety to the Law Civil.’¹⁰³ Hobbes correctly saw that Cicero thought liberty as well as property depended on government guarantees, given the lack of security in the state of nature. Not living under legal liberty meant for Cicero, no less than for Hobbes, a return to the state of nature.¹⁰⁴

Note that so far, there has not been any mention of political rights and their connection with liberty. Having law, properly qualified, secures ‘legal liberty,’ the only kind of liberty worth having on Cicero's account – but what about political participation? Is not having the franchise part of liberty, or at least a necessary condition to securing liberty? While it is clear that not having a permanent ruler with absolute power – not having a master, not even a just one – is a requirement for there to be liberty, the extent to which liberty depends, on Cicero's view, on an extensive franchise, or on voting rights at all, is far less clear.

It is true that Cicero at one point calls the (secret) ballot a ‘defender of liberty’ (vindex libertatis) for the people (populus),¹⁰⁵ yet he is of course very much in favour of the timocratic structure of the assembly of the centuries (comitia centuriata) and the heavy dilution of the votes of the poor.¹⁰⁶ Cicero, after dubbing the ballot a defender or protector of liberty, goes on to say that under his own proposal for a voting law in the Laws it is merely a species libertatis that will be given to the people.¹⁰⁷ It is not clear how this should be understood; it could mean simply a particular kind of liberty, or, more cynically, merely the appearance of liberty. While it is clear that Cicero thinks that every adult male citizen should have the right to elect magistrates and to legislate (ius suffragii), as a matter of justice, the connection with liberty is generally subdued or even absent: it is consistent with liberty that the people decide little (pauca per populum) and that what the popular assembly decides requires ratification in the senate.¹⁰⁸

The following seems the most natural interpretation of Cicero's overall view: participating in the political process, having political rights, is a matter of right and part of what justice requires. Justice also requires that there be a certain amount of liberty in the state, but there does not seem to be a conceptual connection between the franchise and liberty – it is simply that justice requires that there be both. In sum, it seems fair to say that liberty for Cicero consists in legal liberty, in having a legal order subject to the inherent morality of law; it is inconsistent with this kind of liberty to have a permanent master, which is why legal liberty requires, instrumentally, that magistrates change every year (term limits), that they not rule alone (collegiality) and that there be ways to appeal their powers (provocatio).¹⁰⁹

Livy, in a passage which is certainly not historical but should be interpreted as the expression of the political thought of Livy's own time, provides us with a very similar conception of liberty. Describing Brutus's actions after the expulsion of the kings, Livy says that he is writing the history ‘of a Rome henceforth free,’ with ‘the authority of her laws more powerful than that of men.’¹¹⁰ The origin of this newly gained freedom is said by Livy to lie in the term limits of the consular office, ‘because the consular power was limited to one year.’¹¹¹ Liberty here resides, as it does in Cicero, in a constitutional feature, the term limits, which are essential to legal liberty.

7. Towards a new history of liberalism: Machiavelli, Rousseau (and Hobbes)

If we have ‘legal liberty’ – i.e. the right kind of law that is constitutive of civil liberty – and therefore have libertas, then we have reached the minimal threshold for a just political order, a res publica in the sense of Cicero's definition of the legitimate and stable state.¹¹² The dependence between constitutional form and liberty runs the opposite way than is usually assumed by today's ‘republican’ political theorists: while these theorists believe that republican self-government implies certain rights, for Cicero the dependence runs the other way around. The legitimacy of a res publica depends on there being legal liberty in the first place, and it is only when this condition is met that we can speak of a proper political order.

This identification of liberty with equality before the law ultimately found its most powerful expression among nineteenth-century liberals. Indeed, A.V. Dicey, the philosopher who theorized the ‘rule of law,’ defined liberty in precisely such terms. In Britain, he claimed, the rule of law meant ‘not only that with us no man is above the law’ – i.e. no individual was legibus solutus – ‘but … here every man, whatever be his rank or condition, is subject to the ordinary law of the realm and amenable to the jurisdiction of the ordinary tribunals.’ The law must be the same for all, and it was in England that ‘the idea of legal equality, or of the universal subjection of all classes … has been pushed to its utmost limit.’ It was only by barring any instances of ‘discretionary authority in the matter of arrest, of temporary imprisonment, of expulsion from the territory’ that the condition of ‘legal freedom’ prevailed. Inclusion in the electoral franchise was not what determined freedom; Dicey opposed the Reform Act of 1867. Freedom was a legal matter, and reflected a certain quality of the laws – that they be equal for all.¹¹³

Dicey's formulation highlights a conceptual confusion at the heart of the theory of freedom as non-domination. Liberty may be incompatible with domination, but it is fully compatible with ‘universal subjection.’ Or to frame this distinction in the Roman law terms that Skinner and Pettit appealed to: there is a crucial difference between dominium and imperium. Free born Romans were not the property (dominium) of the emperor, like slaves were a form of property; but they did remain under the command (imperium) of imperial power.¹¹⁴ The subjects of any regime – monarchic, aristocratic, or democratic – would always be ‘subjected’ to the commands of the state. Freedom, in fact, depended on our universal subjection to these commands; the alternative was an inequitable regime, in which our freedom was held hostage by individuals who threw off the yoke of subjection (the oligarchic or tyrannical threat).

It is precisely this understanding of liberty as legal equality that we can identify among ‘republican’ authors whose accounts of liberty have frequently been analyzed in terms of non-domination. To be sure, these authors proposed theories that could exhibit substantial differences from one another, and often strayed from the full Ciceronian package. But as conceptual overlaps go, we hold that their conceptions of freedom were closer to the Roman idea of legal liberty than to the neo-Roman idea of non-domination.¹¹⁵ For the sake of brevity, we will focus here on two such ‘republican’ authors, Machiavelli and Rousseau, while drawing in Hobbes – whom no-one would accuse of republican sympathies – as a helpful point of comparison.

In his Discourses on Livy, Machiavelli follows the eponymous Roman historian in defining freedom as equality before the law. Drawing on Livy's famous account of the episode, Machiavelli defends Brutus's execution of his sons,

who, as history shows, had conspired with other Roman youths simply because under the Consuls they could not have the same extraordinary advantages they had enjoyed under the kings; so that the liberty of the people seemed to have become their bondage.¹¹⁶

Following this logic, a state where the people retained their liberty is one where no-one enjoyed any ‘extraordinary advantages’ or privileges, i.e. where everyone was on the same level of legal equality.¹¹⁷ Machiavelli singled out for particular criticism those moments in Livy's history when powerful and honourable patricians were pardoned for their crimes: ‘in a well ordered state a man's merits should not compensate for his crimes,’ he insists apropos of Horatius Cocles.¹¹⁸ Failure to punish the most illustrious citizens would result in the loss of political liberty: ‘a state that properly observes this principle will long enjoy its liberty; otherwise, it will speedily come to ruin.’¹¹⁹

This particular concern with the unequal application of the law reflects the historical conditions of late medieval and Renaissance Florence. The overarching goal of the successive popolo governments in Florence had been to rein in the magnates (i magni), notably through ostracism, bans on political participation, and the creation of the gonfaloniere di giustizia, an official who represented the rule of justice in the city.¹²⁰ To prevent the lawless behaviour of the magnates required domination, or, more precisely, subjection: hence, the name of executives in the Republic, i signori (‘the lords’). Freedom in republican Florence was not conceived as the absence of subjection, but rather as the equal submission to the domination of justice.¹²¹ Or as the chancellor of Florence, Coluccio Salutati, phrased it: ‘To abide by the law in order to preserve one's freedom can be hard: it can even look like a kind of slavery.’¹²² Salutati was perhaps paraphrasing Cicero's statement that ‘we are all slaves (servi) of the law so that we may be free.’¹²³

But where Cicero had paid particular attention to the formal qualities requisite for legal liberty, Machiavelli focused instead on the execution of the law. Indeed, freedom can be lost in two ways, either through the introduction of privileges and bills of attainder (formal violations of legal liberty), or through the uneven application of the law (practical violations). Machiavelli also departed from Cicero the theorist in accepting that there were exceptional times when it might be necessary to violate the law, a point that Cicero the opportunist statesman had acknowledged in practice, all the while denouncing it as a theorist in his mature works of political theory.¹²⁴ All the same, when the salus populi was not in jeopardy, Machiavelli believed that legal uniformity was constitutive of liberty.

There was, however, an important exception to this rule, which explains why it can prove confusing to pin down Machiavelli's theory of freedom. One of the reasons why Roman patricians (like the Florentine grandi) posed the greatest threat to legal liberty was because they clung to an alternative vision: this small elite ‘wish[ed] to be free for the purpose of commanding,’ he writes. Here was an aristocratic concept of freedom, an ideal that could produce greatness (grandezza), which Machiavelli also admires and which is antithetical to the idea of legal liberty outlined above. But this kind of aristocratic freedom was an ideal that posed severe risks for the many ‘who constitute an immense majority.’ The popolo only ‘desire liberty so as to be able to live in greater security.’¹²⁵ With respect to the people, then, Machiavelli made a ‘liberal’ case that freedom was a matter of individual (and family) security, not active participation in government. In ‘a free government,’ he writes in a famous passage, one should ‘be able freely to enjoy one's own without apprehension, to have nothing to fear for the honor of his wife and daughters, or for himself.’¹²⁶ Machiavelli is not far away here from Judith Shklar's definition of ‘liberalism of fear,’ whose purpose, as she defined it, was ‘to secure the political conditions that are necessary for the exercise of personal freedom.’¹²⁷

For Machiavelli, participation in lawmaking played a minimal role in the production of liberty and is not conceptually connected with it. Like Cicero (and more particularly Polybius), Machiavelli also rejected pure forms of government in favour of a balanced constitution.¹²⁸ Both grandi and popolo had a place in the republic: the balancing act required for its preservation entailed managing the ambition of the former in order to protect the freedom of the latter. Typically, this process entailed some degree of popular involvement in the administration of the state, though importantly for Machiavelli, popular participation was not a necessary condition for either the establishment or preservation of freedom. He acknowledged the possibility that an aristocratic government (such as that of Sparta or Venice) could also serve to safeguard liberty, but simply argued that in his view a more popular (but not a democratic) government was a better guardia della libertà.¹²⁹ Again, this opinion aligns with Cicero's claim that the res publica must be a res populi, though not a democracy. In any case, popular participation was not necessarily about lawmaking at all. As Aristotle before him, Machiavelli took a negative view of frequent legislative action: ‘changed laws did not suffice to keep men good (buoni).’¹³⁰

This account of Machiavelli, according to which liberty is not premised on republican government, does not mean that there is no daylight between his position and Hobbes's famous claim that ‘Whether a Common-wealth be Monarchical, or Popular, the Freedome is still the same.’¹³¹ But it may well mean that when it comes to their conception of liberty, narrowly conceived and independent from their broader theories of value and politics, the paradigmatic republican Machiavelli and the proto-liberal Hobbes have more in common than is usually thought.¹³² The citizens of Lucca might have had good reason to inscribe LIBERTAS on their turrets, but neither Machiavelli nor Hobbes would have claimed that this was for reasons specific to Lucca's form of government (recall aristocratic Venice). And while Hobbes is often associated with a view that men rule, not laws, we must not forget that Hobbes ‘does place law-making at or near the heart of his account of political authority,’ and that this entails that for Hobbes, the sovereign must rule ‘through law: the political realm acts through the legal realm.’¹³³

The argument is that for Hobbes, the sovereign's will can be identified as such ‘only if it meets certain general and publicly recognisable criteria,’¹³⁴ and this means, according to David Dyzenhaus, that ‘all acts of sovereignty must comply with the law to be recognisable as acts of sovereignty.’¹³⁵ This has consequences for Hobbes's ‘true liberty of a subject,’ or ‘civil liberty,’ as he calls it, in that there are things, ‘though commanded by the Soveraign,’ the subject ‘may nevertheless, without Injustice, refuse to do.’ Famously, Hobbesian subjects are always at liberty to defend themselves, and may under certain circumstances disobey when their sovereign enlists them as soldiers. Subjects are also entitled to rules of due process and therefore at liberty to resist incriminating themselves.¹³⁶ As opposed to the caricature of Hobbes as a thinker of unconstrained absolutism, this yields a Hobbes closely aligned with the idea of legal liberty, where ‘civil liberty’ is not simply liberty from the law, as natural liberty in the state of nature is; rather, Hobbes’ civil liberty means a specific quality of liberty which is a condition of political obligation and achieved by the bonds of law.¹³⁷ And although Hobbes may be said to be far more of a specifically legal thinker than Machiavelli, they can both very broadly be interpreted as upholding a concept of legal liberty as described in our article.

It is therefore not the case that for Machiavelli ‘the people's desire for freedom … is so truncated in scope that it can be satisfied by a tyrant who provides for their security.’¹³⁸ The one example of a monarchic state that successfully preserves freedom in the Discourses is France, where ‘the king there has bound (obligati) himself by countless laws that provide for the security of all his people.’¹³⁹ The French king, as sovereign, has the power to make and unmake laws, but is still constrained by legal obligations, including contracts. A sovereign who did not rule through law, or who ruled in a seigneurial or despotic fashion (in Bodin's sense), and who was therefore unconstrained by legality – a sovereign, in other words, who could act as a constituent power doing things without legal warrant – was exactly the hole in the fabric of the state from which the entire condition of liberty could unravel.¹⁴⁰

It is for this very reason that Rousseau, in the ‘Dedication’ to his second Discourse, defined liberty in opposition to a regime where a single individual was princeps legibus solutus.¹⁴¹ Setting up his thought experiment about the ideal state (which conveniently matches Geneva), Rousseau wrote: ‘I would have chosen to live and die free, that is, so subject to the laws that neither I nor anyone else could throw off their honorable yoke.’¹⁴² He develops this point further later on: ‘I would not have wanted anyone in the state to claim they were above the law […] For regardless of a government's constitution, if one person is not subject to the law, everyone else is necessarily at their mercy.’¹⁴³ Here as well, we can observe how the particular form of government for Rousseau was not constitutive of freedom, contrary to what his republican interpreters have argued.¹⁴⁴ Rather it is the absence of any individual who might be ‘above the law’ – that is, whose legal rights are unequal or superior to those of others – that defines freedom. In the Social Contract, Rousseau of course insisted that sovereignty must always rest with the people, and thus law-making, as an exercise of sovereignty, could not be delegated to a government by the few or the one. But popular sovereignty is not what gives us liberty.¹⁴⁵ If a single faction controls the assembly, it will no longer express the general will; the state will then seek ‘a master, not a liberator,’ and the people will lose their freedom.¹⁴⁶

So what exactly constitutes political liberty for Rousseau? It is by undergoing the transformation brought about by a social contract that humans can enjoy political liberty. The social contract achieves this alchemical miracle of turning natural into political liberty by apportioning equal rights to all members: since ‘everyone giv[es] themselves entirely, the condition is equal for all.’¹⁴⁷ This equality is first and foremost legal (as opposed to, say, material or physical); it is, essentially, the aequabilitas iuris that Cicero had identified with free peoples. In Rousseau's formulation, all citizens should enjoy ‘such an equality’ that they ‘all enjoy the same rights.’¹⁴⁸ When we are all equal before the law then we are free.

It might seem excessively revisionist to argue that liberty in Rousseau depends on formal features of the law, rather than on popular sovereignty. To some degree, there is an unresolved tension, identified by Ernst Cassirer, between voluntarism and rationalism in Rousseau's thought.¹⁴⁹ But while Rousseau, unlike Machiavelli, rejects the possibility of an aristocratic guarantor of liberty, popular sovereignty in and of itself is clearly an insufficient condition for liberty, as his own criticism of Athenian democracy (like Cicero's) also indicates.¹⁵⁰ What's more, in the idealized rural societies that Rousseau imagines in book 4 of the Social Contract, the popular sovereign nearly withers away: ‘the first [citizen] who proposes a law merely expresses what all others have already felt.’¹⁵¹ Here Rousseau is closer to a virtue theorist than a voluntarist. Even admitting for more ambiguous passages, participation in lawmaking is clearly an insufficient condition for political freedom in Rousseau. An entire half of the population is in fact barred from lawmaking, but nonetheless deemed free by Rousseau: women.¹⁵² In his description of the Genevan citoyennes (note the term!), they are depicted as the ‘guarantors’ of the ‘august liberty’ of the Republic. They owe this title to their role in perpetuating ‘a love of laws in the state and harmony among citizens.’¹⁵³ Here as well, the preservation of liberty is predicated on the rule of law, not its expression.

8. Conclusion

These excessively brief case studies hopefully suggest that the history of political liberty is not so much a story about discontinuity (Constant and Berlin) or about disappearance (Skinner and Pettit), but about myopia: we have failed to see that our ‘modern’ ideas of liberty are not as modern as they look.¹⁵⁴ Recognizing the classical and early-modern arguments for rule-of-law liberalism also encourages us to question the utility of dividing the political canon into ‘liberal’ and ‘republican’ camps.¹⁵⁵ And it challenges the standard narratives of political liberalism, most of which seek an origin point in the early-modern period, either with the wars of religion, natural law philosophers, or free-market economists.¹⁵⁶ In fact, the pursuit of ‘equal justice under law’ (as inscribed on the U.S. Supreme Court building) has been a hallmark feature of Western political thought since antiquity.

To be clear, it is not our contention that liberalism without democratic government and popular sovereignty is desirable. Normatively, we share a basic commitment to liberal democracy. Historically, however, we believe that the philosophical and political foundations of liberalism have been overlooked and misrepresented.¹⁵⁷ Where contemporary theorists of neo-Roman or republican liberty seek its basis in politics (direct participation in self-government), we identify an older and arguably more influential tradition that locates it in law. This tradition severs the tight conceptual connection between voluntaristic participation and liberty that republicans typically uphold, and instead emphasizes the rationalist underpinnings of law-making. If neo-republican theorists have missed this tradition, it is largely because law tout court is an insufficient condition for liberty; there is a structural requirement for law to function in this liberal way, which is that all must be equal before it. The law ‘must be the same for everyone, whether it protects, or whether it punishes,’ affirms the Declaration of the Rights of Man and of the Citizen (art. 6). It is the history of this fundamental principle of the rule of law that we have sought to excavate here.

Acknowledgement

We would like to thank audiences at the University of Zurich and Cambridge University for criticism and suggestions, especially Carol Atack, Chris Brooke, John Dunn, Jeff Dymond, Peter Garnsey, Signy Gutnick Allen, Andreas Gyr, Nikolas Hächler, Adam Lebovitz, Amira Moeding, Enrico Piergiacomi, Malcolm Schofield, Max Skjönsberg, Michael Sonenscher, Sylvana Tomaselli and Adam Woodhouse. Many thanks to Lars Vinx for his insightful comments and to two anonymous reviewers.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Benjamin Straumann's work on this article received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 864309).

Notes

1 Benjamin Constant, “The Liberty of Ancients Compared with that of Moderns” (1819), in Political Writings, ed. and trans. Biancamaria Fontana (Cambridge: Cambridge University Press, 1988), 309–28; for Hobbes's comment on Lucca, see Leviathan, ed. Noel Malcolm, vol. 4 of The Clarendon Edition of the Works of Thomas Hobbes (Oxford: Oxford University Press, 2012), 2.21; p. 332.

2 Berlin, “Two Concepts of Liberty” (1958), in Liberty: Incorporating Four Essays on Liberty, ed. H. Hardy (Oxford: Oxford University Press, 2002); for the quote, see Montesquieu, The Spirit of the Laws, ed. Anne M. Cohler, Basia C. Miller, and Harold Stone (Cambridge: Cambridge University Press, 1989), 11.2 (p. 155).

3 See e.g. Chaim Wirszubski, Libertas as a Political Idea at Rome during the Late Republic and Early Principate (Cambridge: Cambridge University Press, 1950); Kurt Raaflaub, The Discovery of Freedom in Ancient Greece, trans. Renate Franciscono (Chicago: University of Chicago Press, 2004); Naomi Campa, “Positive Freedom and the Citizen in Athens,” POLIS 35 (2018): 1–32; Valentina Arena, Libertas and the Practice of Politics in the Late Roman Republic (Cambridge: Cambridge University Press, 2013); Jed Atkins, Roman Political Thought (Cambridge: Cambridge University Press, 2018). See also Michelle T. Clarke, “Doing Violence to the Roman Idea of Liberty? Freedom as Bodily Integrity in Roman Political Thought,” History of Political Thought 35, no. 2 (2014): 211–33, an account of libertas as bodily integrity and freedom from coercion that is compatible with ours.

4 Quentin Skinner, Liberty Before Liberalism (Cambridge: Cambridge University Press, 1998); Philip Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Oxford University Press, 1997); on its extension into Roman political thought, see Arena, Libertas, and Atkins, Roman Political Thought; for Rousseau, see Annelien de Dijn, “Rousseau and Republicanism,” Political Theory 46, no. 1 (2018): 59–80.

5 On the reception of Aristotle's Politics (as we will see, a critical source for this democratic theory of freedom as non-domination), see notably Sophie Smith, “The Language of ‘Political Science’ in Early Modern Europe,” Journal of the History of Ideas 80, no. 2 (2019): 203–26.

6 Montesquieu, Spirit of Laws, 11.3 (p. 155).

7 We do not discuss Sallust here because virtue is far more central to his analysis of liberty than it is for Cicero or Livy. While for Sallust, law and equal rights are necessary conditions for liberty as well, they are never sufficient; virtue, especially military virtue, is necessary as well and cannot be guaranteed by law. See William Walker, “Sallust and Skinner on Civil Liberty,” European Journal of Political Theory 5, no. 3 (2006): 237–59; Benjamin Straumann, Crisis and Constitutionalism: Roman Political Thought from the Fall of the Republic to the Age of Revolution (Oxford: Oxford University Press, 2016), 261f.

8 Dom. 33.

9 Dom. 77–80 for the limits imposed by legal liberty on the power (potestas) of the Roman people, which is said (80) to be most extensive (maxima) in all other regards.

10 See Straumann, “Justice and Republicanism,” in F. Lovett, M. Sellers (eds.), The Oxford Handbook of Republicanism (Oxford: Oxford University Press, forthcoming); and below.

11 For an insightful discussion of the instability of neo-Roman or republican liberty, and its tendency to collapse either into liberal negative liberty or participatory positive liberty, see D. Dyzenhaus, “Critical Notice of On the People's Terms: A Republican Theory and Model of Democracy, by Philip Pettit,” Canadian Journal of Philosophy 43, no. 4 (2013): 494–513.

12 Skinner, Liberty Before Liberalism, 41; Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Oxford University Press, 1997), 31–2. See also Hannah Dawson and Annelien de Dijn, “Introduction,” in Rethinking Liberty Before Liberalism, ed. Dawson and de Dijn (Cambridge: Cambridge University Press, 2022).

13 Politics, 6.1317b13, trans. modified.

14 Politics, 5.1310a29-34 (Rackham trans.); see Campa, “Positive Freedom.”

15 Athenian Constitution, 2.2–6.1. More generally, see Josiah Ober, Mass and Elite in Democratic Athens: Rhetoric, Ideology, and the Power of the People (Princeton: Princeton University Press, 1989), 61–6; P. Brook Manville, The Origins of Citizenship in Ancient Athens (Princeton: Princeton University Press, 1990), 5; Mogens Hansen, “Democratic Freedom and the Concept of Freedom in Plato and Aristotle,” Greek, Roman and Byzantine Studies 50, no.1 (2010): 1–27; Campa, “Positive Freedom”; Annelien de Dijn, Freedom: An Unruly History (Cambridge: Harvard University Press, 2020), chp. 1; René de Nicolay, “The Birth of Unlawful Freedom in Plato's Laws 3,” Polis 38 (2021): 494–511.

16 Politics, 3.1279a21-22.

17 Politics, 5.1310a28-34.

18 Herodotus, Histories, 3.31.4 (for Cambyses) and 3.80.3 (Otanes, who equates the “monarch” with a tyrant at 3.80.4).

19 Sophocles, Antigone, l. 506–7.

20 Politics, 5.1310a35.

21 Politics, 5.1301a20.

22 Politics, 4.1292a20, 27, 34. See Kinch Hoekstra, “Athenian Democracy and Popular Tyranny,” in Popular Sovereignty, and Ivan Jordovic, “Aristotle on Extreme Tyranny and Extreme Democracy,” Historia 60, no. 1 (2011): 36–64. See also Federica Carugati, Creating a Constitution: Law, Democracy, and Growth in Ancient Athens (Princeton: Princeton University Press, 2019).

23 Xenophon, Hellenica, 1.7.12; see also Memorabilia, 1.1.18. See Dustin Gish, “Defending Demokratia: Athenian Justice and the Trial of the Arginusae Generals in Xenophon's Hellenica,” in Xenophon: Ethical Principles and Historical Enquiry, ed. Fiona Hobden and Christopher Tuplin (Leiden: Brill, 2012), 161–212.

24 Hellenica, 1.7.15 (Socrates); 25–26 (Euryptolemus); see also Plato, Apology, 32b, where Socrates declares (apropos this precise episode) that he could never do anything parà toùs nómous.

25 Plato, Crito, 51c, 53c.

26 Plato, Phaedo, 115a.

27 See Brian Tierney, “'The Prince Is Not Bound By the Laws’: Accursius and the Origins of the Modern State,” Comparative Studies in Society and History 5 (1963): 378–400.

28 Politics, 2.1269a21-24.

29 Thucydides, Peloponnesian War, 2.37.1-2. See Josiah Ober, Political Dissent in Democratic Athens: Intellectual Critics of Popular Rule (Princeton: Princeton University Press, 2008); more generally, W. R. Connor, Thucydides (Princeton: Princeton University Press, 1984).

30 Peloponnesian War, 2.65.10.

31 Peloponnesian War, 6.15.3; see also Plutarch, “Life of Alcibiades,” 16.2-3, 17.2

32 Peloponnesian War, 6.89.5.

33 Isocrates, Panathenaicus, 131; Pseudo-Xenophon / “Old Oligarch,” Athenian Constitution, 1.5, 10.

34 Republic, 8.557b, 8.561d.

35 Republic, 3.404e.

36 On akolasía as a disease of the soul, see also Republic, 10.609b-c. See also Fred D. Miller, Jr., “Platonic Freedom,” The Oxford Handbook of Freedom, ed. David Schmidtz and Carmen E. Pavel (Oxford: Oxford University Press, 2016).

37 See e.g. Apology, 26e9; Gorgias, 508a; and Republic, 2.364a3, 3.403a10, 4.444b7, and passim.

38 Politics, 3.1277a.

39 Politics, 5.1301b.

40 Herodotus, Histories, 3.80.6. See also Gregory Vlastos, “Isonomia,” American Journal of Philology 74, no. 4 (1953): 337–66; Ober, Mass and Elite, 74–5; Gerald Stourzh, Modern Isonomy Democratic Participation and Human Rights Protection as a System of Equal Rights, trans. Cynthia Peck-Kubaczek (Chicago: University of Chicago Press, 2021).

41 Politics, 5.1301b.

42 Republic, 8.558c. Cicero would pick up on the same critique in De re publica: see below.

43 See Bernard Manin, The Principles of Representative Government (Cambridge: Cambridge University Press, 1997).

44 Nicomachean Ethics, 5.1131b20.

45 Peloponnesian War, 2.37.1.

46 Nicomachean Ethics, 5.1132b30-1133a1 (“The very existence of the state depends on proportionate reciprocity; for men demand that they shall be able to requite evil with evil— if they cannot, they feel they are in the position of slaves … .”). Hobbes similarly insisted that “Justice be equally administered to all degrees of People,” where criminal law was concerned: see Leviathan, 2.30 (p. 534). We return to Hobbes below.

47 Republic, 3.404e-5c.

48 Polybius, Histories, 6.57.9.

49 Histories, 6.44.

50 Histories, 6.43.5. On Polybius's political theory, see Straumann, “Leaving the State of Nature: Polybius on Resentment and the Emergence of Morals and Political Order,” Polis 37 (2020): 9–43.

51 Cicero, De re publica, 3.23.7. On licentia in Cicero, see René de Nicolay, “Licentia: Cicero on the Suicide of Political Communities,” Classical Philology 116, no. 4 (2021): 537–62.

52 Cicero, De re publica, 1.44.17; see also 1.66. More generally, see Malcolm Schofield, Cicero (Oxford: Oxford University Press 2021), 40ff.

53 See e.g. Tacitus on how “really great and famous oratory is a foster-child of licence, which foolish men called liberty (licentiae, quam stulti libertatem vocitant),” and was in fact “an associate of sedition, a goad for the unbridled populace,” Dialogue on Oratory, 40.2-3; in Tacitus, Agricola, Germania, Dialogue on Oratory, trans. M. Hutton, W. Peterson, et al. (Cambridge: Harvard University Press/Loeb, 1914), 342–3.

54 De re publica, 2.43.

55 See for example the plebeians’ distrust of Cassius's first proposed agrarian law: “those lands would bring servitude [servitutem] to the men who should receive them, and were being made a road to monarchy,” Loeb trans., 2.41 (353). See also Schofield, Cicero, 29.

56 Arena, Libertas, 9.

57 Atkins, Roman Political Thought, 40–1.

58 De re publica, 2.54.

59 De re publica, 1.48. This is from the presentation of democracy, but elements of it are retained in the constitutional law-governed state Cicero holds up as the best and most stable: De re publica, 2.57. In De officiis, Cicero suggests that the first Roman kings were similarly chosen because “they managed by establishing equitable conditions to hold the higher and the lower classes in an equality of right” (aequitate constituenda summos cum infimis pari iure retinebat), 2.41 (Perseus trans).

60 De re publica, 1.49 (Loeb 75); Atkins, Roman Political Thought, 52–3. More generally, see Dan Edelstein and Benjamin Straumann, “Roman Rights Talk: Subjective Rights in Cicero and Livy,” History of Political Thought 43.4 (2022): 637–59.

61 See Elaine Fantham, “Aequabilitas in Cicero's Political Theory, and the Greek Tradition of Proportional Justice,” Classical Quarterly 23.2 (1973): 285–90. See also Henrik Mouritsen, Politics in the Roman Republic (Cambridge: Cambridge University Press, 2017), 14.

62 Cicero, De officiis, 1.64 (Loeb 66).

63 Cicero, De officiis, 1.88 (Loeb 88). [In liberis vero populis et in iuris aequabilitate exercenda]

64 De re publica, 1.43. This comes up again in 1.49: “For if we cannot agree to equalize men's wealth, and equality of innate ability is impossible, the legal rights at least of those who are citizens of the same commonwealth ought to be the same … ” See Fantham, “Aequabilitas,” 285.

65 De re publica, 1.53 (Loeb 81). [aequabilitas quidem iuris, quam amplexantur liberi populi].

66 History of Rome, 2.3.3, trans. rev. Canon Roberts, Perseus ed; discussed in Atkins, Roman Political Thought, 47; see also Skinner Liberty Before Liberalism, 44–5.

67 See Daniel Kapust,“Skinner, Pettit and Livy: The Conflict of the Orders and the Ambiguity of Republican Liberty,” History of Political Thought 25, no. 3 (2004): 377–401.

68 See Wirszubski, Libertas, 17.

69 History of Rome, 3.53.9; Atkins, Roman Political Thought, 49.

70 History of Rome, 3.56.9; Atkins, Roman Political Thought, 48.

71 History of Rome, 8.4.3-5; 8.5.4; and 8.14.3-11.

72 History of Rome, 38.50.8 (Loeb 171); qtd. Atkins, Roman Political Thought, 47.

73 See e.g. Machiavelli, Discorsi, esp. 1.40.

74 Atkins, Roman Political Thought, 51.

75 Skinner, Liberty before Liberalism, 85. See Hobbes, Leviathan, 2. 21 (p. 332). Interestingly, this converges with a view held by many Romans with regard to the early principate under Augustus, where Augustus’ prerogative “was wide, but constitutional and limited” and where “the essential rights and liberties of Roman citizens remained untouched,” allowing them to live “again under a system of law and order which safeguarded their rights,” in Wirszubski, Libertas, 122.

76 Clu. 146.

77 Hawley does not think that the neo-Roman distinction between republican and liberal liberty is stable, and accordingly does not think that liberal freedom is compatible with having a benevolent master: Hawley, Natural Law Republicanism, 178. We also harbour doubts regarding the stability of that distinction, but need not take a stance on this well-worn issue here.

78 Hawley, Natural Law Republicanism, 34; 34–38.

79 Off. 1.70.

80 Rep. 2.43, trans. Zetzel.

81 Rep. 2.43.

82 Rep. 1.43

83 Rep. 2.44.

84 Hawley, Natural Law Republicanism, 34, 37.

85 Hawley, Natural Law Republicanism, 37–40; and Off. 1.107–120; cf. 1.13. Note that this is not necessarily inconsistent with perfectionism, but it is inconsistent with paternalism. It may well be that we simply have to find out more about our individual natures in order to perfect them. Such an account has much in common with J. S. Mill's anti-paternalist, individualist perfectionism.

86 Rep. 1.47; and Att. 15.13.3.

87 Leg. 2.12, 1.48.

88 Clu. 146.

89 Leg. 3.44; cf. Dom. 43.

90 Leg. 3.44.

91 Caec. 95, trans. Yonge.

92 Caec. 96.

93 Leg. 2.11. Cf. Lon Fuller's 8 requirements for the rule of law, namely generality, promulgation, that laws be prospective, clarity, non-contradiction, that they be possible, constant, and that there be congruence between official action and declared rule, which implies due process. Fuller, The Morality of Law, rev. ed. (New Haven: Yale University Press, 1969).

94 Leg. 3.2, 3.6; cf. Rep. 2.54; 2Verr. 5.143.

95 Dom. 33.

96 We follow Karl Büchner's interpretation of ius autem legis aequale; aequale “muss heissen, dass das ius in Hinsicht auf das Gesetz in dessen spezifischer Eigenschaft, für alle gleichmässig zu gelten, die Gleichheit des Gesetzes ist.” Aequale has this meaning rarely, but it does so at Leg. 1.49, too: societas hominum et aequalitas et iustitia per se est expetenda. See Büchner, M. Tullius Cicero, De re publica: Kommentar (Heidelberg: Universitätsverlag Winter, 1984), 136f.

97 Cic. Off. 2.42: Ius enim semper est quaesitum aequabile; neque enim aliter esset ius.

98 See Fantham, “Aequabilitas”; A. Dyck, “On the Interpretation of Cicero, De Republica,” Classical Quarterly 48, no. 2 (1998): 564–8.

99 Straumann, Crisis and Constitutionalism, 54–62, 286–95. See Cic. Off. 1.26; 2.24, where Caesar by overthrowing iura and suppressing leges also suppresses by the same token libertas and makes the state unfree.

100 Leg. Agr. 2.102.

101 Off. 2.41.

102 Caec. 70.

103 Leviathan, 2.24 (p. 388).

104 See, for this view of Hobbes as a champion of legality and a kind of legally constituted civil liberty, D. Dyzenhaus, The Long Arc of Legality, 115–20.

105 Leg. 3.39.

106 Rep. 2.39f.

107 Leg. 3.10; 3.39. See V. Arena, “Popular Sovereignty in the Late Roman Republic,” in Q. Skinner, R. Bourke (eds.), Popular Sovereignty in Historical Perspective (Cambridge: Cambridge University Press, 2016), 73–3, for the view that this represented a “quasi-alienation” of the people's sovereignty.

108 Rep. 2.56. Indeed, the one instance where one can find an unmistakable association of participation and liberty is at Rep. 1.47, but here Scipio presents the argument for democracy; liberty as participation is once again seen as a specifically democratic idea, and the examples are Athens and Rhodes. Romans, by contrast, are considered free only by name (verbo). Many thanks to Malcolm Schofield for discussing this passage.

109 See Lintott, Constitution of the Roman Republic, ch. 7. The resulting picture of the res publica as a res populi, and of the control exercised by the people over their property, is one in which that control looks rather Schumpeterian: the people in assembly can vote people into office, and the magistrates have to step down after a year; the people can and do legislate, but it is only the magistrate who has the initiative and can put forward proposals. For the debate on the extent to which the Roman republic had democratic features, see Millar; Hölkeskamp; and now Yakobson.

110 Livy 2.1.1: imperiaque legum potentiora quam hominum.

111 Livy 2.1.7: libertatis autem originem inde … quia annuum imperium consulare factum est.

112 Note that according to Scipio's definition in Rep. 1.39, a proper res publica requires agreement about the law, iuris consensus. This is not, however, a contractarian view, where agreement (consensus) creates what is normatively binding (ius); rather, it is the rationalist view that rational insight into what the natural law requires creates the agreement. See Schofield, Cicero. Also, note that what is required is a iuris consensus, not a consensus about happiness or the proper conception of the good (i.e. not beatitudinis consensus, nor summi boni consensus).

113 See Dicey, Introduction to the Study of the Law of the Constitution (London: Macmillan, 1889), 181, 176; Gertrude Himmelfarb, “The Politics of Democracy: The English Reform Act of 1867,” Journal of British Studies 6, no. 1 (1966): 97–138; and Greg Conti, “Introduction,” to Dicey, Writings on Democracy and the Referendum (Cambridge: Cambridge University Press, forthcoming).

114 On the importance of this distinction, see Daniel Lee, The Right of Sovereignty: Jean Bodin on the Sovereign State and the Law of Nations (Oxford: Oxford University Press, 2021), chap. 3.

115 For the idea of conceptual overlap and conceptual change, see Straumann, “The Energy of Concepts: The Role of Concepts in Long-Term Intellectual History and Social Reality,” Journal of the Philosophy of History 14 (2020): 147–82.

116 Niccolò Macchiavelli, Discorsi sopra la prima deca de Tito Livio, ed. Franceso Bausi (Rome: Salerno, 2001), 2 vols.; 1.16 (pp. 102–3); English trans. (modified) in Discourses on the First Ten Books of Titus Livius, vol 2. of The Historical, Political, and Diplomatic Writings of Niccolo Machiavelli, trans. Christian E. Detmold (Boston: Osgood and Co., 1882), 138.

117 J. G. A. Pocock discussed the political tradition that recognized “liberty defined by law,” but argued that it was “of the empire rather than the republic,” and thus not central to Machiavelli (it was a language “he had altogether ignored”): see “Virtues, Rights, and Manners: A Model for Historians of Political Thought” (1981), in Virtue, Commerce, and History (Cambridge: Cambridge University Press, 1985), 43, 46. In an early work on Machiavelli, by contrast, Skinner did connect Machiavelli's theory of liberty with law: see Machiavelli (Oxford: Oxford University Press, 1983), 9–10; disc. in Carl K.Y. Shaw, “Quentin Skinner on the Proper Meaning of Republican Liberty,” Politics 23, no. 1 (2003): 46–56 (49).

118 Discorsi, 1.22 (p. 127); Discourses, 150 (trans. modified).

119 Discorsi, 1.24 (p. 133); Discourses, 153 (trans. modified).

120 See John Najemy, A History of Florence, 1200-1575 (Oxford: Blackwell, 2008).

121 See Gabriele Pedullà, “Humanist Republicanism: A New Paradigm,” History of Political Thought 41, no. 1 (2020): 43–95; and Peter Stacey, “Liberty and the Rule of Law: Freedom in Renaissance Republicanism,” in A Cultural History of Democracy in the Renaissance, ed. Virginia Cox and Joanne Paul (London: Bloomsbury, 2021), 41–61.

122 “A Defense of the Roman Origins of Florence,” in Images of Quattrocento Florence: Selected Writings in Literature, History, and Art, ed. and trans. Stefano Ugo Baldassarri and Arielle Saiber (New Haven: Yale University Press, 2000), 7.

123 Clu. 146: legum denique idcirco omnes servi sumus, ut liberi esse possimus. Trans. ours. It is possible that Salutati knew the Pro Cluentio, a copy of which had been given to Petrarch in 1355 by Boccaccio: Silvia Rizzo, La tradizione manoscritta della Pro Cluentio di Cicerone, 26. But for an argument that Salutati did not know the Pro Cluentio, see Peter L. Schmidt, “Zur Rezeption von Ciceros politischer Rhetorik im frühen Humanismus,” in Renaissance-Rhetorik, ed. Heinrich F. Plett (De Gruyter, 1993), 23–42. It is certainly possible that Salutati knew the passage in question.

124 When dealing with Catiline's conspirators, Cicero (as consul) ordered them executed without appeal (thus violating the right of provocatio). He would subsequently be exiled for this action. See Straumann, Crisis and Constitutionalism, chs. 2–3 for the political and institutional history and ch. 4 for Cicero's political thought.

125 Discorsi, 1.16 (p. 105); Discourses, 139–40.

126 Discorsi, 1.16 (p. 102); Discourses, 138.

127 Shklar herself would probably not have agreed with this remark, as she refers to “the deeply illiberal prerevolutionary republican tradition of which John Pocock has reminded us so forcefully” in The Machiavellian Moment. See “The Liberalism of Fear,” in Liberalism and the Moral Life, ed. Nancy Rosenblum (Cambridge: Harvard University Press, 1989), 22 (and for the definition, 21).

128 Discorsi, 1.2 (pp. 22–6), largely paraphrasing Polybius book 6.

129 Discorsi, 1.5 (p. 37), also for the examples of aristocratic states that have preserved political liberty. Cf. John McCormick, Machiavellian Democracy (Cambridge: Cambridge University Press, 2011).

130 Discorsi, 1.18 (p. 113); Discourses, 114 (trans. modified). For Aristotle, see above, and Pol. 2.1269a; see also JJR, SC, 4.1.

131 Leviathan, 2.21 (p. 332).

132 See Nikola Regent, “Quentin Skinner, Contextual Method, and Machiavelli's Understanding of Liberty,” History of the Human Sciences (2022), 1–27: “Regarding this kind of liberty Machiavelli seems to be agreeing with Hobbes … ” (7). For the crucial differences in underlying assumptions, apart from the narrower issue of liberty, see Straumann, Crisis and Constitutionalism, ch. 7. The strongest claim for a radical distinction between republican and liberal theories of liberty can be found in Skinner, Liberty Before Liberalism.

133 Noel Malcolm, “Thomas Hobbes: Liberal Illiberal,” 131, 132. Emphasis Malcolm's.

134 Ibid, 130.

135 David Dyzenhaus, “Hobbes on the Authority of Law,” 198 (n.); id., The Long Arc of Legality: Hobbes, Kelsen, Hart (Cambridge University Press, 2022), ch. 2.

136 Hobbes, Leviathan, ch. 20; see Susanne Sreedhar, “Interpreting Hobbes on Civil Liberties and Rights of Resistance,” in Interpreting Hobbes's Political Philosophy, ed. by S. A. Lloyd (Cambridge, UK: Cambridge University Press, 2019): 141–55.

137 See Dyzenhaus, Long Arc of Legality, 115–20, esp. 120: “the bonds of the law create civil liberty … . Hobbes's concern was thus more the quality than the quantity of civil liberty. That a universal quality is secured … is important to understanding why for Hobbes sovereign rule is not arbitrary in the way contemporary ‘republican’ philosophers today allege.” For liberty as a condition of obligation, see Hobbes, Leviathan, ch. 20, where slaves are not given liberty and thus “have no obligation at all,” whereas the subject “hath corporal liberty allowed him, and upon promise not to run away, nor do violence to his master, is trusted by him.”

138 Vickie B. Sullivan, Machiavelli, Hobbes, and the Formation of a Liberal Republicanism in England (Cambridge: Cambridge University Press, 2006), 46.

139 Discorsi, 1.16 (p. 106); Discourses, 140 (trans. modified); emphasis added.

140 See Lee, Right of Sovereignty.

141 See Brian Tierney, “'The Prince Is Not Bound By the Laws’: Accursius and the Origins of the Modern State,” Comparative Studies in Society and History 5 (1963): 378–40; Kenneth Pennington, The Prince and the Law, 1200-1600: Sovereignty and Rights in the Western Legal Tradition (Berkeley: University of California Press, 1993).

142 See Rousseau, Discours sur l’origine et les fondements de l’inégalité, in vol. 3 of Œuvres complètes, ed. Bernard Gagnebin and Marcel Raymond (Paris: Gallimard, 1964); henceforth, JJR; 112. On the Dedication, see Helena Rosenblatt, Rousseau and Geneva: From the First Discourse to The Social Contract, 1749-1762 (Cambridge: Cambridge University Press, 2007), chp. 2.

143 JJR, 112; emphasis added. Cf. also Montesquieu, Spirit of Laws, 11.3: “La liberté est le droit de faire tout ce que les loix permettent ; et si un citoyen pouvoit faire ce qu'elles défendent, il n'auroit plus de liberté, parce que les autres auroient tout de même ce pouvoir” (emphasis added).

144 See e.g. Pettit, “Two Republican Traditions,” in Republican Democracy: Liberty, Law and Politics, eds. Andreas Niederberger and Philipp Schink, (Edinburgh: Edinburgh University Press, 2013), and by the same author, “Rousseau's Dilemma,” in Engaging with Rousseau: Reaction and Interpretation from the Eighteenth Century to the Present, ed. Avi Lifshitz (Cambridge: Cambridge University Press, 2016). See also de Dijn, “Rousseau and Republicanism.”

145 This is the more common argument among Rousseau scholars: see e.g. Joshua Cohen, Rousseau: A Free Community of Equals (Oxford: Oxford University Press, 2010); Frederick Neuhouser, “Jean-Jacques Rousseau and the Origins of Autonomy,” Inquiry 54, no. 5 (2011): 478–93. For an interpretation more in line with the one proposed here, see Robert Wokler, “Rousseau's Two Concepts of Liberty,” (1987), in Rousseau, the Age of Enlightenment, and Their Legacies, ed. Bryan Garsten (Princeton: Princeton University Press, 2012), chp. 10.

146 Du contrat social, 2.3 (on the will of all vs the general will), in JJR, 3:371–72; and 2.7, in JJR, 3:385 (for quote; see also 4.2, in JJR, 3:440).

147 Du contrat social, 1.6; JJR, 360–1; see also 2.4, in JJR 3:374.

148 Du contrat social, 2.4; JJR, 374.

149 Ernst Cassirer, The Question of Jean-Jacques Rousseau, trans. Peter Gay (New York: Columbia University Press, 1954).

150 Du contrat social, 2.4; JJR, 3:374.

151 Du contrat social, 4.1; JJR, 3:437.

152 On Rousseau's gendered norms of citizenship, see notably Joel Schwartz, The Sexual Politics of Jean-Jacques Rousseau (Chicago: University of Chicago Press, 1985); Penny A. Weiss, Gendered Community: Rousseau, Sex, and Politics (New York: NYU Press, 1995); and Catherine Larrère, “Jean-Jacques Rousseau on Women and Citizenship,” History of European Ideas 37, 2 (2011): 218–22.

153 “Dédicace,” Discours sur l’inégalité, in JJR, 3:119–20.

154 For an earlier, narrower attempt at showing this, see Straumann, “Is Modern Liberty Ancient? Roman Remedies and Natural Rights in Hugo Grotius's Early Works on Natural Law,” Law and History Review 27, no.1 (2009): 55–85.

155 On the need to question the division between liberal and republican lines thought, see notably Sullivan, Machiavelli, Hobbes; and Richard Bourke, “Rights, Property and Politics: Hume to Hegel,” in The Cambridge History of Rights, vol. 4, ed. Dan Edelstein and Jennifer Pitts (Cambridge: Cambridge University Press, forthcoming). Bourke is responding to Pocock, “Virtues, Rights, and Manners.”

156 In The Lost History of Liberalism: From Ancient Rome to the Twenty-First Century (Princeton: Princeton University Press, 2018), Helena Rosenblatt briefly considers Cicero and Seneca, though focuses on their ideas about liberality (see 9–11). For notable accounts that focus on the post-Renaissance, see Friedrich Hayek, The Road to Serfdom (1944), vol. 2 of The Collected Works of F. A. Hayek (Chicago: University of Chicago, 2007); Pierre Manent, An Intellectual History of Liberalism (1987), trans. Rebecca Balinski (Princeton: Princeton University Press, 1995); Annelien de Dijn, French Political Thought from Montesquieu to Tocqueville (Cambridge: Cambridge University Press, 2008); and Alan Ryan, The Making of Modern Liberalism (Princeton: Princeton University Press, 2012).

157 The same arguably holds true for democracy: see e.g. Stephen Sawyer, “The Forgotten Democratic Tradition of Revolutionary France,” Modern Intellectual History 18, no. 3 (2021): 629–57.