Polybius and the Roman Political System

No study of Roman politics can ignore Polybius (c. 200–120 BCE), not just because of his status as a contemporary source, but also because of the lasting impact his approach has had on subsequent – ancient and modern – analyses of the Roman constitution. A Greek statesman from Megalopolis, Polybius was exiled to Rome in 167, where he spent the following seventeen years, developing close ties with the leading men of the time. During his exile he began work on a monumental history of Rome that would trace her conquest of the Mediterranean world in forty books, the first five of which survive intact, the rest only in fragments of varying length. The work is remarkable for its ‘factual’ style of reporting, which makes it a prime example of pragmatic, ‘didactic’ historiography.Footnote ⁴ The approach is analytical rather than rhetorical, always looking for general causes behind historical occurrences. Polybius’ discussion of the political system, presented in the fragmentary book six, offers the first, indeed only, original attempt to subject Roman institutions to theoretical analysis. Polybius’ impact can hardly be overestimated and his legacy arguably lives on to the present day; many scholars still regard Polybius’ model as the best guide to understanding the Roman constitution.Footnote ⁵

The central tenets of Polybius’ analysis are well known and need only be broadly outlined. Essentially, the Roman political system is presented as the embodiment of the ‘mixed constitution’, blending elements of monarchy, oligarchy, and democracy into a single, well-balanced entity.Footnote ⁶ The internal stability ensured by the mixed system allowed Rome to direct her energy outwards towards military expansion, thereby explaining her remarkable drive and success. According to Polybius, the purpose of his work was to explain Rome’s conquest of the civilised world to a Greek audience – although he may also have hoped to impress his Roman hosts. In antiquity constitutions were generally assumed to hold the key to the success or failure of states since they determined their degree of stability and upheaval, freedom and tyranny, and it was this linkage that provided the rationale for Polybius’ constitutional digression. Moreover, given the traditional Greek ideal of the golden mean, a ‘mixed’ constitution that blended different types of government and balanced opposing interests against each other was – in analogy to the doctrine of the bodily humours – naturally deemed the best, and the only guarantee of long-term stability.Footnote ⁷ Therefore, since Rome’s military superiority was indisputable, it followed logically that it must have a matching superior constitution, which from a Greek perspective meant a mixed constitution.

This premise lends Polybius’ analysis a certain aprioric aspect but, as importantly, the particular version of the ‘mixed constitution’ he presents is in many respects unusual, not just in its application of Greek theory to Roman institutions but also in the way he uses his theoretical models. Polybius’ basic framework was provided by the traditional tripartite division of constitutions into monarchy, democracy, and oligarchy that can be found already in Herodotus’ constitutional debate, and even earlier in Pindar.Footnote ⁸ Later it became a standard fixture of Greek political thinking and theorists further refined the model by dividing each of them into good and bad forms. The terminology and categories varied, but in Polybius’ work the positive forms appear as monarchy, aristocracy, and democracy, and their negative versions as tyranny, oligarchy, and ochlocracy (mob rule). The constitutional archetypes were considered to be in a permanent state of flux, invariably degenerating into their negative forms and causing upheavals that would lead to their eventual replacement by another type of government. In Polybius’ model the different types succeeded each other in a fixed cyclical movement, the anakyklosis, which represented a variant of the Platonic idea.

Faced with the problem of endemic political instability Greek theorists came up with a number of solutions. In Plato’s ideal constitution the state was ruled by an enlightened guardian, whose benevolent reign would overcome the conflict between rich and poor. Later he moved towards a mixed system which combined elements of the different constitutions. Aristotle, on the other hand, while in principle accepting Plato’s elevation of kingship, also developed a less utopian ideal, the so-called Politeia, which blended oligarchic and democratic features. Although presented as the perfect form of democracy, Aristotle’s ‘polity’ was in practice closer to a moderate oligarchy. This constitution was intended to bring an end to political upheaval and class conflict by offering a measured compromise between the interests of the elite and the populace. In Aristotle’s analysis the polis was naturally torn between the well-to-do (euporoi) and the indigent (aporoi). By focusing on the structural causes of political instability and stressing the significance of economic inequality, Aristotle’s solution thus remained firmly rooted in the social reality of contemporary Greek poleis. The conflict between rich and poor was to Aristotle the root cause of instability, and his solution involved a broad compromise between the two groups and the application of balanced, bipartisan policies that respected the concerns of both sides.

Polybius’ ‘mixed constitution’ is a very different creature from Aristotle’s ‘Polity’. To Polybius the anakyklosis did not reflect the instability caused by the competing interests of different social groups but was entirely ethical in nature. It was brought about by the moral corruption of the rulers which inevitably followed within ‘pure’ systems, leading to their overthrow and replacement. Thus, as Nippel observed, with Polybius Greek constitutional theory lost its sociological dimension, and it was this analytical shift which enabled Polybius to combine all three ‘positive’ archetypes, kingship as well as aristocracy and democracy.Footnote ⁹ The inclusion of monarchy altered the character of his model, for while monarchy evidently existed as a constitutional type, it did not reflect the rule of any particular group in society in the same way that democracy and oligarchy did. Monarchy is in a sense the primitive default option that emerges when an open, participatory political process has broken down and power is left in the hands of a single individual. To Aristotle monarchy was therefore essentially tyranny, although he accepted that the enlightened rule of one man, defined as kingship, in theory represented an ideal and possibly supreme form of government. While elite and masses were the given constituents of any ancient society, whose – self-interested – rule was conceptualised as oligarchy and democracy, respectively, there could in the nature of things be no specific constituency behind a monarchy. Polybius’ notion that an effective ‘mixed constitution’ required a monarchical element is therefore based on a mechanistic construction of the problem. Kingship was included in order to create a correspondence between the perceived problem – the constitutional cycle – and the solution – the ‘mixed constitution’, which combined all the ‘pure’ forms and neutralised their individual weaknesses. Or, in other words, since monarchy was part of the problem – the anakyklosis – it must also be part of the solution, i.e. the ideal constitution. This schematic logic, however, remains divorced from socio-economic considerations of the causes of political instability.

In this context the example of Sparta, the classic Greek paradigm of a ‘mixed’ constitution to which Polybius explicitly – and favourably – compared Rome, becomes significant. Sparta’s unique combination of monarchy, democracy, and oligarchy meant that in order to carry that comparison through a monarchical element had to be found also in Rome. That was obviously difficult in a system founded on the explicit disavowal of all things regal, and Polybius’ attempt at identifying kingship in the role of the Roman consuls remains weak. It ignores a number of features: that there was more than one consul, that they held equal powers, were elected by the assembly for one year only, held no legislative powers, answered to the senate which could issue instructions, and might be held responsible for their actions after their time in office. Some Romans, as we shall see, may have interpreted the consuls as the inheritors of the king’s imperium, but that did not make their role monarchical, especially since they also saw the office as deliberately conceived in opposition to the ousted kings.

Polybius’ ‘monarchical’ consuls highlight the mechanical nature of his approach, which shows little direct engagement with the nuts and bolts of the Roman institutions and their practical functioning. Each element of the constitution is identified as the embodiment of one of the conventional Greek archetypes. Thus, while the consuls are made to represent kingship, the assembly is labelled the democratic element and the senate the oligarchic. That approach differs radically from that found in Aristotle, for whom the basic distinction in any society was that between the wealthy who had sufficient means to take active part in government and those without. Oligarchy was therefore defined as the rule of the well-to-do, while democracy existed when poorer people held power despite their lack of resources. Crucially, Aristotle did not operate with an either/or distinction, since each type represented a continuum of constitutional forms; Aristotle envisaged no fewer than four degrees of oligarchy.Footnote ¹⁰ The character of a constitution depended on a number of variables, of which the most important were property assessments (timema) for citizenship and office holding, the level at which the thresholds were set, the provision of payment for officials and the degree of access to the assembly. In addition, Aristotle’s distinction between oligarchy and democracy took into account the general distribution of wealth in society, the rule of law, and the extent of inherited powers.

None of these factors are considered by Polybius, who simply attaches to each institution a constitutional label that sums up its political ‘value’ and defines it as the incarnation of a particular constitutional principle. According to Polybius their political character was so unequivocal that when each element was viewed in isolation, the constitution appeared either to be fully democratic, monarchic or aristocratic. Polybius therefore does not allow for the possibility that they might contain internal contradictions or compromises, thus ignoring what Nippel called ‘Intraorgankontrolle’, the internal control mechanisms that play such a prominent role in the models of Plato and Aristotle.Footnote ¹¹ There is no hint that assemblies, magistracies, and councils might themselves cover the full range from democratic to aristocratic. Thus, we search in vain for a discussion of the property qualifications which determined the structure of the assemblies, access to public office and membership of the senate. Polybius seems unaware of the fact that councils, assemblies and magistracies had no definite political character but could fall at any point within a broad continuum of popular and aristocratic politics.

We might therefore ask whether Polybius’ Roman constitution was not a ‘composite’ constitution rather than a ‘mixed’ one, combining elements of different political systems which remain essentially separate. However, this is precisely where Polybius makes his original contribution to ancient political theory. Since each element was ‘pure’, the stabilising ‘mixture’ lay entirely in the interrelation between individual institutions, whose powers were moderated through a system of ‘checks and balances’. The interdependency of the Roman institutions represents the central tenet of Polybius’ analysis, which argued that none of them could act independently but had to rely on the co-operation of the others, a co-operation that was forced upon each element and essentially was motivated by self-interest tempered by fear.Footnote ¹²

Demonstrating this interdependency in practice is not without difficulty, however, and many scholars have noted the problems Polybius faced when trying to apply his thesis to the specifics of the Roman political system. It is, for example, hard to accept the consuls as counter-balancing the senate, since consuls and senators all belonged to the same socio-economic class; indeed the consuls were themselves members of the senate and would remain so after their tenure. Some of the interdependencies claimed by Polybius also seem forced, not least the people’s reliance on the senate, which exaggerates the importance of public contracts and the proportion of citizens sharing in them (6.17).

Examples such as these cast doubt on the assumption that the analysis grew out of Polybius’ own personal observations of Roman politics and his close acquaintance with leading protagonists. Polybius’ constitutional digression should in this respect be treated as distinct from his ‘pragmatic’ historical reporting. It is perhaps better understood as an intellectual ‘tour-de-force’ which served to divert and impress his readership – and rival historians – with a dazzling display of political theorisation. Polybius embraces the challenge of making Roman institutions comprehensible to a foreign audience that would often have found them baffling, and he does so by fitting them into a conventional Greek theoretical framework. As a result the unfamiliar becomes familiar – to the Greeks as well as to us; Rome emerges as the perfect embodiment of the long-standing Greek ideal of the mixed constitution, able to take its rightful place alongside Sparta.

The question is where this leaves the study of the Roman political system. Polybius’ analysis is too schematic to be of much help in understanding the location of power in republican Rome, and despite recent attempts to ‘rehabilitate’ his model, most scholars have recognised its basic shortcomings.Footnote ¹³ There is, however, a much wider legacy of Polybius’ work, and to grasp that we may shift our focus from the specifics of his analysis to his overall approach; for what was truly innovative in Polybius’ constitutional digression was his application of Greek concepts to Roman institutions. This example of ‘interpretatio Graeca’ has raised few eyebrows, probably because most modern scholars consider these analytical tools to be universally applicable. But as Ostwald noted, the classification of constitutions as the rule of one, few and many, is now so conventional, ‘that we tend to forget how uniquely Greek it is’.Footnote ¹⁴ The concepts employed by Polybius – and indeed his entire analytical framework – were developed in a Greek context and shaped by experiences specific to the Greek polis. Given their cultural contingency, we may consider the possibility that they in fact impose an alien perspective that obscures what was uniquely Roman.

Polybius’ approach was basically a functional one that perceived Roman politics as a logical system, in which each element fulfilled a specific role in maintaining the all-important stability. As components of political machinery they were endowed with rational purpose as a means of negotiating power within the state, balancing the interests of different groups and upholding social peace. Polybius’ identification of a distinct political ‘value’ in each Roman institution has wide implications, for in the end it is this approach that has defined the comitia as essentially ‘democratic’ in their rationale.

Cicero’s Res Publica

Polybius’ analysis of the Roman state was that of an outsider, which raises the question how a Roman writer would apply the Greek models. This is where Cicero’s contribution becomes important, for in the De re publica and De legibus he offers a glimpse of a different conceptualisation of Roman institutions.Footnote ¹⁵ Unlike Polybius, Cicero was not a detached foreign spectator, but a Roman aristocrat with long experience in dealing with the institutions he describes. Although he builds on the same Greek theoretical foundations as Polybius, whom he acknowledges as an authority (Rep. 1.34), and uses identical models and categories, there are slight but significant shifts in emphasis and the addition of crucial new accents.Footnote ¹⁶

The De re publica, written between 54 and 51 during Cicero’s self-imposed withdrawal from the political scene, belongs to a period of upheaval and instability in Rome. The text has a clear remedial function, exploring possible solutions to the disunity and instability that currently threatened the republic. It therefore presents a positive and uplifting vision of the res publica, with particular stress on the importance of consensus, co-operation and unity of purpose as the original impulses that had first led to its creation.Footnote ¹⁷

Cicero’s description of the constitutional types and the advantages derived from a balanced mixture does not depart significantly from his Greek sources, but the overall conception of the constitution is nevertheless quite different. He abandons the idea of opposing constitutional elements locked in perpetual competition over power and resources and kept in equilibrium only through mutual fear. Strikingly, he uses a Platonic image of dissimilar sounds creating musical harmonies to illustrate the distinctly Roman consensus he envisages. Importantly, this is a harmony between social classes held together by common accord, not distinct constitutional forces or principles.Footnote ¹⁸ The resulting concordia is therefore very different in nature from Polybius’ checks and balances.

In line with conventional Roman thought and political practice, Cicero readily accepts the people’s right to libertas and stresses its overall centrality to the res publica, but it was also self-evident to him that the elite would provide leadership. He therefore introduces a concept absent from his Greek models, namely senatorial auctoritas, the pervasive influence that went far beyond what was formally enshrined in law (Rep. 2.57). Essentially therefore, Cicero’s constitution is both aristocratic and bipartite, with the populus holding the power but exercising it on the advice of experienced councillors and delegating executive functions to appointed magistrates.Footnote ¹⁹

Thus, in contrast to Polybius the central virtue of the mixed system was to Cicero its ability to satisfy the people’s desire for libertas without compromising the power of the elite or the strong magistracy. But because the basic principle of the people’s libertas was beyond contention, its precise articulation became much more crucial. A key concept was aequabilitas – equity, equality or fairness – which represents the particular kind of equality that is proportional to merit and thus responds to the fundamental aristocratic objection against the egalitarian principle which is its failure to take into account naturally occurring differences in status or ability.Footnote ²⁰

Cicero’s other constitutional treatise, the De legibus, goes into greater detail with regard to the practicalities of politics and here the discrepancies between his and Polybius’ approach become even more apparent. The popular tribunate is presented as a means of pacifying the people, a necessary concession which ensures that the common people make no desperate attempt to assert their rights (Leg. 3.24–5). Cicero further outlines a procedure designed to give voting procedures the ‘appearance of liberty’, ‘libertatis species’, a point echoed in his praise for the republican hero Publicola, who by granting the people a moderate amount of liberty had entrenched the power of the elite.Footnote ²¹ Thus, the popular institutions and their prerogatives emerge as institutionalised guarantors and symbolic representations of the people’s liberty, serving to integrate the people into an aristocratic regime – without conveying any real power.Footnote ²² In Cicero’s version of the ‘mixed constitution’, institutions could exist to create a semblance of representation, being gestures of inclusion and shared citizenship rather than vehicles of actual popular influence.

Cicero’s model remains a negotiation between, on the one hand, the Greek conceptual framework he had adopted and, on the other hand, his personal vision and experience of the Roman state. As such it hints at a more complex and subtle understanding of institutions than Polybius’ model of formalised checks and balances. In particular, the suggestion that institutions could have symbolic functions may give us a clue as to how to resolve the paradoxes inherent in the Roman political system. Polybius’ functional approach has, as noted, encouraged a ‘normalisation’ of Roman institutions, but a better understanding of the Roman political system and the particular logic which governed it may be gained by focusing on those elements that do not fit easily into his model. In that context the ambiguous role of the populus attracts particular attention.

‘The Power of the People’: The Comitia in the Roman Republic

The Roman People was, as most textbooks explain, ‘sovereign’; only the populus could make laws, declare war and make peace, appoint leaders of the Roman state and condemn a citizen to death. The ‘sovereignty’ of the populus Romanus was implied already in the fifth-century law code, the Twelve Tables, which according to Livy stated that: ‘whatsoever the people decreed last should have the binding force of law’.Footnote ²³ The role of the people as the source of public legitimacy was undisputed; as Giovannini noted, the supremacy of the populus was, judging from our sources, never called into question.Footnote ²⁴ Certainly, in historical times the principle of the people’s ‘sovereignty’ was for all Romans the self-evident foundation of political life. The concept of popular ‘sovereignty’ is, of course, an anachronism in a Roman context, invented as it was in the early modern period.Footnote ²⁵ But Roman notions of the ‘power of the people’ may not only have differed from anything we would today recognise as ‘sovereignty’; it was probably also quite unlike the ‘democratic’ principles debated by ancient Greek theorists. In fact, the people’s ‘sovereignty’ may have been rooted in a very different construction of the Roman state. In order to investigate that issue we will look first at the ways in which the ‘will of the people’ was expressed and articulated in the popular institutions.

The classic Roman republic had a bewildering variety of popular assemblies, the comitia curiata, centuriata, tributa, and the concilium plebis, in addition to the non-decision taking meetings known as contiones. They were created at different moments in Rome’s history, and for different purposes. They did, however, share the same formal structure and procedures, particularly concerning the relationship between the assembly and the presiding magistrate. Political gatherings, needless to say, always require some form of organisation and leadership. In Rome, however, the role of the organiser went far beyond the strictly practical. The presiding official was in full control of every aspect of the proceedings. Unlike classical Athens, Rome had no statutory assembly days, allowing the people to come together on a regular basis to deliberate current issues. The various assemblies and meetings convened only when called by an authorised official. An assembly thus came into being through the actions of an official, who could dismiss it again at any time, thereby preventing it from reaching any decision.

The fact that it was the magistrate who constituted the people politically also determined the workings of the comitia. Again the contrast with Athens is striking, since there was no possibility of popular initiatives or proposals emerging from the meetings themselves, which were not even allowed an open debate. Discussions were formally separated from the decision-taking meetings (comitia), and relegated to so-called contiones, which were defined as formal public gatherings where no vote was taken. Contiones were called by an official, who held the required authorisation, the so-called potestas contionandi, and they might be held on their own or precede a comitia. But even these occasions did not offer the opportunity for a free or open debate. The proceedings were strictly regulated and remained under the direct control of the presiding official. It was he who decided not only the time and place, but also the issue(s) to be debated and crucially who were permitted to speak – usually arranged in advance.Footnote ²⁶ The contio was therefore more an address to the people than a consultation of it.Footnote ²⁷ There was no formal interaction between speakers and populus, just a one-way communication, which was fully controlled by the magistrate who convened the meeting, selected the speakers and could dismiss the crowd at any point. It follows that the Roman people had no legitimate ways of convening or expressing its views without formalised leadership. When this was not forthcoming the people could not voice any opinion; in other words, the populus did not exist as a political body independently of its leaders.

The essentially passive role of the populus is also reflected in the legislative procedure. The presiding magistrate would present proposals to the assembly, whose response was limited to yes or no without any opportunity to modify or change what was put before it. The lack of active input into policy-making means that a legislative comitia can be defined as an ad hoc meeting convened by a magistrate for the purpose of ratifying a specific proposal (rejections, as we shall see, hardly ever happened). The limitations imposed on the Roman comitia may not strike modern observers as that exceptional, perhaps because of the similarities between the Roman procedure and that of a modern-day referendum, where politicians put a simple proposition before the electorate for its approval or rejection – a process which is commonly perceived as a fair and democratic solution to the practical problem of consulting large numbers of people. The parallel is misleading, however; for the modern equivalent to a Roman legislative assembly is not a referendum but a meeting of parliament. The basic principle in contemporary political systems is that of representation, referendums usually being the rare exception. The modern popular assembly is therefore a parliament, and to appreciate fully the constraints imposed on the Roman comitia we must envisage a parliament with no regular meetings, no debates, and no ability to formulate or modify policies; in short, a parliament completely controlled by the executive.

The relationship between magistrate and assembly in Rome also differed fundamentally not just from that found in current political systems, but also in, for example, classical Athens. Unlike Athenian office holders, the magistrates were not simply officials in charge of specific public functions and accountable to the ekklesia. They were, as Meyer noted, superior to the populus, as indeed underlined by the etymology of their title magistratus, derived from magis – greater.Footnote ²⁸ The power of the people was expressed via – and only via – the actions of its elected officials. The unique position of the Roman magistrate was also underlined symbolically. Thus, his powers were embodied and directly expressed in the symbol of the fasces, which had no parallel in the Greek world.Footnote ²⁹ Moreover, the relationship between magistrate and populus manifested itself spatially in the fact that the magistrate presided over the assembly from an elevated tribunal, physically standing above the populus which gathered before him.

Given the extensive powers of the magistrates, the method used to appoint them becomes significant. The process is conventionally described as popular election, but the formal procedure is revealing as to the underlying logic of the occasion. Again we find the populus playing a peculiarly passive role. As Badian has demonstrated, the people were neither ‘sovereign’ nor active in the elective assemblies.Footnote ³⁰ They were convened by a magistrate, who formally guided the comitia, and the technical language used for the appointment of magistrates indicates that it happened through the joint action of the presiding magistrate and the populus.Footnote ³¹ Without the guidance of a magistrate the populus could not appoint new leaders. In this context it is significant too that the new appointment formally was made only when the leader announced it. This feature gave the process an element of internalised succession, where the new magistrate was ‘created’, ‘creatus’ by the incumbent rather than by the popular vote.Footnote ³² It was the magistrate not the people who was responsible to gods and humans for the succession and its consequences.Footnote ³³ When the chain between them was broken, an interrex had to be appointed to carry out the ritual handing over of power to the successor. In formal terms the occasion was therefore one where the outgoing magistrate presented his successor to the assembly, asking it to grant its approval.Footnote ³⁴

The procedure is commonly assumed to go back to regal times and originate in the traditional acclamation of the new king by the comitia curiata, which the ancient tradition identified as the oldest Roman assembly, predating the republic.Footnote ³⁵ Supposedly, the kings had received their powers through a lex curiata de imperio, which in effect was an oath of allegiance sworn to the new ruler, expressing the people’s consent.Footnote ³⁶ That interpretation is supported by the terminology used. Thus, it has been suggested that etymologically suffragium may refer to the sound made by an assembled crowd banging their weapons as a sign of approval. That would indicate a primitive act of acclamation by a crowd collectively responding to a request, rather than an elective process where votes were cast and counted.Footnote ³⁷ The political language of the Romans points in the same direction, since Latin had no word expressing the exercise of an active political choice by the citizens. The word for voting, vote, and voter – suffragor, suffragium and suffragator – are all positive terms of approval and support, not of choice. The vocabulary thus reflected a system where the populus could either express its approval or withhold it.

Later real elections offering a choice of several candidates were introduced, but there is no evidence that the assembly was then free to vote for whomever it wanted. As Badian concluded: ‘The comitia had no right of initiative’.Footnote ³⁸ The magistrate was still formally and in actual fact in charge, and he could refuse to accept a vote by the populus, as happened on several occasions.Footnote ³⁹ Thus, in 215 the consul Q. Fabius Maximus threw out the result of the first century, the centuria praerogativa, and asked it to reconsider its decision, which it duly did.Footnote ⁴⁰ In 184 the election of Q. Fulvius Flaccus as praetor was rejected, because he was already designated aedile.Footnote ⁴¹ Later, at the consular elections of 148, the presiding magistrate refused to acknowledge the election of Scipio Aemilianus, who was formally a candidate for the aedileship.Footnote ⁴² When on some occasions the magistrate backed down and compromised, it was because the candidates in question forcefully invoked the principle of popular ‘sovereignty’, which always had the potential to create conflicts between different constitutional and legal conventions.Footnote ⁴³

The ‘sovereignty’ of the Roman people in important respects turns out to be restricted by and subject to the power of the magistrates. So how are we to understand the nature of this ‘sovereignty’? The relationship between populus and senatus may give us an idea, suggesting as it does that the people were subjected to another much more fundamental authority. According to Roman tradition the senate was, as the name suggests, a council of elders which had been more or less informally constituted in order to advise the kings and – after their expulsion – the consuls. In accordance with its advisory role, its formal powers were strictly limited. Its position in relation to the assemblies is therefore intriguing, since our sources insist that any decision by the assembly required the formal approval of the senate, the so-called auctoritas patrum. As Cicero stated: ‘ … no act of the popular assembly should be valid unless ratified by the auctoritas of the Fathers’.Footnote ⁴⁴ Little is known about the precise nature of this nebulous concept, and the clearest evidence comes from two laws which reformed its application. The lex Publilia of 339 prescribed that it must be given to new laws before the vote was taken in the comitia centuriata, while the lex Maenia (third century) required that auctoritas patrum be given before, not after, the elections.Footnote ⁴⁵

The practical significance of the auctoritas patrum is unclear: there are no reported instances where it was withheld, and many scholars have therefore suggested that it later became a formality.Footnote ⁴⁶ But its existence is nevertheless important, not least because of the implicit claim that the will of the people was not the highest authority but subject to the approval of an aristocratic body of elders. Without senatorial consent the ‘will of the people’ was null and void, and interestingly this relationship was also reflected in the layout of the political venues. Thus, the Curia occupied an elevated position above the people’s meeting place, the Comitium, which it dominated visually and symbolically. It might be tempting to see the senate’s role as part of a ‘Polybian’ compromise which balanced the interests of the aristocracy against those of the masses. But the power of the senators was not, as one might have expected, founded on conventional aristocratic claims to leadership, such as superior wealth, responsibility, and capability. It was a claim of a very different nature.

‘Religion’ and ‘Politics’

The approval of the senators was, as we saw, expressed as auctoritas patrum, and this concept is a key to understanding their power. Auctoritas carries strong religious associations, and it has been suggested that the ancient expression describing the senators’ approval ‘patres auctores fiunt’, goes back to sacral law. Moreover, the senators who gave their auctoritas were historically the patres, i.e. the patrician senators.Footnote ⁴⁷ The existence of this particular group played a central role in the development of Rome’s political system. Two aspects are crucial. First, it was a hereditary status, and, second, its members held special privileges primarily of a religious nature. Although much remains obscure, it appears that at some early stage in Rome’s history a group of families, gentes, established themselves as a separate class with particular religious rights and responsibilities. Above all they seem to have assumed the right to interpret the will of the gods through signs and rituals, and it was that role which underpinned their political ascendancy.Footnote ⁴⁸

In order to understand this connection we may briefly consider the religious construction of the Roman state as a whole. Our understanding of Roman religion has been vastly enhanced by recent scholarship which above all has stressed that modern notions of ‘politics’ and ‘religion’ as discrete and ideally separate spheres of life did not apply in Rome; in fact the conceptualisation of the two was alien to Roman mentality and indeed absent from their language.Footnote ⁴⁹ As Scheid stated: ‘ … in the Roman world there was no difference between “secular” life and religious life. Every public act was religious and every religious act was public’, and it therefore followed that ‘a magistrate was invested … with a function that extended to two complementary fields of action, namely relations with the gods and relations with men’.Footnote ⁵⁰ Both ‘politics’ and ‘religion’ were ultimately concerned with preserving the community.

The Roman res publica was founded on a partnership between the populus Romanus and its gods, and the continued support of the latter was considered vital to the well-being of the community. This ‘peace with the gods’ (pax deorum) was ensured through public ritual and sacrifice, and through the continuous consultation of the gods before any collective action was taken. Obtaining the gods’ consent was essential for any public decision-making, and a complex system of divinatory rituals allowed the Romans to interpret the will of Jupiter. The Roman preoccupation with divination may seem obsessive to modern observers, but the continuous consultation of the gods gave the Romans the freedom and confidence to act without unduly worrying whether a particular action had the gods’ approval or not.Footnote ⁵¹

The practice of divination in Rome involved the gods far more directly in the political process than in almost any other known political system. It meant that the people did not decide autonomously, but in close co-operation with the gods. Every decision was taken in consultation with Jupiter, who responded to inquiries whether meetings should go ahead or be called off. The decision-making thus involved a crucial third party, since it was the augurs, the magistrates and the senate who managed the relationship between the people and the gods. It meant the people could never act on its own, but only in conjunction with the two other partners. Legislation passed or elections conducted without favourable auspices would be vitiated and could for the sake of the community not be upheld.

In the early republic priestly and magisterial functions were, according to the ancient tradition, monopolised by the patrician order, whose main characteristic was their inherited religious authority. Significantly, it appears that the right to consult the gods, the auspicia, formally lay in the hands of the patrician senators, though normally delegated to the holders of imperium, who later came to include also non-patricians. On the death of a king the auspicia had apparently passed to the patres during the interregnum, ‘auspicia ad patres rederunt’, and after the end of monarchy the patricians still resumed control whenever the normal handover of power was interrupted.Footnote ⁵² In that situation a patrician senator would be appointed interrex in order to restore the magisterial succession. Since the auspices were essential to the exercise of power, to the passing of legislation and to military leadership, the patrician hold over the auspicia in effect meant a monopoly on executive power. But although the patricians may have monopolised the public priesthoods, the senate remained the highest religious authority, settling all disputes in that sphere. Cicero could thus with full justification describe ‘auspicia et senatus’ as the foundation of the res publica.Footnote ⁵³

The existence of the patriciate provides a clue to understanding central aspects of the republican constitution; the patrician privilege represented an autonomous source of authority – embedded in religious concepts and practices – and allowed them to claim formal ascendancy over the populus. The magistrate and his council therefore held an independent authority, distinct from the popular mandate granted by the assembly. Their religious power allowed them to intervene in any decision of the populus. The augural prerogative of the patres was always exercised indirectly by magistrates assisted by augurs, which was the reason why Roman magistrates could halt or annul popular decisions (Cic. Leg. 3.27). Simply by declaring: ‘alio die’ – on another day – they could bring comitial proceedings to an end – a right unheard of in democratic Athens. The presence of a hereditary class, controlling the mediation between gods and state, shaped Roman public life and institutions. It placed the focus of the ‘constitution’ firmly on the (patrician) magistrates, who in turn were ‘greater’ and endowed with authority to nominate their successor, who would hold similar personal authority. It also limited the pool from which the leaders could be drawn and ensured broad aristocratic homogeneity, setting the leaders formally apart from the populus.

The relationship between populus and magistrates, priests, and senate thus casts a revealing light on the nature of the ‘sovereignty’ of the Roman people. In terms of legislation it meant that every law had to be approved by the populus, but not that every wish of the people would become law. Likewise, in the case of appointments all magistrates required the approval of the people but not all those approved would necessarily become magistrates. What emerges is therefore a very different notion of political institutions from that found in the writings of Greek theorists. To gain a clearer understanding of the people’s ‘sovereignty’ we may briefly consider the Roman ‘state’.

As Cornell observed, the Romans had no ‘concept of the state in the modern abstract sense … as an impersonal entity that stood apart from the individuals who composed it. Rather it was simply the Roman people, the collectivity of the citizens’.Footnote ⁵⁴ The famously untranslatable concept res publica, often rendered as ‘state’, summed up all the affairs pertaining to the public, the populus, its antonym being res privata, the property/affairs of individual citizens. Res publica was, as Cicero stated tautologically, identical with res populi, and it has been argued that the res publica in a very literal sense belonged to the people.Footnote ⁵⁵ The ‘power’ of the people was therefore not a question of popular ‘sovereignty’ as a distinct constitutional principle in the modern sense, simply because the people were the state. It followed axiomatically that in a free res publica the populus were the only source from which law and legitimacy could be derived. Accordingly, the authority of the senate and the patricians was never a power to legislate or to appoint leaders of state but one that entitled them to a mediating role – between the Roman people, its gods, and foreign powers. Decisions might have to be approved by gods and senate but they carried no legitimacy without the formal ratification of the populus. As far as we know, this fundamental understanding of state and populus was never disputed, and it shaped both political ideology and practice throughout the republican period. But the Roman notion of the free res publica and the central role which it accorded the comitia did not involve any recognition of basic ‘democratic’ principles or the people’s right to self-government. It was essentially an acknowledgement that there could be no legitimacy without the people’s consent.

This conclusion puts the spotlight on the nature of this consent, its significance and its practical expression. To what extent was this merely a formal requirement with little or no political content? An interesting answer was suggested by Scheid, who compared the consultation of the gods with that of the people.Footnote ⁵⁶ He noted that: ‘Divination was a deliberate and precise human technique which consisted not so much in an empirical and direct consultation with the gods, but rather in the recitation of a kind of prayer that revealed the gods’ agreement with whoever was consulting them. In a way, that consultation with the gods was comparable to the magistrate’s consultations with the people’. The similarities between the two procedures are indeed striking, both carefully designed to produce affirmative outcomes; as we saw, the entire format of the assembly aimed at eliciting a positive response to the magistrate’s request for approval. In both spheres the magistrate can be seen as the originator of the action, for which he seeks the consent of the two other bodies involved, the populus and its tutelary gods. This rationale lends the comitia a ritual aspect that is brought out even more clearly when we look at the ways in which the populus expressed its consent.

Voting and Assemblies

In antiquity political participation was, in the absence of the modern concept of representation, by definition direct; people could be politically active only by turning up in person and casting their vote in the assembly. In Rome, however, popular participation was structured in a unique way, which minimised the role of the individual citizen. Whenever a vote was to be taken the citizens were organised into groups, which voted together as blocks casting a single set of votes each. Different assemblies were based on different units, curiae, centuriae or tribus, but the fundamental principle remained the same.

The Roman system of block voting had no direct parallel in the ancient world,Footnote ⁵⁷ and its origins remain obscure, although it has been linked to the structure of Roman society in pre-historic times. Thus, the oldest Roman assembly, the comitia curiata, divided people into curiae, and it has been suggested that these groups, ‘co-viria’, may have been the original units from which the Roman population had been formed. The traditional designation of the Roman citizens as Quirites goes back etymologically to the members of curiae, which Prugni argued were not artificial political constructs but autonomous units existing before the creation of the state. Supposedly, all Roman citizens were members of a curia, which retained some religious functions under the leadership of a curio.Footnote ⁵⁸

In the late republic the comitia curiata had very limited functions, and its original responsibilities are largely a matter of speculation. However, the curiae were involved in granting imperium, executive powers, to the chief magistrates, as they had previously done to the kings in an act of formal acclamation. Perhaps the curiae had once met separately to acclaim the new king, and only at a later stage were turned into a single assembly in which each unit continued to deliver a separate vote of consent.Footnote ⁵⁹ The hypothesis would provide a historical explanation for this unique invention. Still, whatever its origins the real significance of the block vote in the comitia curiata lies in the pattern it came to provide for all political participation in the Roman republic. As other types of assemblies were created they invariably adopted the ancient practice of the block vote, which persisted long after the archaic structures that first inspired it had vanished.

There are wide implications of the block vote, since it meant that it was no longer the mass of assembled citizens who collectively gave their backing to a proposal or a candidate, but the largely artificial units to which each of them had been assigned. Constitutionally the vote cast by individual citizens did not count, only the vote of their units. The block vote thus introduced an element of abstraction into the people’s participation, which marked an important modification of the ‘direct’ principle that otherwise prevailed in the ancient world. Ultimately, it enabled the populus to convene and act constitutionally without large-scale participation. It separated the populus as a political concept from its physical reality by allowing the former to be formally present in all its constituent parts without the mass of Roman citizens actually being there.

The most startling illustration of this separation comes from the late republic when the comitia curiata had been reduced to pure ritual. At that time the highest state officials would still be granted their powers by a vote in the ancient assembly of the curiae. The traditional procedure would be followed as before, the magistrate would call the curiae, votes would be cast and counted, results declared, but with the important difference that each of the thirty curiae now was represented by a single lictor.Footnote ⁶⁰ The procedure was, of course, a formality and the outcome never in doubt, but constitutionally it still carried validity and was in some way deemed necessary for the formal exercise of power. This requirement could have serious practical implications, not just, as we saw, during the civil war but also in 54, when the consul Ap. Claudius Pulcher apparently found himself unable to assume his duties as proconsul and take up the governorship of Cilicia because no curiate law had been passed. While he himself seems to have claimed that none was needed, he still took the radical step of procuring a false lex curiata through bribery and fraud. Thus, two of the consular candidates made a compact with the consuls promising that ‘if they were themselves elected they would furnish three augurs who would state that they had been present at the passing of a lex curiata which had never been passed … ’.Footnote ⁶¹ Although the precise significance of the lex curiata is contested in this case, the incident shows that a vote by thirty lictors could have real political implications. It may be tempting to dismiss the late republican comitia curiata as a constitutional anomaly, an archaic relic from the distant past preserved because of the Romans’ innate conservatism. But the important lesson to be drawn from this story lies in the fact that the people did act constitutionally and delivered a binding votum of real political significance – without actually being present. Because the vote was taken by abstract units, all that was required was the representation of these units.

The comitia curiata was, of course, unusual in its complete formalisation of political participation, but it was not entirely anomalous; it merely expressed in extreme form the underlying principle behind the people’s participation in all Roman assemblies. The block vote, whether taken in centuriae or tribus, meant that the actual number of participants became irrelevant from a constitutional point of view. A vote was valid when passed by the units, not when a representative section of the population had given their approval. The implication was that statutory quorum, a common feature in Greek constitutions, and other measures to ensure numerical and social representativity were redundant in Rome.Footnote ⁶² In effect, any crowd convened according to the formal rules and divided into their respective units was the Roman people. The validity of a law or an appointment was not dependent on how many people had taken part, and logically no efforts were therefore made to ensure a representative turnout. In fact large-scale participation was rather discouraged, as indicated by a number of features: the venues were kept small; complex and time-consuming procedures were never effectively rationalised; no provisions were made to promote lower class participation by e.g. remuneration; and comitia were even banned on market days, thereby preventing visiting rustici from taking part.Footnote ⁶³ The result was that by the late republic only a tiny proportion of the citizen population could participate, although that did not affect the perceived legitimacy of legislation or appointments; even highly contested bills or elections were never challenged on the grounds of poor or unrepresentative turnouts.

Political legitimacy was achieved through proper procedure, which meant a mandate given by the units into which the Roman people were divided. The formalisation of consent had paradoxical consequences, for while block voting made actual numbers irrelevant, it remained crucial that all units be represented. We happen to know through an aside of Cicero’s that if no one had turned up from a tribus (itself an indication of the small numbers involved) a few voters would be transferred from other tribus to represent the empty one (Sest. 109). Just as striking as the creation of fictitious tribules was the requirement that all tribes should deliver their votes even when they no longer counted. Tribes voted successively in legislative assemblies but when a majority had been reached and the bill effectively was passed, all the remaining tribes still had to come forward and cast their votes.Footnote ⁶⁴ Again the practice suggests that the popular vote had originally been unanimous, since there would be no point in obliging tribes to deliver a redundant vote if it expressed no agreement. The aim was for all units to state their consent, which was symbolically important not just in terms of its formal legitimacy, but perhaps also for the relationship between the law and the citizens, who in a sense became bound by the vote of their unit.

The same logic could also be applied to the direct opposite – but equally paradoxical – effect, namely to deny that a particular decision had been handed over to the populus when in fact it had. In 104 the lex Domitia transferred the election of the higher priesthoods to the comitia, presumably to make it less divisive and limit the opportunities for politicking and intrigue which the previous co-optive process had offered – and Domitius personally experienced. However, for religious reasons this prerogative could officially never be taken away from those with privileged religious authority and an ingenious solution was therefore found which left the choice to just seventeen of the thirty-five tribus, to be selected by drawing lot.Footnote ⁶⁵ Although the assembly now de facto elected the priests, the decision had not formally been taken by the populus since a majority of the tribus had not taken part in the proceedings.

Overall we are dealing with a system in which there was a formal requirement that laws, verdicts, and appointments be approved by the populus in order for them to hold any legitimacy. It is logical therefore that the populus which gave its approval was a highly formalised, even symbolic, version of the people, and one that was not expected to act but to respond affirmatively, i.e. give its suffragium. But there was one important exception to this pattern, which related to the judicial role of the populus. When the assembly convened as a court, the outcome was not given beforehand. Instead of simply ratifying a proposal put before it, the assembled citizens had to make a decision, and a hugely important one to the individuals concerned. The need to reach a verdict made participation in judicial comitia a very different issue, and it is here that we come across the only recorded Roman concern about assembly attendance.

Emilio Gabba drew attention to a passage in Varro, which preserves a rule from M’. Sergius commentarium vetus anquisitionis (third/second century) that provides detailed instructions on how to call judicial assemblies dealing with capital cases.Footnote ⁶⁶ It gives elaborate prescriptions on the announcement of the meeting, which must be called from the Rostra, and on the closure of the tabernae of the argentarii. The presence of senators should also be ensured, and the text reveals how seriously trial attendance was regarded, not just in terms of numbers but also of the quality of the participants. The measures aimed at creating a well-attended assembly dominated by the well-to-do, and the importance of large-scale attendance in capital trials is documented already in our earliest legal source, the Twelve Tables, where a maximus comitiatus was required for such occasions.Footnote ⁶⁷

The official concerns about participation at capital trials highlight the general absence of such considerations. The block vote gave the people’s political role an abstract quality, which reduced the significance of the individual citizen as a political agent. This point is underlined also in the different Greek and Roman definitions of the state. Eder noted that while Aristotle defined the polis as a koinonia ton politon, and the citizen according to his functions in the polis, Cicero famously defined the res publica as res populi.Footnote ⁶⁸ Thus, whereas the Greek definition mentions individual citizens, the Roman refers to a totality, the populus. Res publica was not a koinonia of citizens but, as Eder observed, an ‘abstract entity’, which might be subdivided into smaller units; Roman citizens could, in short, be politically active only as members of a group, as ‘a small part of a part’ – ‘ein Teilschen eines Teiles’.Footnote ⁶⁹

These differences in definition are further reflected in the sharply contrasting citizenship policies pursued in Rome and the Greek world. Athens was extremely restrictive with its citizenship, largely because of a political system which allowed and expected extensive participation of the citizens who were defined in terms of their political activity.Footnote ⁷⁰ Rome, on the other hand, was famously open with her citizenship, which was extended to foreign communities as well as freed slaves, feasible precisely because her citizenship was not conceived of in terms of political participation or direct personal influence.Footnote ⁷¹ It enabled the Roman state to develop along very different lines and to expand through the large-scale incorporation of outsiders into its citizen body.

Briefly returning to our opening question concerning the ‘power of the people’, the way forward may be to distinguish between formal powers and powers that were to be actively exercised. Thus, the people could in one sense do virtually everything and in another very little. Formally, the ‘people’ had unlimited powers but in practice hardly any means of exercising them. Because of the simple equation of the state with the populus, the people were by definition the only source of legitimacy in a free res publica. But in Rome the people was also defined as profoundly passive and reflective. Things happened for and through the populus but not by the populus. It might therefore be tempting to dismiss the people’s consent as a mere formality, but that would be a mistake, for in Roman public life there were no ‘mere formalities’. We are again reminded that observance of correct procedure was essential to the validity of any public action and the people’s consent was as important as that of the gods. Both parties – gods and populus – could in principle withhold it, although that would defy the logic and rationale of the ritual.

The paradox of the Roman popular assemblies may ultimately be rooted in our own approach to political institutions, which remain deeply influenced by Greek conceptualisation and Polybius’ constitutional labelling. His model did not envisage the possibility of formal, even symbolic, powers, and he therefore missed the peculiar nature of the Roman assemblies. To him each institution served a well-defined, rational function in spreading or concentrating power, and on that premise the contradictions of the Roman assemblies become inexplicable. Since it made little sense to Polybius that the Roman people had total power but no way of exercising it, he fitted the Roman situation into a model of balancing forces, which implied that the powers of each institution were complete and subject only to the limitations imposed by the powers of competing institutions. In that way the apparent lack of internal logic could be construed as a logic on a different level, one where the rationale of the system lay not in the articulation of the individual institutions but in the whole ensemble of institutions and above all in their interaction.

The ‘Struggle of the Orders’ and the Evolution of the Roman Political System

The basic principles of the Roman constitution outlined in the previous pages remained largely constant and were over time elevated to venerable ancestral custom. No constitution remains static, however, and also in Rome new institutions and offices were introduced and old ones modified, though rarely abolished. Since few strands of reliable information survive from the early republic and most of our knowledge has to be inferred from later institutions and practices, any attempt at tracing constitutional developments must in the nature of things have to remain tentative.

Our main, indeed only, contemporary evidence, the Twelve Tables (mid-fifth century), suggests a fairly ‘primitive’ rural society, which only very slowly developed centralised powers, laws and institutions.Footnote ⁷² Power appears to have been in the hands of great families whose position was based on land ownership. As noted, one group of families had early on established itself as a ruling class – the patricians, who claimed religious privileges and prerogatives that served to underpin their social and political ascendancy. Magistrates were appointed annually – how many and under what titles remains uncertain – and their powers were confirmed by popular acclamation. Alongside the magistrates a body of elders existed, drawn from the leading – mostly patrician – families who advised the magistrates and oversaw their actions. Collectively, the patrician senators claimed the right to annul any appointment or piece of legislation, presumably in order to check the magistrates rather than the assembly, which held no political initiative. Membership of the senate may not have been formalised until the late fourth century, but we should not overestimate the fluidity of the early Roman senate. It did after all hold substantial collective powers, and even an informally constituted senate represented the collective authority of the great families, whose senior members presumably could expect a seat on the magistrates’ advisory council.

The aristocratic nature of early republican society is hardly open to doubt, and the political system appears to have been designed primarily to check the powers of individual members of the ruling class. This was achieved through short-term tenure of public office and the collegiality which also limited the scope for independent action. Paradoxically, it could be argued that the elevated position of the magistrates and the immense powers invested in them suggest a considerable degree of cohesion and internal discipline within the early Roman elite, given that aristocratic systems normally delegate such authority only when there is a reasonable expectation that office holders will act in their collective interest.

We have no reliable information about the selection of magistrates in the early period. Presumably, it happened consensually through more or less formalised negotiations among the patrician elders – who already had the power to block unsuitable appointments.Footnote ⁷³ It would therefore appear that the early republic was ruled by a small hereditary class of families, which took turns to fill the magistracies and perform the largely military responsibilities that came with them. The populus, organised in its various subdivisions, was called upon to ratify proposals and declarations of war and approve their new leaders.

Changes to the system happened mostly as a result of the challenge presented by the so-called plebeians, which has become known as the ‘Struggle of the Orders’. Reconstructing this conflict poses fundamental problems because of the almost complete lack of contemporary sources. It has come to us through the filter of later republican history, and whatever information was available to the writers of that period – and that was probably quite rudimentary – would inevitably have been coloured by more recent experiences of political conflict, casting fundamental doubts on most aspects of the traditional narrative. As Linderski noted, ‘the only thing not in contention is that it did take place’.Footnote ⁷⁴ And the most compelling evidence for its existence comes from the specifically plebeian institutions which grew out of the ‘Struggle’ and eventually became integral parts of the Roman constitution.

The plebeian institutions were created in opposition to the established political structures dominated by the patricians, and included plebeian officials, tribuni plebis and aediles plebis, and an assembly, the concilium plebis, which issued formal resolutions, known as plebiscita. The most striking aspect of these institutions is the fact that they represent a direct mirror image of the official institutions of the Roman state. Thus, the nature of the relationship between tribune and concilium appears to have been similar to that between magistrate and comitia. The concilium was called and controlled by the tribune and subject to limitations identical to those imposed on the comitia. We also find the same formalisation of popular attendance, using block votes rather than individual votes. In historical times the concilium was convened in tribus but there are indications that it may originally have been organised in curiae, further suggesting that the concilium was modelled directly on the comitia curiata.Footnote ⁷⁵ The main difference between the two systems lay in the role of the tribunes and in the source of their powers. They had the right, perhaps even the obligation, to intervene on behalf of the plebeians and protect their interests against (patrician) magistrates, and their power to do so came from a lex sacrata, a formal oath taken by the plebeians to protect their tribunes, which conferred sacrosanctitas and made them inviolable.

The plebeian ‘movement’ has traditionally been seen as a genuinely popular reaction against patrician domination and exploitation. It is therefore often assumed that the ‘Struggle’ in the early stages was focused on the socio-economic issues concerning ordinary plebeians, and that only later, as a plebeian elite gradually emerged, did demands for full political equality and participation arise.Footnote ⁷⁶ The plebeian institutions, however, are difficult to reconcile to this model, since they appear to have been no more ‘democratic’ in their structure than those of the supposedly oppressive patrician state, which they replicate in almost every respect. Like the comitia the concilium plebis was entirely controlled by their leaders and allowed no independent initiative.

This puzzling situation opens up two lines of explanation: either the plebeians were unable to conceive of any other form of political organisation and believed that only this structure would command sufficient authority to enforce their demands; or, alternatively and perhaps more plausibly, the plebeian institutions reflected a social structure which was as hierarchical and unequal as that of Roman society as a whole. In other words, they point to the formation of a plebeian elite already during the early stages of the ‘Struggle’, an elite which used the institutions as vehicles for their claim to authority and legitimacy equal to that of the patricians. That interpretation would help explain why the leadership structure of the ‘popular’ concilium plebis was not qualitatively different from that found in the existing assemblies, presumably so unresponsive to the needs of the populus. The plebeians appear to have entertained the same basic notions of the state, legitimacy and power as other Romans at the time. The implication is that a truly ‘democratic’ movement may never have existed in Rome, the plebeian ‘state within the state’ being closer to a paternalistic aristocracy which embraced the interests of the masses, in part perhaps to its own advantage.Footnote ⁷⁷

The plebeian resistance was, of course, rooted in discontent – with the rule of the patricians or perhaps just with specific social issues. In the early stages the sources suggest it was focused on alleviating particular problems: debt and bondage; access to public law, personal security, and land. But the late introduction of constitutional issues, including the demands for plebeian access to state offices, does not necessarily make it a popular movement ‘from below’. Initially, the religious monopoly of the patricians may have seemed so fundamental an obstacle that any attempt to gain full equality appeared futile. Moreover, the major disadvantage suffered by leading plebeians – their inability to hold military commands – may have become a pressing concern only as Rome began her Italian expansion in the fourth century.

The ‘Struggle of the Orders’ has often been seen as uniquely Roman in its origins, character, and articulation, probably because of the special status of the patrician elite, but comparative studies of city-states in medieval Italy have revealed interesting parallels. There we find similar patterns of social conflict with oligarchic monopolies challenged by il popolo, whose council, the consiglio del popolo, appointed its own leaders and protectors, the capitani del popolo, separate from the governing institutions. They would typically gain official recognition over time and become established parts of the government. Moreover, despite their popular rhetoric these leaders were generally of relatively high economic standing, i.e. members of an excluded class of property owners who championed the cause of the people in order to gain access to the inner circle of the ruling class.Footnote ⁷⁸ A similar scenario cannot be ruled out in early Rome.

In 367 a compromise appears to have been reached with the passing of the Licinio-Sextian laws, which formally granted plebeian access to the chief magistracy. However, the fasti suggest that it only took effect in 342 when the plebeians were finally guaranteed one of two annual consulships.Footnote ⁷⁹ It may not have been until this moment that the dual consulship with equal powers came into existence, i.e. as an ingenious solution to the problem of accommodating the claims and interests of two distinct constituencies while at the same time keeping effective check on the plebeian office holders. According to the annalistic writers and the fasti the collegiate system went back to 509, the first year of the republic, but Livy’s famous reference to a praetor maximus who would hammer a nail into the Capitoline temple each year casts doubt on this tradition (7.3.5). As many scholars have argued, the story implies that the early republic was headed by a single senior magistrate, perhaps assisted by two lower officials.Footnote ⁸⁰

After the settlement the auspices may formally have remained in the hands of the patricians, who also continued to control the interregnum throughout the republic. The plebeian magistrates, as Linderski suggested, would have been allowed to use the auspices, although they would never ‘have’ them.Footnote ⁸¹ Since the consulship primarily was a military office and the holders were away on campaigns most of the year, a third magistrate, known as the praetor, was appointed.Footnote ⁸² He was expected to stay in the city and be in charge of jurisdiction, while the consuls were in the field. The compromise also changed the role of the plebeian tribunate and the concilium plebis. The tribunes became state officials with the power to propose laws binding for the whole community – and to block all public proceedings. However, the tribunes’ lack of auspicia meant that their resolutions had no divine approval, a problem that might have been solved by extending the auctoritas patrum to plebiscita.Footnote ⁸³ It is probably in this context that the changes to the auctoritas patrum, mentioned above, should be interpreted.

Membership of the senate was formalised by the lex Ovinia, passed between 339 and 318.Footnote ⁸⁴ It also allowed plebeian access to the censorship, and may be seen as part of the general compromise, which established the senate as a permanent institution, independent of the consuls, for the newly formed aristocracy of patricians and plebeians. It laid down new rules for the lectio senatus, the censors’ selection of senators, and according to Festus (290L), senators should now be drawn ‘ex omni ordine’ to ensure proper representation for plebeians. In principle, all citizens might have been eligible but the poor were, of course, never considered. In theory, the censors ought to pick the best men from every rank, while in practice only aristocrats were selected. However, the senate’s formal openness remained important to the elite ideology that evolved during the middle republic, as we shall see in the next chapter. At this stage the tribunes may not have been regularly enrolled into the council, which counted fewer than 300 members. Cornell has argued that the senate would have been a much weaker body prior to the lex Ovinia, which ‘emancipated’ it from the magistrates by ensuring that senators could no longer be expelled for political reasons.Footnote ⁸⁵ Exclusion now required a special written justification, a censorial nota. In reality, however, the lex Ovinia may have changed little, since membership of the senate had probably already become relatively stable before its passing.Footnote ⁸⁶

Another important outcome of the ‘Struggle’ was the creation of a tribal assembly, the comitia tributa, convened by consuls or praetors to pass leges and elect curule aediles. Since the tribal division of the voters appears to have been first introduced in the concilium plebis, we are faced with the paradox of a ‘state’ institution apparently modelled on an ‘alternative’ plebeian one. The existence of a comitia tributa has therefore been called into question by scholars, who argue that the concilium plebis remained the only assembly entirely based on the tribus until the Sullan reform.Footnote ⁸⁷ However, the reported instances of consuls convening the tribes undermine this hypothesis. The paradox of the comitia tributa might be solved by rethinking what a Roman assembly was. It may be a misconception to see them as ‘institutions’ in the modern sense of permanent constitutional bodies. They had no existence independently of the magistrates, who – quite literally – created them ad hoc to ratify a specific proposal, deliver a verdict or approve of/elect his successors. According to context and purpose the magistrates would call up the populus in one of its various divisions: the centuries primarily for military and foreign matters, the curiae for religious purposes and the tribes for most domestic affairs. Therefore, rather than being separate bodies these assemblies were simply the populus convened in one configuration or another. This point is illustrated by the famous assembly in 45 when the consul first convened the populus as a comitia tributa to elect the quaestors, then changed his mind and decided to call a consular election, in turn reconfiguring the crowd into a comitia centuriata.Footnote ⁸⁸ Although the case is exceptional, the procedure was perfectly legal since it was the magistrate who constituted the assembly for a task which he himself defined.

The appearance of tribally organised comitia reflected the growing importance of the tribes during the early republic. The process is obscure and much disputed, but we can observe how these geographically defined units came to replace the curiae as the basic division of the Roman people. From the late fourth century this role was assumed by the tribes, which became the units used for both the census and for conscription (perhaps from 332).Footnote ⁸⁹ Membership of a tribus eventually came to signify Roman status and identity, to the point that it even entered the standard nomenclature of male citizens. Following this shift the tribunes began to organise the plebs into these, presumably more authoritative, units when calling a concilium. Viewed against this background it should come as no surprise if the consuls chose to convene the people along similar lines. In doing so, they may not necessarily have taken their cue from the tribunes; most likely their actions reflected a general trend which accorded ever greater prominence to the tribal divisions of the Roman people and, as we shall see, eventually would affect also the military assembly, the comitia centuriata.

The nature of this development is difficult to gauge, and there is no agreement as to the origins and significance of the tribus.Footnote ⁹⁰ The question has long been approached from a ‘political’ perspective and placed within a discourse on ‘democratic’ and ‘aristocratic’ values in early Rome. Historians have focused on the significance of the tribes as a potential means of promoting the people’s interests, but that assumes the assembly actually served as a vehicle of popular influence. As we saw, this cannot be taken for granted and the growing prominence of the tribes may therefore not be indicative of any real shift in power. It has been speculated that the tribes had greater ‘democratic’ potential since they in principle offered all citizens an equal vote, while the curiae remained dominated by patrician gentes. Supposedly, the tribes were originally groupings of landed estates, some of which were also held by plebeians, thereby making them the natural choice of unit for their assembly.Footnote ⁹¹ On this interpretation the tribes replaced the curiae by the late fourth century as a result of Rome’s expansion, since newly enfranchised citizens could not be inscribed in gentes and hence in curiae. However, the identification of curiae as ‘aristocratic’ and tribes as ‘democratic’ is questioned by evidence suggesting that all citizens – patricians as well as plebeians – were members of a curia, although many did not know which one they belonged to.Footnote ⁹² If membership of the curiae was universal in Rome, there was no reason why newly enfranchised citizens might not be inscribed in them.

What distinguished the tribus from the curiae was their geographical definition, often seen as yet another ‘democratic’ quality. Indeed, it has been argued that they gave rural voters a relatively greater say by compensating for their longer travel and allowed them to be represented despite the practical obstacles they faced.Footnote ⁹³ The tribus, however, had no defined political interests, and no internal organs, local assemblies or elected representatives, which questions the – in itself quite anachronistic – notion of tribes as means of ‘representation’. There is, moreover, little evidence that the Romans were concerned with geographical representation or ever contemplated introducing a fairer system that, for example, correlated the size of tribes with their distance from Rome. Their only concern was that members of all tribes be present and take part in proceedings.

The general shift towards tribal voting does not lend itself easily to a ‘political’ reading and the rationale behind it effectively escapes us. But the trend was so universal that the military configuration of the Roman people, the comitia centuriata, would also eventually be affected by it. The development of this assembly is crucial for our understanding of the Roman political system as a whole.

The Comitia Centuriata

The comitia centuriata was the Roman people organised as an army and divided into military units, centuries. As a military assembly it met outside the pomerium on the Campus Martius, and originally the gathering may simply have marked the occasion when the army formally swore an oath of loyalty to its new commander, while later it delivered verdicts in capital cases, approved declarations of war and peace settlements, and passed other legislation, although to what extent remains uncertain.Footnote ⁹⁴ Its internal structure and division into centuriae reflected the same economic distinctions that determined military service. Therefore, since conscription remained closely linked to property ownership, the assembly had a built-in social bias in favour of the better-off. Initially, the property distinctions may have been quite basic and directly mirrored the primitive organisation of the army. Thus, the centuries were at first divided into equites, the classis – hoplite-type legionaries – and those placed below the classis (infra classem) the light-armed infantry or velites. The original number of centuries in each of these categories is not known, but later this simple organisation was turned into a far more complex timocratic structure consisting of 193 centuries arranged in an intricate hierarchy. At the top and voting first came the eighteen centuries of equites, including the six ‘sex suffragia’ (probably the senators – originally just the patricians). Then came the eighty first-class centuries, twenty centuries for each of the second, third and fourth classes, thirty for the fifth, followed by four non-armed military units, and finally the single century of proletarii who were too poor to serve in the army. Although ascribed to Servius Tullius by the ancient sources, this elaborate structure seems to have been introduced in the late fourth or early third century.

The purpose evidently was to grade the influence of individual voters according to rank and social standing, and its introduction may have been a response to a functional change which transformed the elective assemblies from acclamatory into decision-making bodies faced with real choices between multiple candidates. This development cannot be traced in any detail or even approximately dated.Footnote ⁹⁵ An early date may seem difficult to reconcile with the monopoly on the chief magistracy which the literary sources insist the patricians held during that period. A limited selection of patrician candidates might, of course, have been put before the assembly, but it is perhaps more likely that the succession was arranged informally in advance before receiving the approval of the assembly. Similarly, the appointment of mostly patrician colleges of ‘consular tribunes’ in the fourth century can hardly have been the result of open elections which offered a – predominantly plebeian – electorate a broad choice of patrician and plebeian candidates.

After the plebeio-patrician settlement it was, in any case, no longer possible to regulate the succession internally within the elite. The political class had not only become much broader but probably also less cohesive. It may therefore not have been until the later fourth century that elections with several candidates contesting each post became the norm. A number of features might support this idea. The new constitutional arrangement reserved one consulship for the plebeians, which almost certainly must have become the object of competition; presumably there was no well-defined plebeian ‘inner circle’ able to organise the succession, comparable to the heads of the old patrician families. The decision was therefore left to the assembly, a solution that held numerous advantages for the elite and may have been instrumental in securing the long-term future of the aristocratic government. Not only did it strengthen the inclusive construction of the res publica that became central to the identity of the emerging plebeio-patrician nobilitas, but the comitia also provided the vital external factor which ensured that the procedures regulating access to executive office remained public and transparent. In addition, leaving the choice to the ‘people’ may have helped address the problem of electoral failure; as Bleckmann noted with reference to Thucydides, defeats are borne more easily when they are not inflicted by one’s peers but by an outside body.Footnote ⁹⁶ In Rome this adjudicatory role was passed to the populus, i.e. the abstract and highly formalised version of the people that constituted itself in the assembly.

The changes to the elite thus had a knock-on effect on the role of the assemblies, and particularly the comitia centuriata, which in turn explains the reorganisation of this institution in the late fourth century – the moment when the plebeio-patrician compromise became effective. The traditional Roman practice of block voting offered a unique opportunity for giving votes different weight, a practice later praised by Cicero as embodying the specifically Roman version of citizen rights, which managed to be simultaneously equal and unequal (Rep. 1.43). The restructuring of the comitia centuriata was an attempt to ensure that it conformed more directly to traditional aristocratic ideals by guaranteeing men of substance a greater say in the decision-making. As such the reform was a practical response to wider political changes and reflected concerns similar to those expressed in the regulations on the calling of judicial comitia. As we saw, the commentarium vetus anquisitionis sought to ensure not just a substantial turnout but also the presence of the ‘right people’, and this preoccupation with the maintenance of traditional social hierarchies also dictated the transformation of the comitia centuriata into a stratified timocratic body.Footnote ⁹⁷

The political implications of this move should not be overstated, however, and there is little reason to perceive it as a concerted attempt to assert oligarchic control over elections. The elite were already in full command of the process through the powers of the presiding magistrates and the senate. Moreover, the political content in elections was probably always negligible; indeed there is no evidence that they were ever the focus of any sustained conflict between elite and populace. The massive under-representation of the poor, confined to just a single unit in the centuriate assembly, has struck modern observers as extreme in its near-disenfranchisement of the lower classes, but the most interesting aspect of the proletarian centuria is perhaps its very existence. Excluded as they were from service in the army, the poorest had logically no place in a military assembly. The presence of the proletarii in the comitia centuriata may therefore be understood in symbolic terms as an inclusive gesture reflecting traditional Roman concerns that the entire citizen body take part in proceedings which confer legitimacy within the res publica. Rather than a ‘trick’ to deceive the masses, as it is often presented, the proletarian century is a reminder of the ‘ritual’ dimension to public proceedings in Rome.

Placing the decision in the hands of the propertied classes automatically reduced the incentive for candidates to campaign widely at elections, which may have been one of the major benefits of the reform. Since open competition poses a threat to the stability of aristocratic systems, it is crucial that the electoral structures and procedures are designed to minimise the potential for clashes. These concerns may also have provided the spur for the next reform of the comitia centuriata, which casts a revealing light on the politics of the middle republic and therefore will be considered in some detail.

The Reform of the Comitia Centuriata

The evidence of Cicero and Livy implies a change to the structure of the comitia centuriata during the later third century. The former indicates a reduction in the number of first-class centuries from the eighty previously recorded to seventy, while Livy adds more detail, informing us that the ancient system of Servius Tullius had since been modified and in his day the centuries had been combined with tribus so that each tribus contained two centuries, one of seniores and one of iuniores.Footnote ⁹⁸ The reduction of first-class centuries reflected this co-ordination, the seventy centuries matching the thirty-five tribus. This figure in turn allows us to date the reform, which must have taken place after 241 when the expansion of the tribes ended and they reached their final total of thirty-five. Moreover, since no reform is mentioned in the surviving books of Livy, it probably predates 221, when the Livian text resumes. It is unclear whether all classes were co-ordinated with the tribes or just the first. The problem was a practical one – how to combine thirty-five units with twenty or thirty units. A highly complex solution was suggested by Mommsen, but as other historians have pointed out, there is in fact no explicit evidence that the centuries outside the first class were co-ordinated with the tribus.Footnote ⁹⁹ The reform probably also involved a change in the voting order, although much remains uncertain. Still, it seems likely that the equites were moved from their traditional position as the first voters and came to vote after the first class.Footnote ¹⁰⁰

The reform has often been interpreted as ‘democratic’, shifting power from the elite to the populace.Footnote ¹⁰¹ The view is based on the fact that the number of first-class centuries was reduced and the second class now had to be called in order to reach a majority, and on the reversed voting order of the equites and the first class.Footnote ¹⁰² We are, however, dealing with very minor adjustments to the distribution of voting power. Only about 5 per cent of the centuries were reallocated in favour of a few slightly less wealthy assidui of the second class. The change had no impact on the overall timocratic structure of the assembly. Moreover, the loss of the joint majority of equites and first class was only really significant if we assume that they had previously voted unanimously and effectively kept the second class away from voting, which seems unlikely.Footnote ¹⁰³

The basic difficulty with ‘democratic’ interpretations of the reform lies with the premise that elections were inherently ‘political’. But not only was the field of candidates restricted to the elite, but there is also no evidence that Roman elections – even in the polarised climate of the late republic – were ever driven by programmes or policies that turned them into ‘political’ events in a modern sense. Some historians have attempted to gauge the political character of the reform by comparing the number of ‘new men’ in the consul lists before and after the changes, assuming that an increase would indicate a strengthened ‘democracy’. Since none could be traced, it was concluded that the reform must have been oligarchic.Footnote ¹⁰⁴ The equation of ‘new men’ with democracy is dubious, however, since there is no reason to believe the newcomers generally would have been any more responsive to the people’s needs than the old elite, whose ranks they joined. A ‘democratic’ assembly need not in principle have elected more ‘new men’ than an aristocratic one.

If the reform is unlikely to be ‘political’ in the sense that it shifted power from one section of society to another, we will have to look for an explanation elsewhere. The central feature was undoubtedly the integration of tribes and centuries, since it was the introduction of a tribal element that triggered the reduction of first class units. And while the political consequences of this change may have been limited, it was highly significant on a formal level. The comitia centuriata was historically the people assembled in its military configuration, but its units, the centuries, no longer represented the populus under arms. As army units they had been replaced by maniples sometime between 340 and 280, thereby removing the correspondence between military units and voting units.Footnote ¹⁰⁵ Meanwhile the tribus had, as we saw, acquired a status as the primary division of the Roman people, used for a wide range of administrative and political procedures. Therefore, by co-ordinating tribes and centuries the formal authority of the centuriate assembly – as well as the mandate it delivered – would be strengthened. This approach does not, however, clarify the likely change in voting order; that requires a different explanation. Moving the equites from their prestigious and influential position as first voters might seem a ‘demotion’, but since their place was taken by the first class, evidently also men of substance, it is perhaps better understood as a result of another innovation, which is the introduction of the so-called centuria praerogativa, itself one of the most intriguing features of the Roman electoral process.Footnote ¹⁰⁶

The Centuria Praerogativa: Lottery and Elections

After the reform one century from the first class, the centuria praerogativa, would be chosen by lot to cast its vote before the rest.Footnote ¹⁰⁷ Only when its results had been announced could the voting of the first class commence. The rationale behind granting one unit a privileged role as ‘pre-voters’ must have been to provide a lead for the rest of the centuries. And according to the ancient sources the example set by the centuria praerogativa was indeed followed by the other centuries, to the extent that the outcome of the election could be predicted as soon as its vote had been declared.Footnote ¹⁰⁸ The nature of the influence of the centuria praerogativa is not entirely clear. Some have seen it as a ‘bandwagon’ effect, where voters keen to support a winner backed the most promising candidates. The idea may seem plausible but does not fully explain how a single unit could create such momentum that it effectively decided the election. Its vote must have carried a very particular significance, and others have therefore interpreted the centuria praerogativa in religious terms. Cicero describes its vote as an omen, i.e. a sign from the gods, and the use of the lot is probably important in this context. If so, that might explain its impact on the rest of the voters.Footnote ¹⁰⁹

In a challenge to this interpretation, Rosenstein noted that the Romans sometimes used the lot simply as a practical means of selection and argued that if it had been an omen it must have been binding, whereas in practice the vote of the first century could be disregarded, seemingly without religious misgivings.Footnote ¹¹⁰ Thus, in 215 and 211 the presiding magistrates intervened after its vote had been announced and asked it to change its mind, which it obligingly did. Rosenstein is right to argue that the vote of the centuria praerogativa was not a formal impetrative omen, i.e. Jupiter’s response to a specific question, but that does not deprive it of religious significance. Although not every lot was regarded as a sign from the gods, in some situations it was employed for divinatory purposes.Footnote ¹¹¹ The context must therefore be taken into account, and here the ‘religious’ dimension of the occasion seems undeniable. Jupiter was a constant presence and partner in Roman public life and never more so than at a comitia. Before the assembly could begin prayers were said and auspices taken. The augurs remained in attendance throughout, and the physical setting for the vote was that of a templum, a formally inaugurated space, freed of extraneous spiritual forces. It therefore seems unlikely that a lot taken at the assembly, presumably within the templum itself, could be regarded as purely practical and ‘secular’.Footnote ¹¹²

The two instances where the centuria praerogativa was asked to reconsider its vote may be less conclusive that it might first seem. In 215 the consul Fabius Maximus recalled the centuria praerogativa and invited it to review its choice of his son-in-law, instead putting himself forward as a candidate (Liv. 24.7.10–9.3).Footnote ¹¹³ In 211 one of the candidates chosen by the centuria praerogativa, Manlius Torquatus, approached the presiding consul and asked him to recall the centuria, excusing himself because of his poor eyesight. The centuria praerogativa then conferred with the seniores of its tribe and two new consuls were elected; it is not clear whether they had been among the original candidates from the first round (Liv. 26.22.3–13). In this case there had been a flaw in the first vote, since the consul was in fact incapacitated. A second vote could therefore go ahead, which was later fully vindicated when it shortly afterwards transpired that the other chosen candidate, T. Otacilius, had actually died (Liv. 26.23.2).

An omen was not an inflexible dictate that could not be circumvented without sacrilege. Signs from the gods were always subject to interpretation and negotiation. In this case it worked through the selection of one unit which would make a choice between the available candidates. Changes to the list of candidates therefore allowed a new vote to be taken without violating the initial omen. In 215 the two eventually elected, Fabius and Marcellus, had not been candidates, and that may also have been the case in 211. The events in 215 also indicate that the authority of the centuria praerogativa went beyond the purely conventional. Thus, when Fabius Maximus tried to prevent the election of his son-in-law, he did not simply address the remaining 192 centuries, recommending a different choice to them, but insisted that the first century change its vote. His actions highlighted not just the power of the centuria praerogativa as a determining influence on the comitia, but also its ritual importance; ideally there should be no discrepancy between the vote of the centuria praerogativa and the final outcome.

The voting procedures did not require a randomly chosen unit of ‘pre-voters’ and the Romans had managed without until the reform.Footnote ¹¹⁴ The purpose of the centuria praerogativa was therefore not practical but political. Randomly selecting a single unit to provide a lead to the other centuries helped minimise divisions and generate a clearer, more unanimous electoral outcome.Footnote ¹¹⁵ But merely asking one unit to vote before the others would not in itself have produced that result; it must have been – tacitly – endowed with a much more fundamental authority most likely derived from a perceived divine aspect to its selection. The reform reinforced the ritual element of Roman elections by involving the gods more closely in the process, but it was done indirectly for the simple reason that the actual decision never could be taken out of the hands of the populus. The ingenious solution was to let ‘divine fortune’ choose a single unit which would then decide the election by its example. It was, in short, the use of the lot that gave the centuria praerogativa its special status.

There were real political benefits for the elite since it retained the principle of ‘popular choice’ while largely reducing the outcome to a matter of chance. Open contests over public office and executive power always pose a risk to aristocratic systems, and the Roman senate, concerned with the collective interests of the elite, therefore tried to put a lid on electoral campaigning and other means of gaining competitive advantages. From an elite perspective the ideal distribution of public office happens either through a closed rotational system or by pure chance, neither of which allows interference in the selection process. The former is a largely theoretical possibility in most societies – and would have been politically unacceptable in Rome – and the introduction of the centuria praerogativa may be seen as an attempt to move the process towards the latter. The reform combined election and lottery, entrusting the decision to a small, randomly chosen subsection, which would have represented an unpredictable and elusive target for electoral campaigners. In order to maximise this effect the pool of centuriae from which the deciding one was chosen had to be as large as possible, which explains in purely practical terms why the first class replaced the equites in the voting order. Lobbying the centuriae of the first class would have been more difficult simply because of its size, and perhaps even more so after additional complexities had been introduced with the co-ordination of tribes and centuries.Footnote ¹¹⁶

The reform can be understood as an attempt to manage the aristocratic republic more efficiently by reducing the incentive for open campaigning. The lottery element also militated against undue concentration of power in the hands of a few elite families, since each of them maintained close links to their own tribus and could normally rely on their support at elections; the random selection of the praerogativa therefore automatically ensured a degree of rotation. Finally, the reform strengthened the formal authority of the assembly as a representation of the populus Romanus.

While the reorganisation of the comitia centuriata may strike us as eccentric in its combination of lottery and election, it does in fact have interesting parallels in the constitutions of later Italian city-states, suggesting that Rome was responding to basic structural problems facing aristocratic republics in general. Venice and later also Genoa devised methods for electing their public officials which bear a remarkable similarity to the Roman system, although no direct inspiration can be traced. In Venice a complex procedure was introduced in 1297 to elect the Doge, which involved multiple rounds of elections and drawings of lot in order to produce a final selection of random electors who would choose the new head of state.Footnote ¹¹⁷ The aim was to create a system that precluded electioneering, officially forbidden, and maintained stability within the closed but highly competitive Venetian elite.

The Roman reform probably shared similar objectives, responding to the new situation created by the plebeio-patrician settlement that had placed the centuriate assembly in a decision-making role. The question is how effective the measure was.Footnote ¹¹⁸ As we shall see, electoral campaigning appears to have become more intense in the following century, and there are signs that the role of the centuria praerogativa may itself have been redefined over time. In the two instances, discussed above, where the presiding magistrate called upon the first century to reconsider its vote, he asked them to change both their choices. But later Cicero implied that only the first candidate to be announced by the praerogativa carried the particular authority that guaranteed his election.Footnote ¹¹⁹ The implication is a significant reduction to its influence, and Cicero further noted that the chosen candidate would be elected at that comitia or ‘certainly that year’, ‘in illum annum’, which usually has been taken as a reference to comitia that overran and had to be completed the following day. It is not obvious, however, why Cicero would highlight such a minor technicality, and Mommsen therefore suggested the attractive reading ‘alium annum’, which would imply that if the candidate had failed that year he could still expect to be elected if he stood again.Footnote ¹²⁰ If the emendation is accepted, the passage would reflect a further weakening of the power of the centuria praerogativa during the later republic.

The institution may have sought to regulate elite competition and ‘streamline’ elections, but with the growth of empire and increased competitive pressures the scope of the first unit’s influence was redefined to allow for wider choice and hence greater opportunities for influencing the outcome. Thus, having once provided a definite guide to the other centuries, only one consul would later be decided by this century. And towards the end of the republic the praerogativa may have given little more than a general boost to the authority of its chosen candidate, improving his chances now or in the future, but no longer able to settle the election as originally intended. It was possible to loosen the electoral ‘strait-jacket’ of the centuria praerogativa in this way because its vote had never been formally defined as an impetrative omen; its authority was established through the collective perception of the assembled voters.

The elite’s attempt to manage the elections appears to have been confined to the comitia centuriata, undoubtedly because it chose the most important magistrates and was the focus of the fiercest competition. Whether the non-elective comitia centuriata operated with a centuria praerogativa is uncertain, but events surrounding the declaration of war on Macedon in 200 do not suggest so.Footnote ¹²¹ Presumably, these occasions were seen as acts of ratification rather than choice. In the tribally organised assemblies the first tribus to vote was known as the principium, but there is no evidence that it functioned as a ‘tribus praerogativa’. It was chosen by lot, but so were presumably all the units that followed during the process of successive block voting.Footnote ¹²²

The use of the lot is again significant, since it draws attention to the fact that Roman elections were not determined simply by the number of block votes each candidate gained but also by the order in which he received them. In theory, more candidates could gain a majority than there were posts to be filled, and the ones chosen would be those who gained a majority first. Therefore, also after the tribal assemblies had moved from successive to a single contemporaneous vote – probably after the introduction of the written ballot in 139 – the announcement of the votes was still done successively and after drawing lot. Electoral assemblies thus had a basic element of lottery to them, which underscores the fundamental point that they were not conceived primarily as vehicles of ‘popular power’. No effort was ever made to ensure that the results reflected the ‘Will of the People’; on the contrary, the outcome was as far as possible left to fortune. The assemblies were part – an essential part, it must be stressed – of the symbolic construction of the Roman state as a community of free citizens, acting in partnership with the gods under the guidance of its leaders. But they were not ‘political’ bodies in the modern sense. They conferred legitimacy, but a legitimacy that was not derived from democratic principles – hardly compatible with the lottery aspect – but from the observance of correct procedure, since the assemblies essentially expressed the ‘Will of the People’ in symbolic form. This point is further illustrated by the rules regulating competition, which suggest a very different conceptualisation of the people’s role.

Breaking the Rules: Ambitus

If the introduction of the centuria praerogativa ever had any substantial impact on the level of electioneering among magisterial candidates, it does not seem to have lasted long. Already in the early second century we find signs of growing competition, as a number of measures were passed to curb electoral malpractice. Regulating campaigning is entirely normal in participatory systems based on general election, but in Rome the underlying concerns appear to have been quite different from those shaping modern legislation in this area. The Roman terminology is itself revealing, since ambitus, usually translated as electoral malpractice, simply refers to candidates going around asking voters for support, an act which from a modern perspective would seem to be an entirely natural part of canvassing.Footnote ¹²³

Although the earliest recorded measures may be of doubtful historicity, they still provide a valuable hint of the nature of the concerns we are dealing with. In 432, Livy reports that a plebiscite banned the use of the toga candida, the whitened toga intended to attract attention to the candidates (4.25.13–14). Perhaps more authentic may be the lex Poetilia from 358, which curbed campaigning on market days and outside Rome (Liv. 7.15.12–13), suggesting that the mere soliciting of votes might have been considered problematic. Later in the fourth century the dictator C. Maenius issued an edict against coitiones (314, Liv. 9.26.9), which is particularly interesting as an electoral offence, since it merely describes a compact between two candidates running for the same offices. Coitiones continued to be a source of concern as documented by, for example, Cicero’s defence of Plancius against the charge of collusion, although the precise legal status of such agreements during this period remains unclear.Footnote ¹²⁴ Still, the fact that alliances could be seen as an issue suggests we consider the nature of electioneering more broadly and in particular the question of payment, which became highly controversial during the middle and later republic.

In 181 the lex Cornelia Baebia de ambitu first addressed the question of payment to voters (Liv. 40.19.11). The previous year the lex Orchia de cenis had laid down rules for private expenditure, while the games of the aediles were regulated in 179 (Liv. 40.44.10–12). In 159 yet another ambitus law was passed (Liv. Per. 47), only to be followed in 149 by the lex Cornelia Fulvia de ambitu. A permanent quaestio de ambitu may also have been established, perhaps in 122, to deal with this offence.Footnote ¹²⁵ During the first century further measures were implemented. In 67, a lex Calpurnia targeted divisores, the agents organising the payment, while in 63 the lex Tullia cast the net even wider and included games and other forms of spending such as hospitality. In 55 the lex Licinia de sodaliciis again dealt with aspects of the organisation of payment.Footnote ¹²⁶ Another attempt was made in 54 to limit the practice and in 52 the lex Pompeia sharpened the penalties for ambitus.

The issue of payment, typically described as ‘bribery’, tends to be evoked as a sign of ‘corruption’ and general political decline. Whether this interpretation is necessarily correct remains debatable, and there is a risk that we project modern notions onto republican Rome. In contemporary polities, paying voters for their support counts as an obvious transgression, but in Rome the situation may have been more complex. Not all forms of payment were, for example, illegal. When candidates confined it to their own tribules, it could be described as a time-honoured tradition, suggesting it functioned as a kind of sportulae, small gifts which patrons traditionally presented to their clients.Footnote ¹²⁷ Ambitus was committed only when candidates paid voters indiscriminately; or, in other words, the offence happened when candidates tried to subvert the position of their competitors within their tribus.

Nowadays the basic objection to electoral fraud is the damage it does to the democratic process by distorting the free expression of the ‘popular will’. The politicians competing for popular favour are, of course, affected when that happens but this is an incidental side-effect of the interference with the democratic choice; the real casualty is ultimately the sovereign people and its right to self-government. In Rome, on the other hand, the primary ‘victims’ of ambitus seem to have been the competitors who lost out.Footnote ¹²⁸ Thus the overriding aim was to maintain a level playing field between the candidates. In this context, the ban on coitiones is particularly telling, since there was nothing inherently ‘undemocratic’ about candidates pooling their resources to support each other; only their fellow candidates suffered when that happened.

In a system like Rome’s the selection of new magistrates was ideally managed with as little involvement by the electorate as possible. To that end the scale of the proceedings was, as we shall see in the next chapter, kept deliberately small and unrepresentative. Attempts were also made to limit the candidates’ ability to reach larger sections of the populus – through personal contact, public largesse or the direct purchase of votes. It was crucial that electioneering did not spin out of control, creating friction within the elite or leading to mass mobilisation of the electorate. Ambitus laws served to prevent precisely these outcomes. For that reason they included measures that went far beyond what is today perceived to be ‘malpractice’ and covered forms of electioneering now considered entirely innocent. For example, at some point before 68 the use of nomenclatores was prohibited, although their role merely was to help candidates remember the names of people in the Forum (Plu. Cat. Mi. 8.2). Likewise, in 64 the lex Fabia limited the number of sectatores, the personal supporters who followed the candidate around during the campaign (Cic. Mur. 71).

Ambitus legislation addressed the distortions caused by uneven distribution of resources and influence in the Roman elite by seeking to prevent the rich and powerful from gaining unfair advantages. It was essentially aimed at maintaining the equilibrium of the ruling class – rather than ensuring the purity of the ‘democratic’ process. As such it reminds us that what happened in the elective comitia was something more complex than simply the populus Romanus choosing its leaders. The assembly performed a key role as an ‘independent’ arbiter in the elite’s ongoing struggle for power and public honores. But to fulfil that function it must ideally be beyond outside interference. Hence the attempt to randomise the outcome as far as possible and, when that failed, the repeated measures to stop the candidates from reaching the voters.