Taxing Wealth in the Just City: Cicero and the Roman Census

Abstract

Cicero claims that states were created for the protection of property, so a statesman should try to avoid levying property taxes. A contrary principle holds that, as long as the state is common to all, those who benefit from it most should compensate those who benefit least to maintain distributive justice. With this frame of reference, the article asks two related questions. First, to what extent does Cicero differ from Dionysius of Halicarnassus, Livy, and the Stoics, who describe compensation or common ownership as a principle of fiscal fairness? Second, how does Cicero's political theory reflect the misgivings of wealthy Romans about state power and redistribution in the absence of compensatory taxation from 167 to 43 b.c.e.? I argue that his interpretation of the Servian census entrenches the ‘pre-fiscal’ distribution of property in the Roman constitution, which compromises the impartiality of the state and weakens its ability to respond to fiscal crises.

I INTRODUCTION

At the end of Book Two of De officiis (44 b.c.e.), Cicero turns from the duties of private individuals to the duties of the state's administrators. He admits that they should provide moderate assistance to citizens as long as they do not exhaust the treasury like Gaius Gracchus did with his grain dole.¹ ‘An administrator of the state (res publica), however, must make it his priority that everyone shall keep what is his (suum) and that the wealth of private citizens is not reduced by the public (de bonis privatorum publice deminutio).’² Nothing could be worse, Cicero adds, than the equalisation of property:

For the main goal in the establishment of states and civil communities (res publicae civitatesque) was that individuals might keep what was theirs (sua). For even if men came together by nature's guidance, they still sought the protection of cities (urbium praesidia) in the hope of safeguarding their possessions (spe custodiae rerum suarum). Efforts must also be made to avoid the levy of a property tax (tributum), which was often done in the time of our ancestors because of an empty treasury and frequent wars, and to act with foresight so this does not happen. However, if the necessity for this duty (munus) arises in any state (res publica) — I prefer not to forebode any evil for our state and am not discussing ours but states in general — then efforts must be made to ensure that everyone realises that if they wish to be saved they must yield to the necessity.³

Tributum was a public duty (munus) but was not counted among the revenues (vectigalia) that Cicero deemed necessary for the state (res publica) such as rents or indirect taxes and the tribute paid by Rome's provincial subjects.⁴ For example, when P. Servilius Rullus introduced agrarian legislation in 63 b.c.e. to parcel out the public land of Campania, which was leased out by the censors for rent, Cicero argued that the state's property (patrimonium) had been crucial to the maintenance of Rome's armies and grain supply during the Social War (91–87 b.c.e.) when other revenues were disrupted.⁵ The state, he thought, ought always to maintain sufficient resources to carry out its functions, above all, the protection of citizens’ property, without having to diminish it by taxation.

This article aims to elucidate Cicero's conception of fiscal fairness within its intellectual and historical context. By means of the census, I argue, Cicero links legitimate political authority to the pre-fiscal distribution of property, laying constraints on the government. Sections II and III examine the intellectual significance of these constraints: Cicero, Livy and Dionysius of Halicarnassus use the first Roman census undertaken by King Servius Tullius to illustrate different principles of fairness (Section II) and Cicero's theory also deviates from the Greek Stoic tradition (Section III). Sections IV and V situate these ideas in relation to two landmarks in Roman fiscal history: the suspension of tributum in 167 b.c.e. (Section IV) and its return in 43 b.c.e. shortly after Cicero wrote De officiis (Section V).⁶ The historical context illuminates Cicero's originality as well as the implications of his theory for the fiscal crisis in 43 b.c.e., which demonstrated the weakness of the late republican state and the relevance of morally exemplary instances of compensatory taxation in early Roman history.

The notion of ‘pre-fiscal’ property rights helps us appreciate the philosophical significance of Cicero's theory. I avoid the term ‘pre-political’ because, as we shall see in Section III, Cicero posits a pre-political state of nature, where those rights do not exist, contrasting it with the civil community (civitas), where they emerge, and the state (res publica), which secures them with its laws and revenues.⁷ The notion of ‘rights’ (iura) has sometimes been considered a later European development but this gives too little credit to Cicero and other Roman writers.⁸ Without rights that protect citizens and their property from the state, they are in danger of being used as means to achieve its autonomous and possibly tyrannical ends. Even if those ends are putatively just and beneficial, without rights each individual must live in fear of being sacrificed to the common good, unless, that is, they can cope with the mental utopia of a Stoic sage. Given particularistic human instincts and economic incentives, such insecurity may have negative effects on the whole community. However, Cicero's solution, I argue, compromises the state's impartiality. As long as rights, law, liberty and the state are common to all, to steal a phrase from Cato the Elder, those who benefit from them most should compensate those who benefit least in order to maintain distributive justice.⁹

Historians usually place the emergence of the ‘tax state’ or ‘fiscal state’ in early modern Europe. Feudal monarchs were supposed to finance their activity from patrimonial domains and personal regalian rights. The rising costs of warfare gradually drove the nobility, clergy and urban elites to consent to the autonomy of the state as representative of the political community.¹⁰ One of its intellectual impulses was the movement known as ‘neo-Stoicism’ led by the Flemish philosopher Justus Lipsius (1547–1606). The Prussian cameralist Johann von Justi (1717–1771) justified direct taxes in this neo-Stoic vein, on the grounds that private property was the mediated property of the state, which secured everyone's welfare.¹¹ Cicero's political theory, by contrast, had placed constraints on the government's authority to levy taxes without consent. The passage from De officiis quoted above was paraphrased by the delegate from Brunswick-Wolfenbüttel at the German Reichstag of 1653/1654, when he opposed the emperor's proposal to have the diet approve taxes by majority vote.¹² It is best known for its influence on John Locke's Two Treatises of Government (1689), where he argued that government was established for the protection of property.¹³

Those debates in early modern Europe were in some respects analogous to the conflicting conceptions of fiscal fairness examined in Sections II and III, which were likewise enlisted to support or oppose the formation of ‘tax states’ in the ancient world. The fiscal history of the Republic described in Sections IV and V was quite different from that of European states in the early modern period. Rome's empire allowed it to pass most of the tax burden on to provincial subjects during the second and first centuries b.c.e. In the final analysis, a fiscal crisis laid bare the structural weakness of the state. Cicero sought to justify oligarchical institutions that prevented it from levying compensatory taxes on those who benefited most from the protection of property rights. Under those constraints, whoever controlled the government, unless it was the property owners themselves, had little incentive to provide that protection.

II KING SERVIUS TULLIUS AND FISCAL FAIRNESS

In a passage with Stoic inspiration, Cicero compares the early kings of Rome with Deioces, whom the Medes chose as king due to his reputation for justice. In the Herodotean tale, Deioces exploited his role as an indispensable judge of legal disputes to convince the people to furnish him with guards and build seven concentric walls around his palace and treasury (θησαυροί). This edifice gave him sovereign power (τυρρανίς) that rendered him autonomous from society, illustrated by his withdrawal into the fortress, where he judged cases only in response to written petitions. Cicero claims that the people (multitudo) of Rome similarly elevated virtuous men to the kingship to protect the weak from the strong with an equal right (pari iure). Laws were established for the same reason, to ensure impartial justice even without virtuous kings.¹⁴ While many of Rome's kings introduced good and useful institutions, Servius Tullius, Cicero thought, ‘had the greatest insight of them all into the state’.¹⁵

It may be tempting to see in Cicero's ideal constitution, with its mixture of magisterial power, senatorial authority and popular liberty, a modern ‘shackled Leviathan’ with the organisational autonomy to maintain the government's impartiality and the institutional guardrails to preserve the people's freedom.¹⁶ I argue, however, that Cicero's conception of the Servian census put Rome under significant constraints with respect to taxation, which undermined its autonomy and entrenched the political domination of the wealthy citizens. To understand what I mean by ‘significant’, it is instructive to contrast how Locke resolved the conflict between private property and public taxation. When Locke published Two Treatises of Government in 1689, England was rapidly emerging as the most precocious fiscal state in Europe.¹⁷ Similarly to Cicero, he believed that governments were created for the protection of private property but he was well aware of the need for public spending:

It is true, governments cannot be supported without great charge, and it is fit every one who enjoys his share of the protection, should pay out of his estate his proportion for the maintenance of it. But still it must be with his own consent, i.e. the consent of the majority, giving it either by themselves, or their representatives chosen by them: for if any one shall claim a power to lay and levy taxes on the people, by his own authority, and without such consent of the people, he thereby invades the fundamental law of property, and subverts the end of government.¹⁸

Locke's transition from the state of nature to civil government proceeds by majoritarian consent.¹⁹ Thomas Hobbes had already used the majoritarian principle, especially in De cive (1641) but implicitly still in Leviathan (1651), to describe the social covenant that justifies people's obedience to their sovereign, though contrary to both Locke and Cicero he did not ascribe property rights any priority before civil government.²⁰

Cicero, on the other hand, emphatically rejects majoritarianism in De re publica when describing the first census conducted by King Servius Tullius. The section is fragmentary but accords with Livy and Dionysius of Halicarnassus: he ranked citizens into six classes based on the valuation of their property and assigned them to 193 divisions called centuries, each of which had one vote in the centuriate assembly (comitia centuriata). The centuries of equestrians and first-class citizens were small with the privilege of voting first, so many elections were finished before the lower classes could vote. The entire sixth class was assigned to a single century, where citizens had virtually no voting power at all. Cicero comments:

He distributed them in this way so that the votes (suffragia) were under the control (potestas) of the wealthy (locupletes) and not the multitude (multitudo). He thus guarded the principle that the greatest number should not be the most powerful, which must always be affirmed in the state (res publica).²¹

Valentina Arena detects a shift in Cicero's thinking from De re publica, where the people are sovereign, to De legibus, where they have only the ‘semblance of liberty’ (species libertatis).²² This passage, however, suggests that the populus properly emerges only after the Servian census, which encodes the distribution of property into the centuriate assembly. It was a creature of the law rather than its sovereign and yielded, by design, votes favourable to the wealthy.²³ The state could not become autonomous from the interests of propertied citizens, could not levy taxes by majority vote and could not use taxation to compensate for the advantages that those citizens derived from state policies. Such constitutional constraints are significant not only for their contrast with Locke and modern political theory. They are also historically significant in their own context because they posed a challenge to Cicero's contemporaries. De re publica was composed c. 54–51 b.c.e., soon after Caesar, Pompey and Crassus made their informal pact known as the first triumvirate in 56 b.c.e. As a senator and former consul, the author was aware of the controversy his ideas would stir up and the danger to himself.²⁴ He was trying to unify and mobilise the wealthy citizens, the ‘good men’ (boni viri) who valued peace and security, as a political force in this tumultuous period.²⁵

Cicero obviously knew the tradition, so prominent in Livy and Dionysius, that King Servius Tullius created the census to distribute military and fiscal burdens fairly, but he barely mentions it.²⁶ The king was also the legendary builder of the city's walls, which required the imposition of public burdens (munera), whether labour or taxes, for the common defence.²⁷ In the late Republic, the Servian legacy was being instrumentalised by competing factions in Rome to justify reforms in conformity with constitutional precedents. Emilio Gabba argues that older democratic legends were replaced by an oligarchical narrative after Sulla evoked King Servius Tullius as a model for his reforms.²⁸ However, the spectre of ‘Sullan kingship’, which some aristocrats, notably Pompey, were accused of reviving, was just as horrifying to Cicero and the boni viri as its Gracchan antithesis.²⁹ It is futile to reconstruct standard oligarchical or democratic versions influencing Cicero, Livy and Dionysius. They were conscious of contradictory claims about Servius Tullius in circulation and reconciled them according to their own rhetorical and philosophical aims. While the census was virtually irrelevant to Roman public finance when Cicero wrote De re publica, the return of tributum in the second triumviral period left its impression on Livy and Dionysius (see Section V).

Dionysius of Halicarnassus came to Augustan Rome around 30 b.c.e. and treats its archaic institutions in his history as those of a Greek polis, but was obviously drawing on earlier Roman sources.³⁰ Before describing Servius Tullius’ reforms, Dionysius has the king give two speeches to win support from the poorer citizens for his usurpation. There he promises to relieve debt, abolish debt slavery, tax the wealthy and redistribute public land.³¹ He vows ‘to make the government (πολιτεῖα) equal and impartial (ἴση καὶ κοινή) and justice (τὰ δίκαια) the same for all and towards all’, to prevent the wealthy from abusing the poor as if they were slaves.³² In response, the assembly applauds him as a lawful, just and democratic ruler, giving the people's consent to his acquisition of kingship.³³ Such a vision corresponds to an autonomous and impartial state that uses taxation to compensate for other forms of social inequality. The king apportions tax burdens and redistributes land to citizens based on criteria of fairness beyond simply the ability to pay: he has wealthy citizens compensate poorer ones for their advantages from the state.

The following passage from Dionysius of Halicarnassus contains a description of how tributum was partitioned and conveys a distinctive conception of fiscal fairness.

Once he calculated the expenditure to be levied for the provisioning of the soldiers and for the other war supplies and he distributed whatever amount would be sufficient in the same way [as in the levy of soldiers] among the 193 centuries, he would order them all to contribute the share allotted (ἐπιβάλλον) to each century based on their property declaration (ἐκ τῆς τιμήσεως). So it happened that those with the greatest wealth, who were fewer in number but distributed into more centuries, had to serve more often and without any intermission and to pay greater taxes than the rest.³⁴

Clearly this scheme of conscription and taxation has been idealised to make a point about justice, though it is conceivable that centuries paid taxes together (see Section IV). Dionysius describes shares allotted (ἐπιβάλλον) to centuries based on their property declaration (ἐκ τῆς τιμήσεως).³⁵ If apportioned equally, as he suggests, over half of the entire fiscal burden would be shouldered by those centuries of the first census class. It also furnished the heavy infantry and served in the front rank, bearing the brunt of the clash with the enemy, while the second class was drawn up behind them, the third, fourth, fifth and sixth classes respectively further towards the rear.³⁶ King Servius Tullius made them compensate the poor by suffering greater hardships ‘with their persons and their money’ because they had greater ‘prizes’ (ἆθλα) at stake in war.³⁷

Following Dionysius closely, John Adams in his Defense of the Constitutions of the United States of America (1787) perspicuously focuses attention on the weakness of the king. He praises his ‘excellent equitable regulations’ with respect to taxation and military service but accuses Servius Tullius of ‘undermining the authority of the people’ in order to ‘appease the fury of the patricians’ with his unequal voting laws: ‘the king had been driven to the necessity of this artful flattery of the patricians, by his not being independent of them, and by their sharing with him in the executive power’.³⁸ The people, in other words, placed their naïve faith in the virtue of the king and failed to establish a government with checks and balances that would be strong enough to protect them from the domination of wealthy patricians. By the moral standard that Dionysius had established in the speeches of Servius Tullius, the king's deception was obviously unjust, breaking his promise to establish an impartial government. Dionysius later states in a more elitist tone that the people eventually acquiesced and presents the same justification found in Livy's version: it was best for those who bore a greater fiscal burden to control the government. Before Servius betrays the people, however, Dionysius has him express sharp criticism of the republican oligarchy with the democratic rhetoric of equality and fiscal fairness.³⁹

According to Livy, by contrast, King Servius Tullius established the census so that posterity would celebrate him as ‘the originator of all distinctions in the civil community (civitas) and of the orders which clearly differentiate the gradation of worthy standing and fortune (gradus dignitatis fortunaeque)’.⁴⁰ On this point Cicero and Livy basically agree, but Livy adds that Servius Tullius compensated the wealthy citizens for bearing all fiscal burdens (omnia onera) with a corresponding honour (honos), namely the gradation (gradus) in voting, which effectively gave them all power (vis omnis) in the civil community (civitas).⁴¹ He has the consul Laevinus articulate a similar argument for fiscal fairness in 210 b.c.e., claiming that magistrates and senators had a special obligation to contribute to the public treasury and perform liturgies such as manning the fleet with their slaves because they were superior in honour (honos). After the crisis subsided, at Laevinus's suggestion, they used their political power to obtain a reimbursement, which testifies to the illusory effect of compensation when it comes to justifying control over the state and its allocation of resources.⁴² Claude Nicolet argues that the principle of ‘geometrical equality’ expressed in these passages was derived from Greek philosophy and constituted the normative ideology of the Roman census. With a touch of moral relativism, Jérôme France further generalises the correlation between fiscal or liturgical contributions and unequal political power: ‘for the ancients, this was a just system’.⁴³ However, the standards of proportionality in Livy are not the same as in Cicero and Dionysius of Halicarnassus.

Cicero justifies the voting reforms of Servius Tullius in De re publica without any reference to tributum or any other compensatory fiscal contributions. The wealthy deserved their voting power in the centuriate assembly, he claims, because they had the greatest interest in the ‘best form of civil community’ (optimus status civitatis).⁴⁴ They could presumably be persuaded to pay taxes, if this state required it, but had no moral obligation to compensate the poorer classes for their voting power, much less for their wealth. Somewhat earlier in the dialogue, he anticipates Livy's language for describing the Servian census when he condemns democratic government: ‘for the resulting equality itself is inequitable, since it allows no gradation of worthy standing (gradus dignitatis)’.⁴⁵ However, Cicero draws a distinction between the populus weighted by wealth and the administrators chosen by merit, who constituted the aristocratic element of the constitution. For its own safety, the populus entrusts itself (se committat) to those superior in virtue and spirit (summi virtute et animo) to serve as managers and caregivers of the state (res publica).⁴⁶ The best citizens (optimates), in other words, choose the best administrators based on their ‘worthy standing’ (dignitas), not their wealth.⁴⁷ Like his peers, Cicero associated wealth with moral superiority, but there were always exceptions, ostensibly ‘good men’ (boni viri) without virtue, so the populus weighted by wealth served as a backstop for the preservation of good government.⁴⁸

Dionysius adopts the same justification for the unequal Servian voting centuries as Livy or their common sources, but only after presenting the king as the democratic champion of impartial justice. The higher classes would have to pay higher taxes to compensate the poorer classes for the public protection of their wealth. To give them greater voting power in addition would amount to domination.⁴⁹ It would undermine the impartiality of the state, allowing the wealthy to reimburse themselves with other benefits. This version could have been inspired by Greek models (Section III), alternative Roman sources for the king's popular legacy (Section IV), an Augustan revival of that legacy (Section V), or all of these.⁵⁰ Since it foreshadows the debate later in his narrative over the legitimacy of the property-weighted centuriate assembly in the trial of Coriolanus, it is likely that his Roman sources also discussed popular challenges to the oligarchical conception of fiscal fairness.⁵¹ The point is that we are dealing with different ‘geometrical’ standards of justice: wealth to taxes in Dionysius (4.19.3); taxes to voting power later in Dionysius (4.21.1) and in Livy (1.43.10); and wealth to voting power in Cicero (Rep. 2.40). Nicolet is not the only scholar to read each of these passages in connection with King Servius Tullius as expressions of the same ideology.⁵² By separating them, we can better appreciate what is distinctive about Cicero's theory of pre-fiscal property rights.

III CICERO AND THE STOICS

As a young man in 78 b.c.e. Cicero visited the Greek island of Rhodes, where he studied with the Stoic philosopher Posidonius, among others. He credits the Stoics, especially Posidonius’ teacher Panaetius of Rhodes (c. 185–109 b.c.e.), as models for his De officiis. Since little survives of their work, modern scholarly opinion is divided over the extent of Cicero's originality, but the tide is turning in his direction, as he emphasises that he relied on his own judgement and was no mere translator.⁵³ Either way, a case could be made for the influence of Roman legal thought on Panaetius, as has been done for Polybius, who both belonged to the intellectual circle of Scipio Aemilianus in Rome.⁵⁴ However, if Panaetius wrote ‘On Duty’ (Περὶ τοῦ καθήκοντος) before 133 b.c.e. or did not discuss the Gracchi, then Cicero's reflections on them, as on his own contemporaries, were independent of him.⁵⁵ In this section, I argue that Cicero's conception of fiscal fairness diverges from Stoicism in ways that reflect concerns of the late republican oligarchy.⁵⁶

The Stoics’ core doctrine was that every human being has by nature true ownership of one thing, namely, himself or herself as a moral subject, which makes him or her fully human. By the psychological process of ‘appropriation’ (οἰκείωσις), one's proprietary care begins with one's body and extends through childhood to our families and further outward to all humankind as one acquires the mature identity of an anthropos. Human beings, male and female, learn justice through appropriation of one another as moral subjects, recognising themselves as fellow citizens in a world city or ‘cosmopolis’ governed by a constitution of rational natural laws. Unlike the Cynics, who renounced political status and material possessions to live in accordance with nature, the Stoics believed that being interested in property acquisition or money-making, which was also scorned by the elitist philosophy of the Academy and Lyceum, was natural and beneficial for the political community. Hence the Stoa was the only school to assign property a key role in human morality, anticipating developments in early modern philosophy.⁵⁷

Material wealth, nevertheless, counted only as one of the preferred ‘indifferents’ (ἀδιάφορα) in Stoic philosophy.⁵⁸ Poverty and slavery might be unnatural states, from which humans are rationally inclined to free themselves, but one cannot attach any moral significance to wealth or legal status. On this point, Long observes, ‘we can see why the ex-slave Epictetus, in contrast with the wealthy Cicero, emphasizes morality as a person's only inalienable property and source of autonomy’.⁵⁹ The German philosopher Hegel (1770–1831) took a position that might have pleased Cicero when he claimed that freedom and self-ownership become fully realised in the ownership of material property and the absence of slavery, but this certainly goes beyond Stoic doctrine.⁶⁰ Cicero implies in his discussion of the census of Servius Tullius that the higher wealth-classes were morally superior to the poorer classes (and a fortiori free men to slaves) because they had greater concern for the ‘best form of civil community’ (optimus status civitatis).⁶¹

The Stoics regarded property as being ultimately communal, its privatisation provisional and contingent on utility to the political community. In De finibus, Cicero has Cato the Younger report the view of Chrysippus (c. 279–206 b.c.e.), the third head of the Stoa in Athens: ‘But just as the communal nature of a theatre is compatible with the correctness of saying that the place each person occupies is his own (suum), so in the city (urbs) or world (mundus) which they share no right (ius) is infringed by each man's possessing what belongs to him’.⁶² As Phillip Mitsis notes, the idea is not comparable to Lockean property rights that impose constraints on government.⁶³ There is nothing ‘pre-fiscal’ in this conception that would prevent the community from demanding compensation from citizens for their private place in the theatre. Similarly, Scipio is a mouthpiece in De re publica for the Stoic doctrine that natural law confers ownership only on virtuous men who use their possessions on behalf of the community.⁶⁴ This, too, imposes public conditionality on the owner's legal right, which seems to allow for taxation or redistribution to the extent that (tax-exempt) private property does deviate from the public interest. However, it runs counter to Cicero's principle that citizens should suffer no reduction of their property by the public, even by taxation, except as a last resort.⁶⁵

In contrast to the Stoics, Cicero assigned moral significance to the ‘pre-fiscal’ distribution of property. Justice, which distinguishes ‘good men’ (boni viri), entails that ‘one use what is common for common interests, what is private for one's own (ut communibus pro communibus utatur, privatis ut suis)’. These ‘common’ things are not necessarily material possessions, as translators and commentators have assumed.⁶⁶ He gives examples later: the forum, temples, colonnades and streets, but also laws, rights, law courts, voting privileges, etc.⁶⁷ Cicero thus adopts a principle of justice in the use of common and private things found already in Cato the Elder.⁶⁸ This passage in De officiis goes on to describe the acquisition of private property by conventional legal norms, insisting that it would be unjust for anyone to violate its distribution. His formulation, ‘nothing is private by nature’ (sunt autem privata nulla natura), resembles Chrysippus’ simile of the theatre, but the emphasis on retaining one's property is apparently Cicero's own.⁶⁹ Brouwer assumes the passage is a fragment of Panaetius, dismissing the objection that Cicero refers to his home town of Arpinum, and points to it as evidence that Panaetius was one of the ‘Roman Stoics’, who transformed property from a preferred ‘indifferent’ into a moral right that could constrain the government.⁷⁰ More likely it was Cicero himself who denigrated communal ownership in the state of nature and defended privatisation as the just and lawful alternative. He took it for granted in his court speech Pro Sestio (56 b.c.e.) that the panel of judges, who were drawn from the wealthiest Roman citizens, would agree with his description of the state of nature (rerum natura), ‘before natural and civil law were laid down’, where only violence secured one's precarious possession of communal goods.⁷¹ Polybius and the Epicurean Lucretius depict a similarly unjust and lawless state of nature, but there is little evidence that any Stoic philosopher did.⁷²

Cicero's evocation of pre-fiscal property rights in court cases suggests that his assumptions were widely shared by wealthy Romans in the first century b.c.e. According to the next stage of political development described in Pro Sestio, wise men assembled humankind into civil communities (civitates) and taught them justice and mildness, allowing them to distinguish what is public or ‘for the common interest’ (res ad communem utilitatem) and, implicitly, what is private. Once these principles had been discovered, citizens took the further step of constructing walls around houses to form cities (urbes).⁷³ Unless that speech was modelled on Panaetius, the corresponding passage in De officiis (2.73) was also Cicero's own. The city (urbs), unlike the civil community (civitas), embodies the state (res publica) with its physical infrastructure and collective agency, with its public revenue and compulsory service (munus), for the protection of private property.⁷⁴ In another early speech, concerning his friend Aulus Caecina's ownership of a farm, he argues that an inheritance alone is not enough to secure one's property: ‘It is not possible for me to retain what has become mine (meum) without civil law (sine iure civili)’.⁷⁵ Thomas Hobbes quotes from this speech to show that ‘even Cicero’ supported his view that the state creates property rights, but Cicero means only that it has a duty to protect rights that exist by nature.⁷⁶ As in De officiis (2.73), these are ‘pre-fiscal’ in the sense that members of the civil community mutually recognise them before they empower the state as an autonomous organisation with the means and authority to secure them.

The subsequent chapters of De officiis, on misuse of public office, condemn populist reformers (populares) who, by proposing land redistribution and debt relief, undermine the ‘foundations of the state’ (fundamenta rei publicae) and its ‘harmony’ (concordia), which cannot exist ‘when money (pecunia) is taken away from some and given to others’. Fairness (aequitas) is ‘completely subverted if one is not entitled to possess what is one's own’ (si habere suum cuique non licet).⁷⁷ This distinctively Ciceronian section, in which he boasts of his own political legacy and reflects on the struggles between populares and optimates, is spiced with an example that probably does stem from Panaetius. After liberating his city in 251 b.c.e., Aratus of Sicyon used public money to compensate those disadvantaged by the inevitable unfairness caused by his reallocation of property rights to the returning exiles.⁷⁸ The funds were a subsidy from King Ptolemy II of Egypt, though the principle would be same if he had used a tax or public debt. Cicero contrasts the virtue of Aratus with Sulla, Caesar and Antony, who sold the confiscated property of proscribed citizens to raise money. They were not Stoic sages. Yet moral examples from Greek philosophy are not sufficient to do away with the legal constraints that Cicero, much like Polybius, deemed necessary to prevent abuses of public office.⁷⁹

It is difficult to imagine on what grounds any Stoic would reject the principle of compensatory taxation articulated by King Servius Tullius in Dionysius of Halicarnassus (4.19.3). The principle was so pervasive in Greek political culture that one need not assume a Stoic provenance. A city's territory was understood as collective property, where the demos specified individual rights and had the authority to levy taxes.⁸⁰ Property taxes or liturgies demanded from the wealthy classes, typically based on census declarations, could be seen as compensation for the public protection of those rights, committing even poorer citizens to the toleration of wealth inequality. Whether those burdens also entitled wealthy citizens to exclude or marginalise them politically, as in the Roman Republic, was a question which distinguished oligarchical from democratic conceptions of fiscal fairness. Dionysius's contemporary, the geographer Strabo, writing at the end of the first century b.c.e., describes the case of Rhodes:

The Rhodians are caring towards the people (δημοκηδεῖς) and though they are not governed democratically still wish to keep the multitude of poor men united. The people are supplied with provisions and the wealthy support the needy by means of a certain ancestral custom. There are certain liturgies for supplying provisions, so that the poor man receives his sustenance while the city does not run short of useful men, especially for its naval expeditions.⁸¹

For generations, wealthy Rhodians, including the family of Panaetius, consented to institutions that placed a burden on themselves to compensate poorer citizens.⁸² This is analogous to the justification of Servius Tullius in the Dionysian version for demanding greater contributions from those with more at stake in war. Rhodes practised a style of politics that was, as Strabo notes, de facto oligarchical since wealthy citizens were regularly elected to office, but the epigraphic evidence suggests that they earned democratic support by engagement with civil-society associations.⁸³

While compensatory arguments could be effective even in relatively oligarchical regimes, they became institutionally entrenched in democratic poleis. The anonymous Athenian known as the Old Oligarch writes that it was considered fair for the people to have political power and to benefit from financial burdens imposed on the wealthy because poorer citizens served as rowers in the Athenian navy. Scheve and Stasavage cite this passage as evidence that mass mobilisation in warfare was a powerful equaliser in ancient Greece, just as in recent European and American history.⁸⁴ Dionysius was familiar with the Athenian discourse on fiscal fairness from his meticulous study of fourth-century Athenian oratory, where the democratic government demanded compensatory property taxes and liturgies from the wealthiest citizens.⁸⁵ His own city of Halicarnassus was one of several local hotbeds of populist democracy around the time of his birth. Cicero's brother Quintus, as proconsul of Asia in 61–58 b.c.e., had to intervene forcefully there in order to ensure that the ‘best citizens’ (optimates) were in control of the state.⁸⁶

The oligarchs who plotted to overthrow the Athenian government in 411 b.c.e. resolved, according to Thucydides, ‘to contribute willingly from their own private resources either money or whatever else should be necessary, since they would no longer be enduring hardships for others but for themselves’.⁸⁷ As long as the constitution was aligned with their interests, they were of course willing to pay taxes. Similarly, Cicero argued that Roman property owners defined by the Servian census were entitled to greater voting power because their interests were aligned with the best constitution and its protection of property rights. If a property tax had to be levied, this required the taxpayers’ consent, which could not be obtained by a simple majority, as Locke later suggested. The senate, composed of former magistrates elected for merit with votes weighted according to property valuations, spoke with authority on their behalf, though it should also try to persuade the citizens individually. Cicero had to modify Stoic political theory in order to justify these inegalitarian constraints, which compromised the state's impartiality.

IV THE END OF TRIBUTUM IN 167 B.C.E.

An empty treasury and frequent wars, Cicero reminds his readers in 44 b.c.e., had made it necessary for their ancestors to levy tributum.⁸⁸ The prevailing opinion is that this tax was partitioned based on the actual or estimated expenditure for each campaign season among taxpayers as a percentage of the valuation in their census declaration.⁸⁹ The spoils brought back after the Third Macedonian War, coupled with sufficient annual revenue from provincial and patrimonial sources, enabled the senate to suspend tributum in 167 b.c.e.⁹⁰ After considering the old system, this section examines how its suspension might have shaped Roman debates about fiscal fairness in the late Republic.

Any reconstruction of tributum before 167 b.c.e. is conjunctural, but three points are worth emphasising, as they relate to the subsequent discussion. First, Rome's domination of Italy and the Mediterranean was achieved by regular military campaigns, for which the taxpaying property owners compensated citizen infantrymen and naval rowers.⁹¹ Their contingent cooperation gave the common citizens a certain power of negotiation or leverage over those who benefited most from Roman victories.⁹² Second, Rome's ability to pay them rested on the cooperation and financial liability of the ‘treasury tribunes’ (tribuni aerarii). The translation is misleading because they were not agents of the treasury: they were the wealthiest citizens in the tribes who paid the soldiers and collected taxes, thereby exercising their own leverage over the aristocratic senate and equestrian order.⁹³ Third, the linchpin of this effective system of collective action was the census, a ritual performed every five years, in which every citizen was sorted according to age and wealth into a century for military recruitment, taxation and voting.

The Servian census that distributed the citizens of each tribe into 193 centuries in six wealth classes probably goes back no earlier than the late fourth century b.c.e.⁹⁴ Those in the top five classes were called assidui, liable for property taxes and military service, while the proletarii of the sixth class were ‘assessed per person’ (capite censi) and exempt from all but naval service.⁹⁵ The minimum valuation for assidui was reduced several times to create a larger pool of potential conscripts and taxpayers.⁹⁶ Age groups were divided, half of the centuries in each class being iuniores (age 17–45) and half seniores (age 46–60), except in the proletarian sixth class.⁹⁷ The monetary units for the minimum property valuations suggest that Livy and Dionysius used the same or related sources, dating around 211–141 b.c.e.⁹⁸ Despite his Greek perspective, Dionysius’ history, just as Livy's, was based on Roman authors who knew the taxation system at first hand before its suspension in 167 b.c.e. Fabius Pictor (c. 270–200 b.c.e.) wrote in Greek and may have described Rome with similar atticising rhetoric or terminology, Cato the Elder (234–149 b.c.e.) was elected censor in 184 b.c.e., and Vennonius (second century b.c.e.) was possibly the author of a fragment with a similar description of the Servian centuries and tribes.⁹⁹

The centuries described by Dionysius and Livy were apparently already obsolete when the source they used was written, some time after 211 b.c.e. The centuriate assembly seems to have been reformed around 241–218 b.c.e. The eighty centuries in the first class (prima classis) were reduced to seventy with one junior and one senior century representing each of the thirty-five tribes. The right to vote first in elections was taken away from the eighteen equestrian centuries and instead drawn by lot by junior centuries of the prima classis.¹⁰⁰ These changes benefited tribes farther from Rome, which had been underrepresented at the assembly but bore a heavy burden of conscription and taxation during the exhausting First Punic War (264–241 b.c.e.). The other 105 centuries of the lower census classes were not adjusted to fit the tribes, which meant that they still had to rely on voters in attendance in Rome with the same property valuation, irrespective of tribe, to represent their interests in the assembly just as everyone had done before the reform.¹⁰¹

The partition of tributum by century that Dionysius describes (see Section II), if historical, must pre-date this reform and could never have been strictly equal among all 193 centuries as he suggests.¹⁰² Conscription would have targeted junior centuries, while taxation would fall harder on senior ones if soldiers on active duty were exempt.¹⁰³ By his own account, the entire sixth-class century was exempt and the two centuries of artisans contributed war supplies instead of money for soldiers’ pay.¹⁰⁴ This detail about artisans lends some credibility to the notion that centuries were used for the assessment of tributum within the tribes. A rhetorical treatise of the late fourth century b.c.e. suggests that it was common in the Greek world for cities to register artisans in the census separately from propertied citizens and to have them contribute weapons instead of money when property taxes (εἰσφοραί) were levied for war.¹⁰⁵

Roman taxation and conscription were organized by the tribes.¹⁰⁶ As Varro writes, ‘tributum was called so from tribus because the money, which was levied on the people, was exacted by tribe (tributim) from individuals in proportion to their property declaration (pro portione census)’. This money, he adds, was ‘assigned’ (attributum) to the tribuni aerarii for paying the soldiers.¹⁰⁷ The jurist Gaius and a fragment of Cato the Elder mention a legal procedure for soldiers to seize a pledge (pignoris captio) from the tribuni aerarii who pay them.¹⁰⁸ When a citizen was punished by Rome's censors with removal from his tribe, one scholiast explains, ‘he became a treasury citizen (aerarius) and was consequently not in the album of his own century (in albo centuriae suae) but remained a citizen only in the sense that he made payments for tributum for his own person (pro capite suo)’.¹⁰⁹ The latter probably paid directly to the treasury, typically with a punitively high valuation.¹¹⁰ Fellow tribesmen on the same century album, on the other hand, could conceivably have been assigned to the same tribunus aerarius. Centuries were small cohorts within the tribes, which were re-evaluated by the censors every five years according to wealth and age (and occupation in the case of artisans) and thus may have been used to identify eligible taxpayers and conscripts.¹¹¹

In the first century b.c.e., the tribuni aerarii were the wealthiest citizens of the prima classis.¹¹² That fact alone best explains the title's continuity from the time before 167 b.c.e., and so the debate whether they could in some sense be called ‘equestrians’ in the late Republic is beside the point. They constituted a social order (ordo) from the same ‘highest valuation’ (amplissimo ex censu) as senators and equestrians. That probably means that they met the equestrian qualification (400,000 sesterces) introduced in the late second century b.c.e., but were not enrolled in one of the eighteen equestrian centuries with a public horse.¹¹³ Henrik Mouritsen argues that all citizens with over 400,000 sesterces were enrolled in the eighteen centuries, while the tribuni aerarii were defined by some lower wealth bracket.¹¹⁴ If the public horse had become symbolic, the cap of one hundred citizens in each equestrian century, which once determined eligibility for the horse stipend (aes equestre), had possibly been lifted.¹¹⁵ However, it does not follow that everyone with over 400,000 sesterces was automatically enrolled and allowed to cast votes with the nobility in the centuriate assembly. The equestrian order had clearly grown larger than 1,800 citizens, but still constituted an effectively hereditary aristocratic honour with privileges and insignia that could have distinguished it from the order of tribuni aerarii.¹¹⁶

The fact that the tribuni aerarii were the wealthiest citizens of the prima classis is consistent with their liability for infantry pay (aes militare) in the tribes before 167 b.c.e. Excluded from the senatorial and equestrian orders in the first century, their identification as plebeians has its corollary in the historiography of the struggle of the orders in the early Republic.¹¹⁷ In the course of Roman state formation, paying infantrymen conscripted from the tribes presumably gave the wealthiest men leverage over the patrician nobility.¹¹⁸ Nicolet has argued that the tribuni aerarii acted as ‘prepayment’ liturgists, as in classical Athens.¹¹⁹ We need not push this comparison too far, but it was common for Greek poleis to use wealthy citizens as intermediaries for collecting property taxes, which in turn strengthened their claims of leadership.¹²⁰ As we have seen, the centuriate assembly was reformed after 241 b.c.e., boosting the influence of the non-equestrian prima classis, which had undoubtedly played an important role in financing the First Punic War.

The centuriate assembly's legislation and elections empowered the magistrates and senate with their authority. The legitimacy of the state, therefore, depended on the justice of the census. At several points in the narratives of Dionysius and Livy, including the description of the Servian reforms, Roman oligarchs claim an entitlement to their control of public offices by pointing to their greater fiscal contributions. The fact that Livy and, with more hesitation, Dionysius endorse this claim does not prove that it was uncontentious, especially for the lower classes marginalised in Roman politics.¹²¹ From 167 b.c.e., however, even this justification melted away, as the institution of the census became increasingly irrelevant to military conscription and taxation, precipitating what Jérôme France calls ‘la crise de la cité contributive’.¹²² The Gracchi and their political heirs proposed ambitious state policies to address social problems and promote social welfare with the public revenue, using Servius Tullius as a model or being used themselves as a model for his kingship. As tribunes of the plebs, their authority came not from the property-weighted centuriate assembly but the relatively democratic tribal assembly. Dionysius’ sources apparently projected the legitimacy crisis back in time and resolved it with various compromises in the conflicts between patricians and plebeians over the Servian reforms and trial of Coriolanus.¹²³

For Cicero and the optimates of the late Republic, the fate of the Gracchi was a prime example of the danger of magisterial power conferred by the people on equal terms, without the distinctions of property enshrined in the census. It was Tiberius Gracchus’ claim that the people were sovereign and could remove an obstructionist tribune that provoked his assassination, rather than his agrarian legislation per se.¹²⁴ Public land was, after all, theoretically at the disposal of the people and Gracchus had proposed to compensate the possessors with imperial revenues, which were poised to surge after the inheritance of the Attalid kingdom in 133 b.c.e.¹²⁵ There was an obvious model in Stoic philosophy in the equitable redistribution of Aratus of Sicyon discussed in Section III. Nevertheless, the senate was suspicious of the powers that would fall to the land commission and Gracchus himself, whom they feared would become a king like the usurper Servius Tullius. Moreover, Cicero would later argue that public land was no longer really at the disposal of the people because those who had appropriated it, including Rome's allies in Italy who were still in possession of territories annexed by Rome, obtained by their long occupancy legal rights tantamount to private ownership.¹²⁶ If the distribution was just according to natural law, then for Cicero it was illegitimate for civil authorities to diminish it.

The Roman philosopher Gaius Blossius used Stoic arguments to support his friend Tiberius Gracchus and the redistribution of land. He remarked that whatever Gracchus did, even if it was to burn down the city of Rome, would be justified because he was acting only in the interest of the people. After Tiberius’ death, Blossius fled to Asia Minor and joined the revolt of the ‘citizens of the sun’ with its radically Stoic proclamation of human equality.¹²⁷ Yet the Gracchi did not propose to levy taxes on citizens’ property in Italy. There was broad consensus across the political spectrum that Rome's imperial resources were sufficient to meet its public expenses, so the debate was rather whether Rome should levy higher taxes on provincials and do more to curb the rent-seeking behaviour of its governors and tax farmers. Even the duty of propertied citizens to perform military service as compensation for a state that served their interests, which the Gracchan reforms sought to keep viable, fell away by the end of the second century b.c.e., as generals began recruiting poorer citizens with promises of money and land. As the philosophers Liam Murphy and Thomas Nagel point out, ‘any convention that is sufficiently pervasive can come to seem like a law of nature — a baseline for evaluation rather than something to be evaluated’.¹²⁸ Tax immunity came to seem natural to Roman citizens, becoming entrenched by the passage of time.

Its fiscal–military purpose obsolete, the census fell into decline in the late Republic. Censors were appointed and had other functions but repeatedly failed to carry out the census. Though scheduled every five years, all but one (in 70 b.c.e.) between 97 and 23 b.c.e. was incomplete. The census had always been a public ritual: the citizens appeared by tribe and were called out alphabetically to give public declarations before the censors’ adjudicators, either on Rome's Campus Martius or in their local municipality. The grant of citizenship to large numbers of Italians after the Social War must have weighed down the pomp and practicality of the census, turning it into an unwieldy bureaucratic exercise.¹²⁹ The census was still highly valued by elites in theory as a foundation of the Roman oligarchy, and features in Cicero's ideal constitution in De legibus, but was no longer scrupulously carried out.¹³⁰ Already in the late second century b.c.e. strict rules about attending in person were being relaxed. By Cicero's time it was commonplace for the wealthy to send proxies to make declarations. One reason for this decline, Rene Pfeilschifter argues, was that the late republican census had lost its egalitarian ritual dynamic. Whereas the rich and powerful disdained to stand in line with the commonfolk, the latter must have witnessed only the inequality of wealth and status on display.¹³¹ Without compensatory taxation as its justification, the pageant was onerous and embarrassing: a reflection of nature's unfairness, not civil harmony.

V THE FISCAL CRISIS OF 43 B.C.E.

When Cicero wrote De re publica (c. 54–51 b.c.e.), tributum was still no more than a theoretical possibility, hardly worth mentioning in his account of King Servius Tullius.¹³² When he wrote about it in De officiis (autumn 44 b.c.e.), on the other hand, the situation was ominous.¹³³ As dictator, Julius Caesar had seized the public treasury, combining it with his own patrimony. After the Ides of March, the senate conducted an investigation to recover public money, while Octavian struggled to obtain his inheritance from Antony or sell his properties due to lawsuits, as he sought to furnish cash gifts to citizens promised in the dictator's will and raise an army.¹³⁴ By the end of 44 b.c.e. Caesar's heir had found an unexpected ally in Cicero, who persuaded the senate to support him against Antony. With an empty treasury and limited access to provincial revenues, the senate voted in 43 b.c.e. to levy tributum in Italy for the first time since the early second century.¹³⁵

Cicero's support for this measure proves that he was not opposed to taxing citizens’ property in principle, as long as it was necessary to save the state in an emergency.¹³⁶ He made one last desperate attempt to rally wealthy citizens (the boni viri, many of whom had made their peace with Caesar's clement dictatorship) behind the constitution by painting Antony as a threat to their property as well as their freedom.¹³⁷ Cicero's tributum set a precedent for the triumvirs, Antony, Octavian and Lepidus, who killed him by proscription after they joined forces against the senate and Caesar's assassins in November 43 b.c.e. Once in control of the government, the triumvirate's taxes were deeply unpopular but also pointed a way out of the fiscal crisis, towards an autonomous tax state.

In June 43 b.c.e., Cicero wrote to his friend Cornificius, proconsul in Africa, that the public treasury was almost empty and tributum was unavoidable, owing to the great expense of paying the ‘soldiers who deserved so well’ (optime meritis militibus).¹³⁸ Cicero's nod to the principle of fairness and the situation's urgency reflects the argument in De officiis (2.74) that administrators of the state must make every effort to convince the taxpayers of the necessity of tributum. The taxes in question are described in greater detail by the third-century c.e. Greek historian Cassius Dio: all contributed a 4 per cent property tax; senators also paid four obols (ten asses) per roof tile for houses owned or rented in Rome; ‘the very wealthy’ (οἱ πάνυ πλούσιοι) paid some additional tax; cities and artisans manufactured arms without pay.¹³⁹

These contributions were given willingly by those who favoured Caesar and hated Antony; but the majority, being burdened alike by the campaigns and the taxes, were annoyed, particularly because it was doubtful which of the two would prevail, and yet quite evident that they would be slaves of the conqueror.¹⁴⁰

Cassius Dio echoes none of Cicero's partisan concern for Octavian's deserving soldiers. The citizens as a whole stood to lose no matter which side won. The supplementary tax on ‘the very wealthy’ presumably refers to the 1 per cent tax that Cicero mentions after tributum in a letter to Brutus in July, 43 b.c.e.¹⁴¹

However, unless I am perhaps mistaken, our knottiest political problem is the shortage of money. The good men (boni viri) become more obdurate (obdurescunt) every day at the mention of a property tax (tributum). The proceeds of the 1 per cent, due to the shameless property declaration (impudenti censu) of the wealthy (locupletes), are entirely absorbed in the bounties of the two legions. Yet limitless expenses hang over us both for the armies which are now defending us and for yours.¹⁴²

Cicero's frustration is evident. We do not know how tributum was assessed and collected but the prospects were dim as long as the boni viri were ‘obdurate’. His political theory and plan of action were put to the test because the Republic had seemingly no effective way of taxing them without their consent and cooperation. As for the 1 per cent, a property declaration (census) was required from the wealthy (locupletes).¹⁴³ Cassius Dio claims that the triumvirs in 42 b.c.e. imposed an even larger 10 per cent charge on the wealthy, this time including ‘not only (οὐ μόνον) senators and equestrians, but also freedmen, male and female, whose names were inscribed on tablets (λευκώματα)’.¹⁴⁴ If he means the same tax at a different rate, it implies that only senators and equestrians paid the 1 per cent in Cicero's letter of 43 b.c.e. Whereas the triumvirs in 42 b.c.e. confiscated property for false declarations (τιμήσεις), apparently the most Cicero could do in 43 b.c.e. was condemn their shamelessness.

The tablets were possibly understood as a sign of honour and allegiance to the Republic. Appian implies that Cicero maliciously entered the names of Antony's supporters but adds that many of them paid the tax right away precisely to avoid that slanderous accusation.¹⁴⁵ There is a striking contrast between the wealthy in Cicero's letter with their ‘shameful property declaration’ (impudenti censu) and the virtuous patricians in Livy's narrative who paid ‘most faithfully according to their property declaration’ (summa fide ex censu) when tributum was allegedly levied for the first time for the siege of Veii in 406 b.c.e. While the plebs at first rejoiced and praised the senate for offering to pay soldiers from the public treasury, the tribunes of the plebs warned that a property tax would be an insufferable burden on all citizens and offered to protect anyone who refused to contribute.¹⁴⁶

The Fathers had made a good beginning and persevered in supporting it. They were themselves the first to contribute, and since there was as yet no silver coinage, some of them brought uncoined bronze in wagons to the treasury, and even made a display of their contributing. After the senators had paid most faithfully according to their property declaration (summa fide ex censu), the chief men of the plebs (primores plebis), friends of the nobles (nobilium amici), began to pay according to what was agreed (ex composito). When the crowd saw that these men were applauded by the patricians and were looked upon as good citizens (boni cives) by those of military age, they quickly rejected the protection of the tribunes and vied with one another who should be the first to pay.¹⁴⁷

Livy has the senators going immediately to deposit their uncoined bronze in the central treasury (aerarium), for ‘there was as yet no silver coinage’. This may have been what the wealthiest citizens inscribed on the tablets were expected to do in 43 and 42 b.c.e. (and perhaps 210 b.c.e.) but it was not how the common citizens would have paid tributum. The group that Livy describes as the ‘chief men of the plebs’, ‘friends of the nobles’ and ‘good citizens’ were possibly the tribuni aerarii, the wealthiest men of the prima classis. Their payment ‘by agreement’ (ex composito) rather than ‘by declaration’ (ex censu) may allude to their role as intermediaries, who collected taxes from citizens in the tribes (see Section IV).

Livy reports that in 210 b.c.e. the senators competed to be first ‘on the public tablets’ (in publicis tabulis) for their payment, setting an example for the equestrian order and the plebs to pay their taxes ‘without coercion’ (sine coercitione).¹⁴⁸ As noted in Section II, Livy's narrative of the Servian reforms and the speech of the consul Laevinus imply that the gradation of worthy standing and fortune (gradus dignitatis fortunaeque) in the census was justified by the wealthy citizens’ financial contributions, though their control of the state enabled them to reimburse themselves with public money or other advantages in the long run. Livy presumably derived these ideas from sources predating the suspension of tributum in 167 b.c.e. However, the moral examples had special rhetorical potency after the fiscal crisis that ended the Republic. Livy, who lived from 59 b.c.e. to 17 c.e., experienced the taxes and related obligations of the triumviral period (for example, a naval requisition of slaves just like the one in Laevinus's consulship), which loaded his narrative with vivid impressions and significance.¹⁴⁹

Whereas tributum before 167 b.c.e. relied on the cooperation of the wealthiest citizens, which was contingent on their control of the state, the triumvirate could enforce compliance by the threat of proscription and confiscation. Hortensia, daughter of the famous orator Hortensius, went with a crowd of women to protest before the triumvirs in the forum in 42 b.c.e. The wealthiest 1,400 matrons had been added to the tablets and ordered to furnish property declarations for taxation. The triumvirs desisted from dispersing them by force amid a supportive crowd, and compromised by scaling back the number liable. In the speech attributed to her, Hortensia railed against the injustice of forcing women to pay property taxes even though they had no share in government.¹⁵⁰ The tenor of the speech, which may have been embellished by Appian or another author, accords with Livy's historical examples of fiscal fairness.¹⁵¹ There was no precedent for taxing women, she argued, not even during the crisis of the Second Punic War, when women made voluntary contributions for fighting foreign enemies, which they refused to do for a civil war. In contrast to their ancestors, they were being coerced and had to fear informants who would challenge their declarations for bounties. The dignitas attached to their material wealth was being robbed from them.¹⁵²

In the late Republic, just as immunity from taxation came to seem natural, the security of property rights became increasingly precarious. Cicero, as we have seen, attributed this insecurity to the moral corruption of ambitious nobles, who overrode constitutional procedure and acted in their particular interests rather than those of the state.¹⁵³ Civil war is admittedly a precarious situation, when powerholders give short-term interests priority. Even when victorious, however, Rome's structural inability to tax its citizens or redistribute land equitably left those in control of the government with limited choices: either devise new methods of taxation or resort to violent proscriptions and confiscations of property. Because the triumvirs, Cassius Dio writes, ‘needed enormous sums of money and had no other source from which to satisfy the desires of their soldiers, they affected a kind of common enmity against the rich’.¹⁵⁴

Despite their unpopularity, the fiscal innovations of the period 43–31 b.c.e. pointed towards a way out of this crisis.¹⁵⁵ The absence of an up-to-date census possibly increased the risk of miscalculation and the sense of unfairness.¹⁵⁶ However, taxing the roof-tiles of senators in 43 b.c.e. borrowed a method of estimation attested elsewhere in Roman Italy that utilised the visibility of urban residences.¹⁵⁷ Cassius Dio suggests that the contributions (συντέλειαι) assessed on land and slaves in 42 b.c.e. were less onerous than the crushing 10 per cent tax on senators, equestrians and wealthy freedmen, whose names were published on tablets, in part because only they had to submit property declarations (τιμήσεις), which could be challenged by informers.¹⁵⁸ In addition to property taxes, other taxes on sales and residential rents apparently extended the fiscal burden even to the lower classes, which helps explain the intensity of popular unrest.¹⁵⁹ An edict of 40 b.c.e. introduced a percentual tax on inheritances, which may have been planned already by Julius Caesar.¹⁶⁰ After defeating Sextus Pompey in the west in 36 b.c.e., Octavian ceremoniously abolished tributum and other taxes in Italy, though he later imposed a tax on the fortunes of wealthy freedmen for his final campaign against Antony in 31 b.c.e.¹⁶¹ As emperor Augustus, he looked back to these earlier innovations for more stable and equitable sources of revenue for his professional army. The inheritance tax (vicesima hereditatium) and the sales tax (centesima rerum venalium) were made permanent for the new military treasury in 6 c.e.¹⁶² These better aligned the emperor's and army's fiscal incentives with the protection of hereditary wealth and market exchange.

The Servian reforms and tributum before 167 b.c.e., as related in the narratives of Livy and Dionysius of Halicarnassus, took on new significance for their readers in the Augustan age. Wealthy citizens were paying more taxes, while the emperor ensured conditions of peace and stability. In comparative historical terms, one can still speak of Rome as a tributary empire rather than an autonomous tax state. What autonomy the emperor possessed was due mainly to his control over the armies, the treasury and his patrimonial domain, rather than any constitutional authority. From a provincial perspective, Rome could adjudicate disputes and assign fiscal burdens on provincial subjects with relative impartiality. On the other hand, its fiscal autonomy was compromised by a minimal bureaucracy, dependent on local city governments controlled by property owners whose fiscal cooperation was predicated on rent-extraction that limited tax revenue. While tributum could be collected from rich provincials, including Roman citizens in the provinces, massive concentrations of wealth in Italy were shielded from direct taxation.¹⁶³

Cassius Dio penned his own Stoic-inspired proposal for turning Rome into an autonomous tax state in the early third century c.e. Put in the mouth of Augustus's advisor Maecenas in dialogue with Agrippa in 29 b.c.e., it shows an affinity with the Stoic legal philosophy of Dio's contemporary Ulpian (c. 170–228 c.e.) and Caracalla's citizenship edict (212 c.e.).¹⁶⁴ The emperor's patrimonial land should be sold, so buyers would have an incentive to improve it, and cheap government loans should be offered to stimulate investment, while anything that yields profit should be taxed, so that the revenues could sustain a standing army and everything that contributed to a well-governed state (πόλις). ‘For,’ Dio's Maecenas claims, ‘it is only just and proper (δίκαιον καὶ προσῆκoν) that no individual or people should have immunity from taxes (ἀτελῆ) as long as they enjoy the advantage (ὠφελία) from them as much as the rest’.¹⁶⁵ This spelled out what was unthinkable in Cicero's time: Romans in Italy paying compensatory taxes for the benefits of the state no differently than provincials. Agrippa's republican speech against an autonomous tax state, by contrast, advocated only voluntary contributions to the treasury, corresponding to Cicero's conception of the just city.¹⁶⁶

VI CONCLUSION

In Cicero's just city, the laws were walls for the protection of property. King Servius Tullius laid those walls’ foundations in Rome with his census. It entrenched the presumptively just ‘pre-fiscal’ distribution of property in the centuriate voting system and thereby aligned the interests of the populus with the supposedly ‘best form of government’. Though Cicero distinguishes between the people and the administration of government, his emphatic rejection of majoritarianism sets him apart from modern theorists of popular sovereignty, including Locke. He was an advocate of rights protecting citizens from state interference and theorised a constitution that would secure them; but his solution, I have argued, compromised the state's impartiality.

There is a normative dimension to this argument because fiscal fairness appears to require compensatory taxation. Murphy and Nagel observe that only a few libertarian philosophers would claim that the state has no legitimate right to interfere with private property and levy taxes. However, Lockean (or Ciceronian) assumptions are often carried over into everyday arguments, whenever the distribution of pre-tax incomes, property holdings, or market outcomes are treated as a baseline for evaluating fairness in terms of taxpayers’ ability to pay.¹⁶⁷ Murphy and Nagel counter that property has no independent existence anterior to the fiscal–legal framework of the state. The justice of taxation has to be evaluated according to the system of property rights and entitlements that it sustains.¹⁶⁸ Incomes and property values depend largely on the state's policies, which inevitably benefit some individuals more than others. Differential tax rates or exemptions are an instrument for an impartial state to compensate citizens who are disadvantaged by certain policies in order to maintain distributive justice.¹⁶⁹ This perspective establishes a frame of reference by which to assess the strengths and weaknesses of theories and institutions in antiquity.

Cicero's ideas were contentious in his own time and competed with other solutions to the crisis of the late Republic. In Section II, Cicero's account of King Servius Tullius was contrasted with the compensatory justifications for taxation in Livy and Dionysius. The three present different standards of fiscal fairness. The speeches in Dionysius emphasise the king's impartiality: those who benefit most from public protection must compensate those who benefit least. Even in Livy, property was not ‘pre-fiscal’, as it was for Cicero, because contributions to the public treasury allegedly justified unequal voting rights and aristocratic leadership. In Section III, it was argued that Cicero's De officiis offers a distinctively Roman and late republican deviation from Stoic philosophy. Greek Stoicism's emphasis on human equality and communal ownership were consistent with compensatory taxation and redistribution. That is not to say that all Romans thought alike. Section IV attempted to explain Cicero's originality by situating him in his historical context. The end of tributum in 167 b.c.e. and changes in military conscription undermined the legitimacy of the centuriate assembly. It was vulnerable to competing claims of authority from popular tribunes and military leaders, who could draw on democratic and monarchical models.

Apart from its philosophical interest, fiscal fairness has an important historical dimension. The political scientists Scheve and Stasavage argue that compensatory arguments similar to those described by Murphy and Nagel were used effectively by progressive tax-reformers in the twentieth century. The First World War engendered a discourse on fairness wherein wealthy citizens acknowledged the greater sacrifices of poorer citizens. The former, who had more money and material goods at stake, could be persuaded that it was their duty to contribute a larger share of their income. Scheve and Stasavage admit that democracy may facilitate taxation, but their data suggest that mass mobilisation in warfare had a more significant impact on tax rates than the degree of democratisation.¹⁷⁰ As we saw in Sections IV and V, Rome regularly levied tributum for military campaigns before 167 b.c.e., even without democracy, and again for the crisis of 43 b.c.e. The prolonged tax immunity during the intervening period weakened the state's ability and incentive to specify and protect property rights. Compensatory taxation arguably enhanced social cooperation before 167 b.c.e. Its elimination made property redistribution, which without adequate compensation turns into violent expropriation, the principal alternative once the spoils of empire failed to resolve distributive conflicts among citizens.

Cicero justifies the Servian census as the foundation for a constitution credibly committed to the protection of property rights at a time when they were becoming increasingly precarious. The fiscal crisis after Caesar's assassination in 44 b.c.e., examined in Section V, reveals the limitations of Cicero's political theory and the weakness of the oligarchical state. Rome lacked any functional equivalent of the Rawlsian ‘veil of ignorance’, under which equal citizens might impartially agree, as a matter of constitutional choice, to impose compensatory taxes on winners to ensure the cooperation of losers in defence of the state.¹⁷¹ Voluntary contributions by the wealthy were contingent on the state's partiality to their particular interests, which the power of Antony and Octavian subverted. Readers of Dionysius and Livy must have paid close attention to the compensatory justifications for tributum in the past after the fiscal innovations of the triumvirate and Augustus. The emperor could look back to the wise King Servius Tullius as a model for Rome's renewal. However, his autocratic solution came at a cost to the constitutional foundation of citizen rights.